Comments on track 1 draft final proposal

1. Please provide your organization's overall feedback regarding the Resource Adequacy Modeling and Program Design Track 1 Draft Final Proposal, which was posted on August 25, 2025.

As stated in prior comments, the Alliance for Retail Energy Markets (AReM) understands that the California Public Utilities Commission (CPUC) is unlikely to adopt the proposed CAISO default counting rules or planning reserve margin (PRM). However, AReM remains concerned regarding the differences between the CPUC and CAISO modeling and market design. As stated in prior comments, AReM recommends CAISO not adopt the Track 1 changes until Track 3B is completed to ensure any changes do not create divergence between CAISO and local regulatory authority (LRA) RA programs. While AReM recognizes that other LRAs are seeking single monthly default qualifying capacity (QC) values in contrast to the CPUC’s slice-of-day paradigm, which adopts 24 hourly values for each resource each month, it is important all LRAs avoid using divergent methodologies. Unless CAISO can show its proposed methodology results in consistent outcomes with slice-of-day, it should not adopt its Track 1 proposal.

Ambient Derates

AReM supports CAISO’s decision to remove the ambient derate from the process of setting net qualifying capacity (NQC) in its Draft Final Proposal. Doing so would have created two ambient derate processes: one by LRAs and one by CAISO, which would be unnecessary and confusing. However, CAISO chooses to ignore the impact of ambient derates in setting default qualifying capacity despite the well-known negative impact of high ambient temperatures on resource capability and the likelihood of reaching peak demand when temperatures are high. This choice compromises reliability by biasing CAISO’s calculation of reliability at times when the system is likely to be stressed. Consequently, it calls into question the validity of CAISO’s reliability assessment. While the CAISO acknowledges ambient derates are an unresolved issue, it does not commit to addressing this issue in the future. AReM recommends that CAISO make a commitment to address this issue in coordination with the CPUC so that reliability assessments are accurate and unbiased.

Model Calibration      

As mentioned at the working group meeting, CAISO’s methodology for monthly model calibration should account for capacity required to meet local RA requirements. The methodology described in the proposal uses gas-fired plants as the means of establishing monthly requirements, by adding and removing gas plants in its modeled areas. However, this methodology ignores capacity in local capacity areas that is always required to be available, and their removal in the modeling exercise does not acknowledge CAISO’s own RA requirements. At a minimum, CAISO should study the impact of this method before finalizing the proposed calibration methodology and consider using a calibration method that adds or subtracts load instead of changing generation, in accord with the CPUC.

Default NQC for Reliability Assessment     

When assessing system reliability, CAISO should use default NQC values and not use LRA-provided values. Currently, CAISO uses the largest value reported by an LRA in performing resource counting. This practice does not reflect CAISO’s assessment of the contribution of that resource to providing reliability services, as CAISO models in setting its default QC values. It also systematically understates the reliability of the system.  

For example, consider a 100 MW nameplate resource that is 50% owned by load-serving entity (LSE) 1 under LRA1 and 50% owned by LSE2 under LRA2. If LRA1 sets the NQC at 100 MW for the resource, then LSE1 can use that resource to meet 50 MW of its RA requirement. However, if LRA2 discounts the resource for outages (such as ambient derates), then LSE2 would have to show its half of the resource plus capacity from other resources to cover 50 MW of its RA requirements. Because it considers the largest LRA-set value, CAISO would count the resource at 100 MW consistent with LRA1. Consequently, CAISO counts 100 MW shown for the resource and the added resource QC shown by LSE2 due to the difference in counting convention.  

Consequently, CAISO should use its default QC values to perform resource counting for system reliability assessments. Persisting in using the largest LRA-set QC value understates the level of reliability and overstates the need for capacity. 

Battery Duration

As discussed at the stakeholder meeting, CAISO should provide greater clarity on battery duration treatment in its default QC counting rules. The Draft Final Proposal would, seemingly, lump all battery capacity together, including eight-hour batteries and four-hour batteries even though the CPUC has ordered LSEs under its jurisdiction to procure eight-hour duration storage resources and the CPUC’s slice-of-day methodology assigns greater value to longer-duration battery resources. If CAISO intends to change this in the future and create a long-duration storage classification once eight-hour storage interconnects, it should explain this in the final proposal document.

2. Please provide your organization’s feedback on the proposed UCAP default QC methodology, specifically regarding the selection of supply cushion hours and the inclusion of natures of work in unit-specific forced outage rates.

The Draft Final Proposal includes a resource-specific unforced capacity (UCAP) valuation methodology. And CAISO has stated its intention to make this UCAP definition consistent with the CPUC’s UCAP mechanism slated for implementation in 2028. AReM recommends CAISO quantify the impact that the UCAP mechanism has on the PRM value. Nominally, UCAP PRMs are lower than PRMs set based on installed capacity (ICAP) because UCAP methodologies remove the impact of forced outages from the PRM. However, CAISO’s default PRM values presented at the stakeholder meeting were much higher than the current default PRM of 15%. To aid stakeholder understanding, CAISO should show indicative ICAP and UCAP PRMs to show the impact and ensure the calculation does not double count the impact of forced outages.

1. Please provide your organization's overall feedback regarding the Resource Adequacy Modeling and Program Design Track 1 Draft Final Proposal, which was posted on August 25, 2025.

The California Community Choice Association (CalCCA) appreciates the opportunity to comment on the Resource Adequacy Modeling and Program Design Track 1 Draft Final Proposal (DFP). The California Independent System Operator (CAISO) proposes to review the default planning reserve margins (PRM) and default qualifying capacity (QC) methodologies in an annual study process and provide draft default PRMs and QCs for stakeholder consideration. CalCCA recommends that the CAISO:

• Continue its goal of maintaining alignment with local regulatory authorities (LRA);  
• Provide additional information needed to assess the draft PRMs; and
• Update the battery storage effective load carrying capability (ELCC) methodology to differentiate by duration.

LRA Alignment

The CAISO should closely coordinate with LRAs to ensure processes related to counting rules and PRMs align. The CAISO should develop its unforced capacity (UCAP) methodology in parallel with the California Public Utilities Commission (CPUC). The direction LRAs take in implementing counting rules such as UCAP will necessarily impact CAISO processes such as its application of the resource adequacy (RA) availability incentive mechanism (RAAIM). The CAISO should therefore reevaluate its default counting rules, and other related processes such as RAAIM, in coordination with the CPUC and other LRAs to ensure holistic alignment on counting rules and availability incentives.

The CAISO should also evaluate the impact of methodological differences between its modeling and the modeling performed by the CPUC. The CAISO and CPUC methodologies appear similar, with the main difference being that the CAISO removes gas capacity in its monthly calibration step while the CPUC adds flat load. In addition, the CPUC allocates more events to non-peak months, which may have different affordability impacts than the CAISO’s approach. The CAISO should therefore test whether differences in the CAISO and CPUC methodologies result in different PRMs and ELCC results. If they do, the CAISO should provide additional justification for the methodological differences and coordinate with the CPUC to seek alignment where possible.

Default PRMs

The CAISO provides monthly draft PRM numbers ranging from 19.27 to 30.09 percent. Because the default PRM has not been updated in many years, these values vary significantly from the CAISO’s current default PRM.  Since any PRM is a function of the counting methodology used, the PRM will necessarily be different. However, the magnitude of this change requires careful explanation. The CAISO should be explicit that the default PRM adopted is in no way relevant to any other PRM that applies to any other accounting methodology. In light of these principles, the CAISO should provide the following additional information for stakeholders to more fully assess the draft PRMs:

• Demonstrate PRMs Resulting from Alternative Loss of Load Expectation (LOLE) Distributions: The CAISO’s calibrated modeling results in LOLE in July, August, and September. The CAISO proposes to use an alternative event distribution by distributing LOLE across months.[1] The draft PRMs presented by the CAISO appear to be based upon the July, August, and September distribution.[2] The CAISO should share the resulting PRMs with and without the alternative event distributions so that stakeholders can assess the impacts of the CAISO’s proposed approach.
• Consider Affordability Implications when Choosing Between LOLE Distributions: Alternative event distributions may reduce the monthly PRM in the months other than July, August, and September but only at the expense of increasing the monthly PRM in July, August, and September.  In recent years, these months have seen the highest prices for RA capacity, with prices in July, August, and September double the prices in other months.[3]  Increases to the PRM in July, August, and September may increase the overall cost of meeting the target reliability level of 0.1 LOLE.  The CAISO should consider the cost implications when weighing the advantages and disadvantages of choosing alternative event distributions.

• Verify that UCAP Adoption Will Result in an Equal and Opposite Adjustment to the PRM: Any change to resource counting should be met with an equal and opposite offset in the PRM, especially since the current default PRM accounts for forced outages and the current default QC values do not. Before implementing UCAP counting rules (or any other change in counting methodology), the CAISO should provide data showing that this equal and opposite effect will materialize in the LOLE study that sets the PRM. If the offset is not equal and opposite, the CAISO should begin a dialogue with stakeholders to examine why. A shift to UCAP counting that does not meet this criterion appears flawed and deserving of further consideration to either perfect the UCAP calculation methodology or determine whether it is suitable in California.

Because the default PRM has not been updated in many years, and these values vary significantly from the CAISO’s current default PRM, the CAISO should ensure a robust stakeholder process that allows stakeholders to vet the results and understand the sources of the differences between the CAISO’s draft values and the legacy 15 percent value.  

Default QC for Storage

The CAISO proposes to establish QC for battery storage resources using the ELCC methodology and derive individual storage resource QC values using unit specific allocations based upon forced outage rates.[4] This methodology appears not to differentiate between batteries with different durations. The CAISO should revise its methodology so that ELCC values are differentiated by duration to reflect differences in reliability contribution that battery storage may have that are driven by duration, in addition to forced outages.

[1]            Resource Adequacy Modeling & Program Design (Aug. 28, 2025), at 20: Presentation-Resource-Adequacy-Modeling-and-Program-Design-Aug-28-2025.pdf.

[2]            Id., at 41.

[3]            2023 Resource Adequacy Report, Table 8: https://www.cpuc.ca.gov/-/media/cpuc-website/divisions/energy-division/documents/resource-adequacy-homepage/2023-resource-adequacy-reportv2.pdf.

[4]            Id., at 38.

2. Please provide your organization’s feedback on the proposed UCAP default QC methodology, specifically regarding the selection of supply cushion hours and the inclusion of natures of work in unit-specific forced outage rates.

See response in section 1.

1. Please provide your organization's overall feedback regarding the Resource Adequacy Modeling and Program Design Track 1 Draft Final Proposal, which was posted on August 25, 2025.

CDWR, in general, supports (while providing a few suggestions and seeking clarification) the track 1 draft final proposal to update CAISO’s default methodologies for qualifying capacity (QC) and planning reserve margin (PRM), and CDWR supports the overall framework of aiming to demonstrate a set of resource adequacy (RA) rules that would strengthen system-wide reliability while maintaining the autonomy of local regulatory authorities (LRA) to set their own rules and PRMs. The default rules will apply where an LRA has not established its own PRM and/or counting rule methodology. Although generally supportive of the goals and approach to Track 1, CDWR notes that subsequent developments in Tracks 2 and 3 may impact its support for the initiative as a whole. CDWR respectfully submits following comments:

LRA jurisdiction on PRM and counting rules:

CDWR strongly supports the commitment to maintain LRAs’ ability to adopt their own PRMs and counting rules.

Default PRM:

1. CDWR appreciates the inclusion of non-spin reserves (Page 13, Figure 5) in the LOLE studies, where Participating Load is capable of being a supply resource.
2. When and where the default PRM values will be posted by CAISO should be clarified as a procedural matter.
3. For RA purposes, CAISO should allow updates to monthly demand forecasts for hydro pumping loads that are primarily hydrology dependent: Default PRM is set based on a Loss of Load Expectation (LOLE) set standard usually applicable to systemwide load. However, for a water operation system like the State Water Project (SWP), the nature of load is different as it depends on hydrology, water demand, and various operational constraints associated with integrated water and power operation. Its loads and resources are hydrology based and can vary significantly from annual forecasts of loads and resources (used for annual RA showings) to monthly forecasts of loads and resources (used for monthly RA showings). In the monthly showing timeframe, changes in both demand and resource capacity (also known as qualifying capacity (QC)) due to hydrological changes can impact monthly RA showings significantly. These changes provide a better picture of operational demand and resource availability which better supports grid reliability. Even though the default rules are being considered based on a probabilistic approach, CDWR’s real time operation depends on its actual forecast of demand and resource capacity, and the operations must be aligned with the latest forecast of its demand and supply.

CDWR’s ability to provide grid reliability support (through RA capacity and beyond, such as Operating Procedure 4420: System Emergency) is entirely dependent on the latest forecast of water pumping demand and integrated hydro resources. Therefore, the proposal should clearly state that hydro pumping load forecast may be updated on a monthly basis and provided to the CEC for RA showing purposes.

4. RA showing granularity: CDWR appreciates CAISO’s commitment to continue RA showings based on the current granularity of annual and 12 monthly showings updated 45 days in advance. With the nature of CDWR’s unique operations compared to other LSEs, a preferred approach for CDWR would be to update the forecast of its demand and resources based on hydrological forecast and water pumping demand on a monthly basis which may or may not align with the one size fits all probabilistic approaches. There is a merit to identifying such hydro resources and water pumping demand as different from other loads and resources and treating it as such. Allowing updated forecasts of water pumping demand and integrated hydro generation on a monthly basis and counting updated hydro generation capacity associated with the water pumping operation in an integrated water and power system will result in realistic representation of hydro pumping load and hydro generation in day-to-day operations, which allows CDWR to better provide the CAISO grid the benefits of its flexible hydroelectric system.

Default Counting rules:

1. CDWR appreciates inclusion of the participating load (PL) as a separate resource in the draft final proposal (Figure 4). The current default counting rule is defined as: Average reduction in demand over a three-year period on a per-dispatch basis. However, CDWR requests that the default counting rule on PL resource be modified.

CDWR would like the PL default rules to be specific to the abilities of PL and suggests that the default counting criteria for PL should follow the standards set out in the Participating Load Agreement (PLA) and base the qualifying capacity (QC) value on the CAISO certified non-spin capability. The current CAISO tariff and Business Practice Manual (BPM) for Reliability Requirement align with the PLA criteria in setting RA must offer obligation for a PL resource that is a hydro Pumping Load. The pro forma PLA requires PL to be certified and tested by CAISO before bidding to provide Ancillary Services.^[1] DWR proposes that the default PL rule should be:

The Qualifying Capacity of a Participating Loads that is a hydro Pumping Load shall be the amount of CAISO certified non-spin capability.

Further, as the Demand and Distributed Energy Market Integration (DDEMI) initiative is considering enhancement to the current PL model to add real time (RTM) demand bidding functionality, CDWR suggests consideration of the real time demand bidding of PL to be eligible for counting towards RA. Further discussion would be needed on this aspect. If consensus is reached through further development, the default QC rule for a PL could be updated with addition of a new provision on RTM load bidding eligibility for QC values. The RTM load bidding capability can relieve some of the unserved energy in the LOLE model apart from being a supply resource for non-spin reserves in the LOLE model.

2. Hydro resource counting rule and monthly qualifying capacity updates:

CAISO proposes to use average expected load carrying capability (ELCC) method based on average hydro conditions (adjusted within the year if conditions are drastically different). CDWR notes that such approach may fit long term capacity counting only. It is essential that the method adopted for the hydro-electric resources of water management systems such as CDWR should be flexible enough to account for hydrological changes in the short term so that increased capacity can be counted for these resources. If updates to monthly QC values are allowed, it will ensure increased capacity (due to better hydrology) will be reflected in the monthly QC values. This can be achieved by allowing updates to QC values monthly for these systems as is done today. Unpredictable hydrology can result in significant differences in resource availability and allowing to update those conditions to reflect the operational availability will improve reliability and reduce cost. CDWR suggests the capacity forecasted to be generating (based on current hydrology) as provided by the resource in the latest round of forecast (as close to the month as possible) would be a better, more accurate option than historical values. Therefore, the current provision to update monthly QC values should be continued to increase efficiency and improve reliability by reflecting actual availability closer to the operational month.

3. Posting of default QC values:

CAISO plans to post default QC values by each resource in August each year (on page 10, Figure 2). Will the NQC values for RA showing be different (e.g. today’s NQC values for the compliance year updated on a regular basis) and posted separately? In the draft final proposal at page 11, CAISO indicates that CAISO can provide the calculated default QC values if requested by LRAs. Does this mean that resources using an LRA’s own counting criteria will not be included in the default NQC list published in August? For LRAs who request the default NQC values, will those be provided by CAISO? If a resource’s NQC value for RA showing is different than the default QC, what will be the implications of differing values?

[1] CAISO Tariff, Appendix B.4, Participating Load Agreement at Section 4.6.2. (citing Sections 8.3.4 and 8.4 of the CAISO Tariff).

2. Please provide your organization’s feedback on the proposed UCAP default QC methodology, specifically regarding the selection of supply cushion hours and the inclusion of natures of work in unit-specific forced outage rates.

d. Unforced Capacity (UCAP) issue:

CDWR understands that UCAP topic will be discussed after the CPUC UCAP process concludes. CDWR will provide further input during those discussions in future. CDWR would like to mention some of those issues here as well:

1. The draft final proposal indicates a supply cushion UCAP is proposed as a default QC methodology for nuclear, dispatchable thermal, and pumped storage hydro resources instead of an average ELCC approach. Does this mean that supply cushion UCAP application is limited to these resources only?
2. UCAP for a PL resource: If UCAP must be applied to a PL resource, forced outage, as it relates to a PL resource, should be defined clearly as PL is not a generating resource. In addition, discrete dispatch operational constraint for a PL pump could limit the Unforced Capacity (UCAP) reduction because a pump is either on or off and cannot be dispatched partially. An alternative approach would be to consider dispatch based performance (similarly to Demand Response resources on page 20) and in absence of historical dispatch, it could be considered as 100% available until historical values are established.
3. It is indicated that UCAP derate will be applied to resources that do not receive QC values from an LRA derived from a probabilistic or performance-based methodology (exceedance, ELCC, UCAP etc.). A process should be developed to identify if an LRA derived QC value uses some probabilistic inputs that will determine if the method is probability based and whether it should be exempt from UCAP derates.

1. Please provide your organization's overall feedback regarding the Resource Adequacy Modeling and Program Design Track 1 Draft Final Proposal, which was posted on August 25, 2025.

The California Efficiency + Demand Management Council (Council) appreciates this opportunity to comment on the Track 1 Draft Final Proposal (Proposal) in the Resource Adequacy Modeling and Program Design stakeholder initiative.  The Council’s comments address the elements of the proposal that directly pertain to demand response (DR).

2. Please provide your organization’s feedback on the proposed UCAP default QC methodology, specifically regarding the selection of supply cushion hours and the inclusion of natures of work in unit-specific forced outage rates.

The Proposal would create a default Qualifying Capacity (QC) methodology for application when a Local Regulatory Authority (LRA) does not have its own DR QC methodology.  This proposed default QC methodology would consist of a performance factor unique to each DR provider, based on the three-year performance history of that DR provider, that would be multiplied by its claimed capacity value for its DR resources.  A performance factor is a generally reasonable approach to ensure that DR providers that are not otherwise subject to a LRA DR QC methodology are able to deliver on dispatched quantities. 

As a matter of principle, the Council would like to stress the importance of the CAISO not applying a performance factor to DR providers that are already being derated by the LRA based on the same performance.  However, should the CAISO intend to apply its proposed methodology as an overlay of DR that has already been derated by an LRA based on the performance of the DR provider, the CAISO should expand the Proposal to include non-Supply Plan DR resources that have Resource Adequacy (RA) value.  The most prominent example of this is investor-owned utility (IOU) DR programs that receive RA value but are not subject to the Resource Adequacy Availability Incentive Mechanism (RAAIM).  Though these programs receive RA value, they are not shown on Supply Plans and are not required to bid their QC values.      

The Council recommends the CAISO explain in the Proposal how the proposed methodology would interact with the RAAIM to ensure that these policies do not interact in unanticipated ways.  As a related matter, the CAISO should carefully consider a possible disconnect that the proposed methodology could create for a DR provider if the three-year retrospective of the performance factor should begin to significantly understate the DR provider’s future performance as past issues that negatively impacted performance are addressed. 

In addition to the above feedback, the Council offers the following technical recommendations:

• Performance factors should be based on the weighted performance of a DR provider’s resources rather than the simple average.  The Council recommends weighting by dispatch quantity (total expected energy) and subLAP DAM LMP.  This will ensure that the performance of a small resource or a dispatch under low-need conditions does not disproportionally impact a DR provider’s overall performance.
• DR provider performance should be calculated at an annual level because any seasonal variability would be distributed throughout the DR provider’s annual performance.  Also, annual weighting is still compatible with resources that are seasonally available or come online or offline throughout the year for other reasons.  The annual calculation in these cases should simply reflect only the months in which they are available.  Finally, a monthly construct presupposes all resources will be dispatched within every month they are shown, which may not always be the case, particularly under mild conditions typical of spring and fall months.

1. Please provide your organization's overall feedback regarding the Resource Adequacy Modeling and Program Design Track 1 Draft Final Proposal, which was posted on August 25, 2025.

The Public Advocates Office at the California Public Utilities Commission (Cal Advocates) is the independent ratepayer advocate at the California Public Utilities Commission (CPUC).  Our goal is to ensure that California ratepayers have affordable, safe, and reliable utility services while advancing the state’s environmental goals. Below, Cal Advocates addresses the following:

• The California Independent System Operator (CAISO’s) definition of a loss of load “event” does not represent an actual expected loss of load (load shedding) and thus the CAISO should continue to conduct RA backstop procurement based on the resource adequacy (RA) requirements of its local regulatory authorities (LRAs);
• The CAISO should evaluate the cost implications of their translation from annual loss of load expectation (LOLE) to monthly requirements;
• Interaction of the CAISO’s proposed unforced capacity (UCAP) default accreditation and the resource adequacy availability incentive mechanism (RAAIM) must be clarified; and
• The CAISO should clarify which outages at energy storage resources affect the resources’ unit-specific accreditation.

The CAISO’s Proposed Definition of “Loss of Load Expectation”

The Track 1 Draft Final Proposal describes the CAISO’s design for the LOLE study process that will inform the default planning reserve margin (PRM) and be informed by the RA portfolio using default accreditation methods.[1]  The CAISO defines the reported LOLE as “the expected number of days per year where the modeled resources are insufficient and represents the risk of entering an Energy Emergency Alert (EEA) Watch condition or needing to call on emergency measures, rather than actual loss of firm load.”[2]

The CAISO’s proposed LOLE model deviates from industry definitions of LOLE by using the event of entering EEA Watch conditions, rather than an event of loss of load.[3]  This proposed event definition may cause general misunderstanding of reliability assessments and the default PRM.  For example, the CAISO’s LOLE model may report that a 16% PRM is required to meet a 0.1 LOLE target[4] but in this example, the 16% PRM is the point at which the CAISO enters EEA Watch conditions, not when load shed is expected to occur.[5]  Capacity and energy not included in the CAISO’s LOLE model would prevent a loss of load at supply conditions equivalent to a 16% PRM.[6]  In LOLE terms, using a 16% PRM would result in an actual, conventionally-defined LOLE much lower than 0.1. 

The CAISO maintains that setting the LOLE event at EEA Watch conditions is appropriate because emergency resources, which would be used to avoid load shedding, are not included in the LOLE model.  The CAISO does not currently plan to use the proposed LOLE model and its default PRM output to independently justify future backstop procurement since the results do not reflect the actual expectation of a loss of load event.[7]  However, the CAISO has previously determined its own PRM and used it to justify reliability must-run procurement.[8]  The CAISO should continue to conduct RA deficiency backstop procurement[9] based on the RA requirements and RA showings of its constituent LRAs.

Improvements to LOLE Distribution of Event Targets

In addition to the issues with LOLE, Cal Advocates is concerned about the distribution of loss of load event targets in the LOLE model and the resulting monthly PRM values. 

In the calibrated version of the Summer Assessment Model, the CAISO modeled all “loss of load” events to occur in July (assigning 11 events in 500 runs), August (11/500), and September (28/500).[10]  Relatively higher PRM values resulted in other months in order to model zero LOLE events occurring in those months.[11]  Cal Advocates notes that this is partly a modeling issue, which the CAISO implicitly acknowledges by proposing an alternative “smoothed” distribution of events, where the same 50 events were spread over more months (May through November).  For example, the limited number of events allowed, even in this smoothed version, still enforce zero LOLE in winter months.

The modeling improvements the CAISO should make include considering two important factors: (1) What distribution of PRM values achieves the target annual LOLE at lowest cost?  (2) To what extent is the limited number of samples driving the results?

Additionally, the CAISO should provide results that reflect modeled outcomes of their preferred analysis.  The Draft Final Track 1 paper did not include any numerical results of what the PRM would be.  The slides provided PRM values for the calibrated Summer Assessment Model, but not the CAISO’s preferred, smoothed event distribution.[12]

[1] CAISO, Resource Adequacy Modeling and Program Design Modeling and Default Rules Draft Final Proposal, August 25, 2025 (Track 1 Draft Final Proposal) at 9-10.  Available at: https://stakeholdercenter.caiso.com/InitiativeDocuments/RAMPD-Track1-DraftFinalProposal-final.pdf.

[2] Track 1 Draft Final Proposal at 13.

[3] For example, the North American Electric Reliability Council (NERC) defines LOLE as:

 “This is generally defined as the expected number of days per year for which the available generation capacity is insufficient to serve the daily peak demand.”

The CAISO’s current design does not meet this definition since additional available capacity, namely emergency resources, are not included in the CAISO’s proposed LOLE model.  NERC, Probabilistic Assessment: Technical Guideline Document, August 2016 at 2.  Available at: https://www.nerc.com/comm/RSTC/PAWG/proba_technical_guideline_document_08082014.pdf.

[4] A 0.1 LOLE is targeting supply and demand conditions that allow no more than 0.1 loss of load events per year, also known as a once every 10-year (1-in-10) standard.

[5] EEA Watch conditions are when day-ahead analysis forecasts that one or more hours may be energy deficient, whereas an EEA 2 is when all resources are in use and emergency load management programs are deployed.  CAISO, Summary of Restricted Maintenance Operations, Flex Alerts, Transmission and Energy Emergencies Issued from May 2022 to Present, June 12, 2025 at 2.  Available at: https://www.caiso.com/documents/grid-emergencies-history-report-1998-to-present.pdf

[6] For example, the CAISO’s proposed LOLE model does not include certain demand response programs (Track 1 Draft Final Proposal at 13) or pseudo-tied or dynamically scheduled imports explicitly (though a portion of those resources are modeled in general import assumptions).  The proposed LOLE model also does not consider other emergency actions like the deployment of the Emergency Load Reduction Program or gubernatorial calls for energy conservation, though Cal Advocates recognizes the challenge of modeling the impact of emergency actions.  CAISO, 2025 Summer Loads and Resources Assessment Technical Appendix, May 5, 2025 at 5.  Available at: https://www.caiso.com/documents/2025-summer-loads-and-resources-assessment-technical-appendix.pdf.

[7] Further, in this initiative the CAISO stated:

The [Capacity Procurement Mechanism] is a core part of the ISO’s backstop procurement authority designed to work in harmony with the forward LRA-administered RA programs. When there is a deficiency in load serving entity (LSE) RA plans or in specific extenuating grid circumstances, the ISO has the tariff authority to conduct backstop procurement to fill the gap and maintain grid reliability.

CAISO, Resource Adequacy Modeling and Program Design Track 3A Draft Final Proposal, August 27, 2025 at 4.  Available at: https://stakeholdercenter.caiso.com/InitiativeDocuments/Track-3A-Draft-Final-Proposal-Resource-Adequacy-Modeling-and-Program-Design-Aug-27-2025.pdf.

[8] In this instance, the CAISO did not utilize an LOLE model to determine the PRM, but Cal Advocates remains concerned that the default PRM derived from the proposed LOLE model could be similarly used by the CAISO to justify backstop procurement.  CAISO, Memorandum Re: Decision on Conditional Approval to Extend Existing Reliability Must-Run Contracts for 2023, August 24, 2022 at 3.  Available at: https://www.caiso.com/Documents/DecisiononConditionalApprovaltoExtendReliabilityMust-RunContracts-Memo-Aug2022.pdf.

[9] Those backstop justifications based upon RA plans and RA deficiencies.  The CAISO may also backstop based on intra-monthly grid conditions.  CAISO, Resource Adequacy Modeling and Program Design Track 3A: Resource Visibility Draft Final Proposal, August 27, 2025 at 5.  Available at: https://stakeholdercenter.caiso.com/InitiativeDocuments/Track-3A-Draft-Final-Proposal-Resource-Adequacy-Modeling-and-Program-Design-Aug-27-2025.pdf.

[10] CAISO, Presentation - Resource Adequacy Modeling and Program Design - Aug 28, 2025, August 28, 2025 (August 28 Slides) at 20.  Available at: https://stakeholdercenter.caiso.com/InitiativeDocuments/Presentation-Resource-Adequacy-Modeling-and-Program-Design-Aug-28-2025.pdf

[11] For example, September was modeled with an event target of 28 LOLE events and resulted in a PRM of 19.27%, whereas June had zero event targets and a PRM of 30.09%.  Cal Advocates notes that Event Targets are one of many variables that drive PRM results, however.   August 28 Slides at 41.

[12] August 28 Slides at 41 and 20.

2. Please provide your organization’s feedback on the proposed UCAP default QC methodology, specifically regarding the selection of supply cushion hours and the inclusion of natures of work in unit-specific forced outage rates.

Clarify Unforced Capacity and Application of the Resource Adequacy Availability Mechanism

The CAISO proposes to adopt new default accreditation methodologies to determine QCs for different resources.[1]  The default accreditation values would only be used by the CAISO for RA compliance if an LRA does not select another QC accreditation method.[2]  The CAISO’s proposed unforced capacity (UCAP) default accreditation method does not describe any changes to the application of the RA Availability Incentive Mechanism (RAAIM).  However, one of the objectives of the resource adequacy modeling and program design (RAMPD) initiative is to develop UCAP and discuss whether availability and performance incentives, like RAAIM, are necessary for resources accredited by UCAP.[3]

UCAP has a multi-year legacy of development.  Previous UCAP designs were paired with an exemption from RAAIM on the basis that an RA resource’s availability would largely be accounted for in a UCAP methodology.[4]  Those past discussions did not consider if UCAP was only an option for LRAs to use, amongst other accreditation options that may not include forced outage or other availability assumptions.  It is reasonable to continue to apply RAAIM to resources that qualify for UCAP accreditation but whose LRA has selected a non-UCAP accreditation method when showing the resource for RA compliance.  RAAIM should continue to be applied to resources that do not use UCAP since RAAIM is one of the few binding tools to enforce availability and the RA must offer obligation.[5] 

In the next iteration of the Track 1 Draft Final Proposal, the CAISO should clarify if RAAIM would apply to resources that use the UCAP accreditation.

Clarify Outage Treatment for Energy Storage

CAISO proposes an average effective load carrying capability (ELCC) default accreditation with unit-specific forced outage adjustments for energy storage resources.[6]  The specific types of outages that would be included or excluded in UCAP adjustments were actively debated in this workshop.  However, forced outages considered for energy storage accreditation and unit-specific adjustments were not discussed in the proposal, and it remains unclear exactly which types of outages would be considered a “forced outage” for energy storage.  Outages, and outage reporting, have also been an active topic in CAISO’s Storage Design and Modeling initiative.[7]  Foldback (where a battery’s charge or discharge rate is reduced at certain states of charge) and outages related to state of charge are particularly salient for the CAISO’s proposed energy storage default accreditation.

[1] “Qualifying capacity (QC) represents a generating resource’s capacity eligible to count towards meeting [local regulatory authority] RA requirements. The CAISO tariff defers to LRAs— including the CPUC—to establish QC criteria.”  CAISO, CAISO RA Processes and CPUC’s Slice of Day, January 2024 at 2.  Available at: https://stakeholdercenter.caiso.com/InitiativeDocuments/White-Paper-ResourceAdequacyProcesses-CPUC-Slice-of-Day-Jan09-2024.pdf.

[2] “The CAISO tariff also specifies a default PRM and default QC criteria, which apply if LRAs do not specify QC values or reserve margins for their jurisdictional LSEs…” Track 1 Draft Final Proposal at 3, 17.

[3] CAISO, Resource Adequacy Issue Paper, November 7, 2024 (RA Issue Paper) at 51-52.  Available at: https://stakeholdercenter.caiso.com/InitiativeDocuments/Issue-Paper-Resource-Adequacy-Modeling-and-Program-Design-Nov-07-2024.pdf.

[4] UCAP considers forced outage rates of resources to derate the resulting QC.  CAISO, Resource Adequacy Modeling and Program Design Working Group Discussion, February 23, 2024 at 2, 21.  Available at: https://www.caiso.com/Documents/ResourceAdequacy-Presentation-Feb23_2024.pdf.

[5] CAISO Tariff Section 40.9.  See also: CAISO, BPM Configuration Guide: Monthly Resource Adequacy Availability Incentive Mechanism, CC 8830 Version 5.4 at Section 2.1.  Available at: https://bpmcm.caiso.com/BPM%20Document%20Library/Settlements%20and%20Billing/Configuration%20Guides/Resource%20Adequacy/BPM%20-%20CG%20CC%208830%20Monthly%20Resource%20Adequacy%20Availability%20Incentive%20Mechanism%20Settlement_5.4.docx.

[6] Track 1 Draft Final Proposal at 12, 20-25.

[7] CAISO, Storage Design and Modeling Revised Straw Proposal on Outage Management Topic Group, August 8, 2025 at 7-18.  Available at: https://stakeholdercenter.caiso.com/InitiativeDocuments/RevisedStrawProposal-StorageDesignModeling-OutageManagement-Aug082025.pdf.

1. Please provide your organization's overall feedback regarding the Resource Adequacy Modeling and Program Design Track 1 Draft Final Proposal, which was posted on August 25, 2025.

Executive Summary

CESA appreciates the opportunity to provide feedback on the CAISO's draft final proposal for the Resource Adequacy Modeling Program and Design initiative Track 1, presented on August 28, 2025. We have been pleased with CAISO's efforts in loss of load expectation (LOLE) studies and its continued stakeholder engagement, demonstrating transparency and openness to improving its modeling. CESA remains committed to actively engaging with CAISO staff as the modeling efforts evolve in the future.

However, CESA holds significant concerns regarding the lack of defined principles and rationales for key elements of the Unforced Capacity (UCAP) design within the Draft Final Proposal. Stakeholders cannot provide meaningful feedback on CAISO’s proposals without understanding the underlying drivers for CAISO’s design decisions. CESA has asked for this clarity throughout the stakeholder process over the course of the last year so that it can provide meaningful and relevant feedback. CESA has surfaced significant issues with CAISO’s UCAP design approach in its comments throughout the past year and these comments have not been addressed in the CAISO’s proposals.

CESA does not support the draft final proposal.

• CAISO management cannot seek Board of Governors approval of Track 1 policy and related tariff changes until CAISO staff can meaningfully engage stakeholders on the principles and rationales underlying its proposals and incorporate these principles, rationales, and stakeholder feedback into a final proposal. 
• CAISO must hold a dedicated workshop to discuss its principles and rationales supporting its insistence on using an inaccurate methodology to determine which hours resources are most needed for reliability (Supply Cushion Hours approach) and its selection of particular nature-of-work fields which should be included in the UCAP calculation. The draft final proposal does not include any guiding principles or rationales underlying these design proposals.
• The proposal must be updated to commit the CAISO to performing a process, at least annually, involving robust stakeholder input to establish the default Qualifying Capacity (QC) methodology, QC values, and Planning Reserve Margin (PRM) to achieve the industry-standard 0.1 LOLE. This annual process, its overarching goals and objectives, and a requirement to seek stakeholder input and review before finalization, must be added to the CAISO Tariff.
• The proposal must be updated to commit the CAISO to performing an annual assessment to determine if the existing Local Regulatory Authority (LRA) RA programs are delivering sufficient capacity to meet the industry standard 0.1 LOLE reliability standard. CAISO's studies demonstrate that existing LRA RA programs are often not delivering sufficient capacity to meet the industry reliability standard, thereby putting the balancing authority area at risk. These results are critical to tracking and understanding the overall reliability trends for CAISO’s balancing authority area into the future.

The CAISO Tariff must establish a principled recurring stakeholder process performed at least annually to set default Qualifying Capacity (QC) methodology, QC values, and Planning Reserve Margin (PRM) to achieve an industry-standard 0.1 LOLE.

CESA understands that it is CAISO’s intent to establish an annual process involving robust stakeholder input to establish the default QC methodology, QC values, and PRM to achieve the industry-standard 0.1 LOLE. As such, CESA understands that CAISO intends to evolve the particular QC methodology in the future. It is important that CAISO’s main intentions, principles, and objectives are captured in its Tariff language to ensure the evolving methodology remains aligned with CAISO’s reliability imperative.

CAISO’s draft final proposal correctly identifies the objective that its default PRM and default counting rules should meet at least a 0.1 LOLE at the CAISO balancing authority area level. The proposal discusses revising the default PRM based on periodic LOLE studies and updating QC methodologies as necessary, aligning with the annual Summer Loads & Resources Assessment process. CAISO proposes an "annual or as necessary" process for stakeholder feedback into the LOLE study.

CESA strongly urges CAISO to strengthen this commitment by explicitly stating that it will perform a process, at least annually, involving robust stakeholder input to establish the default QC methodology, QC values, and PRM to achieve a 0.1 LOLE. Further, CAISO must establish certain principles that ensure CAISO staff execute the process appropriately.  The CAISO Tariff language establishing this process must include:

• Establishment of a process performed at least annually to establish the default QC methodology, QC values, and PRM.
• Establishment that its default methodology and resulting values shall be designed to achieve an industry-standard 0.1 LOLE.
• A requirement for CAISO to seek and address stakeholder feedback on the study inputs and assumptions, QC methodology, and PRM on a timeline that reasonably allows incorporation of stakeholder feedback into CAISO’s studies and results.
• A requirement for transparent documentation and rationale. CAISO must clearly document the settings and rationale for all parameters used in its LOLE modeling, especially for energy storage, to ensure accuracy and allow for meaningful annual stakeholder review and discussion.
• A requirement for the resource accreditation methodology to provide an accurate and comparable representation of each resource’s contribution to reliability.

Explicitly codifying the annual process, its principles, and objectives within the Tariff, rather than simply allowing updates "annually or as necessary" would ensure:

• Consistency and Predictability: A clearly defined annual process in the Tariff provides certainty for stakeholders regarding when and how these critical reliability parameters will be established and reviewed.
• Transparent Stakeholder Engagement: Formally embedding the stakeholder feedback process in the Tariff would reinforce CAISO's stated objective of transparent and open dialogue.
• Accountability: A Tariff-based commitment provides a stronger foundation for ensuring that the default QC methodologies, QC values, and PRM are consistently updated to reflect evolving system conditions and maintain the 0.1 LOLE reliability standard.

CESA believes that this clear, Tariff-mandated annual process is essential for ensuring robust, transparent, and accurate resource adequacy going forward.

The CAISO Tariff must establish an annual reliability analysis determining whether the RA received from load-serving entities achieved a 0.1 LOLE in the past year

CESA recommends that CAISO commit to conducting an annual assessment to determine whether the resources provided by load-serving entities under LRA resource adequacy programs are meeting reliability needs. This is critical because CAISO's own studies within this initiative have shown that existing LRA RA programs are often not delivering sufficient capacity to meet the industry-standard 0.1 Loss of Load Expectation (LOLE), thereby putting the balancing authority area at risk

Performing this assessment annually, including back-cast analysis, provides CAISO with consistent, transparent, and timely information on the sufficiency of the RA fleet. This information is essential for tracking overall reliability trends, benchmarking annual projections, and understanding whether the committed capacity aligns with actual reliability needs, ultimately enabling CAISO to take necessary backstop procurement actions if local regulatory authority program rules fall short.

2. Please provide your organization’s feedback on the proposed UCAP default QC methodology, specifically regarding the selection of supply cushion hours and the inclusion of natures of work in unit-specific forced outage rates.

The inaccuracy of the Supply Cushion Hours approach in determining when resources are in-demand will yield a fleet of RA resources that put reliability at risk at times CAISO needs resources the most

Consistent with CESA's past comments, CESA remains deeply concerned that the UCAP methodology's reliance on a Supply Cushion Hours approach will jeopardize reliability and inaccurately penalize resources for lacking performance during times when the resource is not needed for reliability. CAISO has not addressed these concerns in its proposals, and it has not provided a rationale for why it continues to pursue this approach, which would allow stakeholders to meaningfully engage on the proposal’s merits/deficiencies.

On August 28, 2025, CAISO presented an alternative supply cushion approach that identifies the 20% of hours with the highest "Supply Need" (Net Load + Outages + Contingency Reserves) by season. This indicated CAISO's openness to exploring methodologies that may yield accreditation values better aligned with actual reliability needs. Although this alternative approach is a step in the right direction, it still suffers from some of the same deficiencies as the original approach, like its incorporation of exogenous outages that CAISO operators have approved. CESA strongly recommends that this alternative method, along with other potential approaches that account for actual reliability needs and unit economics, be pursued in a dedicated workshop to thoroughly evaluate their accuracy and impact on reliability.

CESA previously proposed[1] a resource-specific methodology for determining "in-demand" hours based on experienced reliability need and unit economics, using PNode and Default Energy Bid (DEB) prices for in-service and forced outage periods, and accounting for the resource's normal operating limitations. This proposal would allow for resource-specific values, capture the moments in which CAISO actually needs the resource to support its reliability and economic dispatch objectives, and can be further streamlined to allow a relatively low implementation effort. CESA continues to seek CAISO’s feedback on the merits of this approach.

Lack of Defined Principles and Rationales for Selection of Forced Outages that Impact UCAP Values Has Severely Impaired Meaningful Stakeholder Review of the CAISO’s Proposals

CESA consistently opposed the selection of specific outage nature-of-work codes for UCAP purposes because CAISO has not yet provided an unambiguous UCAP Forced Outage definition, guiding principles, and rationales for its selections. Without a clear definition describing what constitutes a forced outage that should count against a resource's UCAP value, it is problematic for stakeholders to provide substantive comments on CAISO's selection of particular natures-of-work that would qualify. CESA previously highlighted that several "Forced" outages in CAISO’s Outage Management System (OMS) would not align with the North American Electric Reliability Corporation (NERC) GADS forced outage definition.

In its March 3, 2025, comments[2] CESA proposed unambiguous forced outage definitions for UCAP purposes, based on NERC GADS definitions, specifically focusing on equipment failures. We also proposed a Forced Outage Classification Verification Process to allow suppliers to review and verify CAISO’s initial outage classification, acknowledging that the definition may not initially map directly to existing OMS codes. The current draft final proposal lists specific nature-of-work codes as "Included in UCAP" or "Not Included in UCAP" without first articulating the foundational principles guiding this selection. This omission continues to make meaningful stakeholder input on the specific codes challenging.

It was apparent from the stakeholder discussion at the August 28, 2025, workshop that the lack of defined principles and rationales for selection of forced outages that should impact UCAP values has severely impaired meaningful stakeholder review of the CAISO’s proposals. At this meeting, stakeholders heard sometimes conflicting verbal descriptions of CAISO's rationales, highlighting the need for its rationales to be written in the proposal itself so that stakeholders can provide meaningful written feedback.

To further this important discussion, CESA recommends the outages that impact a resource’s UCAP value should adhere to the following principles, aligned with the purpose of the UCAP value to capture the reliable and dependable capacity value and incentivize appropriate resource maintenance:

• The forced outage rate used to determine the UCAP value should reflect a resource’s tendency to experience equipment failures or imminent equipment failures leading to loss of capability. This principle recognizes that forced outages due to equipment failures occur suddenly and can severely strain reliability causing the need for extraordinary operator actions within a relatively short timeframe. The need for additional capacity procurement to address these instances is reflected through a derate to each resource’s capacity value.
• The forced outage rate used to determine the UCAP value should not reflect outages used to ensure an accurate dispatch of a resource operating within its design specifications. This principle recognizes that all types of resources are available to the CAISO within their design specifications. If CAISO happens to dispatch resources into known operating limitations due to deficiencies in its market/resource modeling, suppliers should not be penalized for expressing these limitations to the CAISO through the submittal of an outage card. Many of these types of limitations are currently directly modeled by CAISO for some resource types, but not all, causing potential disparate treatment between resource types if this principle is not maintained.
• The forced outage rate used to determine the UCAP value should not reflect outages that are outside of management control. Certain outages that are completely outside of management control, such as transmission system limitations or gas pipeline limitations, are not the fault of a failure of generating equipment. The generator is fully available and capable of responding to a CAISO dispatch but is being limited by factors outside of its purview.

[1] See Response to Question 4, https://stakeholdercenter.caiso.com/Comments/AllComments/a06ca1e0-f50f-4780-b24b-ec300d9c7cce#org-73987ee8-6095-4451-a4aa-61aea36d10e9

[2] Id.

1. Please provide your organization's overall feedback regarding the Resource Adequacy Modeling and Program Design Track 1 Draft Final Proposal, which was posted on August 25, 2025.

Microsoft appreciates CAISO’s effort in RA Modeling and Program Design Track 1 and the robust discussion about elements of the Draft Final Proposal during the August 28^th working group. While we supported an average ELCC resource accounting method for all resources in the past, we do support a UCAP accounting methodology for thermal, pump hydro, and nuclear resources as important for alignment with the CPUC and to account for resource reliability contributions. While no methodology is perfect, we support an improved UCAP calculation that aligns capacity accreditation more closely with what consumers need, capacity that can be counted on during stressed conditions. By adopting a UCAP framework that is both reliability focused and CPUC aligned, CAISO can modernize RA counting in a way that enhances consistency and confidence in the RA program. 

2. Please provide your organization’s feedback on the proposed UCAP default QC methodology, specifically regarding the selection of supply cushion hours and the inclusion of natures of work in unit-specific forced outage rates.

None

1. Please provide your organization's overall feedback regarding the Resource Adequacy Modeling and Program Design Track 1 Draft Final Proposal, which was posted on August 25, 2025.

While MRP appreciates the progress made in Track 1, MRP does not believe Track 1 is sufficiently developed to take to the Board of Governors in October for all the reasons stated below.  MRP requests CAISO further consider MRP's and other stakeholder comments and revise its Track 1 proposal.

2. Please provide your organization’s feedback on the proposed UCAP default QC methodology, specifically regarding the selection of supply cushion hours and the inclusion of natures of work in unit-specific forced outage rates.

UCAP Assessment Hours.  MRP still remains concerned about the CAISO’s proposed “supply cushion hours” UCAP approach, which assesses a resource’s availability in hours with the lowest “supply cushion”: (i.e., capacity margin).  MRP shares other parties’ (notably, CESA’s) concerns that a supply cushion approach could assess resource availability during hours in which there is no demand for the resource and, likely, where there is little or no risk of loss of load.  MRP encourages CAISO to instead consider using ELCC as a default capacity counting methodology to ensure that its default counting values are compatible with a standard capacity product.  Alternatively, evaluating UCAP performance across the net load peak hours during which a resource would be in demand could also reasonably reflect a resource’s availability.  MRP respectfully encourages the CAISO to not only consider this set of assessment hours for its “default” UCAP QC calculations, but also to incorporate this assessment hour philosophy into what the CAISO intends to advocate with respect to UCAP methods at the CPUC. 

The CAISO’s proposal to apply UCAP rules only as the CAISO’s default QC counting rules also helps mitigate MRP’s practical concern about how long it will take for a resource that suffers a significant forced outage to recover the capacity accreditation that will be lost (because the likelihood of the CAISO’s rules setting MRP resources’ QC values will be reduced), MRP is still concerned, however, about this issue as it will apply to the CPUC’s UCAP rules.  MRP respectfully urges the CAISO to consider advocating for a “best three of four years” approach to setting UCAP values or to consider a paradigm in which a unit could undergo CAISO-directed testing to restore lost QC value after repairs have been made more quickly. 

Consistent with this philosophy, MRP also encourages the CAISO to remove Short-Notice Opportunity Outages (SNOOs) from the sample of hours over which UCAP would be determined.  By definition, if CAISO grants a SNOO to a resource, CAISO has determined that resource is not required to maintain reliability, and such hours should be excluded from the UCAP assessment. 

Interplay between UCAP Capacity Values and the Must-Offer Obligation.  A critically important topic not discussed in the Track 1 Draft Final Proposal is the interplay between the Must-Offer Obligation (MOO) that attaches to resources selling RA capacity and the RA UCAP capacity values.  For example, in the situation in which a resource has a 100 MW nameplate value and a UCAP-based QC of 90 MW, mathematically that resource should be required to offer 100 MW (for the CAISO to be able to “access” the 90 MW UCAP value), but if that resource experiences a 10 MW ambient derate, it cannot perform to the MW level to which it would be expected to offer. The MOO may be even more complicated for resources that sold less than the unit’s entire capacity as RA capacity and for Multi-Stage Generating Units.  Given that the CAISO assesses Resource Adequacy Availability Incentive Mechanism (“RAAIM”) penalties based on how a resource selling RA capacity complies with its MOO, the interplay between UCAP capacity values and the MOO must be carefully considered.  While MRP understands the CAISO has scoped this issue for Track 2, the linkage to Track 1 is important to determine whether CAISO’s default counting rules may impact the use of the term NQC or result in the creation of a new term specific to CAISO’s own default counting rules, such as Default Qualifying Capacity or DQC.

Monthly LOLE Distribution.  MRP supports the “smoothed” loss of load distribution shown in Figure 6 of the Draft Final Proposal.  While MRP agrees that, under current conditions, the probability of loss of load is likely concentrated in July, August and September, the ongoing proliferation of solar resources, coupled with aggressive building and transportation electrification policies, will shift reliability risk outside of these three months into other months.  The CAISO’s monthly default PRM should reflect such shifts; further, when that happens, the CAISO’s “Summer Assessment” of reliability risk should be renamed.   

Mid- and Long-Term Loss of Load Expectation Studies.  In the Draft Final Proposal (at pages 10-11, and 26), CAISO describes an annual process for conducting Loss of Load Expectation studies to inform Effective Load Carrying Capability-based default QC values and CAISO’s default near-term PRM.  While this process is reasonable, the Draft Final Proposal does not discuss the CAISO also conducting mid- to long-term LOLE analysis which was raised earlier in the initiative process.  MRP requests CAISO clarify whether lack of a discussion on these mid- to long-term studies means CAISO no longer intends to conduct these studies.  MRP hopes this is not the case, because MRP sees value in these mid- to long-term studies to help resource owners consider the long-term future of their resources.   

1. Please provide your organization's overall feedback regarding the Resource Adequacy Modeling and Program Design Track 1 Draft Final Proposal, which was posted on August 25, 2025.

Local Regulatory Authority Jurisdictional Authority

Northern California Power Agency (NCPA) strongly supports CAISO’s current proposal whereby the proposed default methodologies for qualifying capacity (QC) and planning reserve margin (PRM) shall apply only in cases where a local regulatory authority has not established its own methodology.  It is appropriate for the CAISO to continue to defer to the authority of LRAs regarding the establishment of QC and PRM methodologies for the customers taking electric service within their jurisdictions.

Counting Rules

As further described in comments previously submitted by NCPA, NCPA strongly believes that a resource should be accredited with a QC value that reflects that resource’s actual operational capability and the support it can contribute to grid reliability during high load conditions.  As such, NCPA does not support bundling dispatchable hydroelectric facilities and run-of-river hydroelectric facilities into a single resource type.  MISO, for example, recognizes the difference between these types of resources and has separate Resource Classes for “Reservoir Hydro” and “Run-of-River Hydro” (in Schedule 53B of its tariff). NCPA also does not believe it is appropriate to accredit a QC value for dispatchable hydroelectric facilities using an average ELCC methodology.  Many, if not a majority, of the dispatchable hydroelectric facilities interconnected in the system have the ability to respond to CAISO dispatch operating instructions during emergency conditions and produce output that is consistent with or near the facility’s modeled maximum output in support of grid reliability. NCPA’s dispatchable hydro has certainly done so, providing valuable reliability to the grid during emergency conditions.  NCPA believes that the proposed default counting rules for dispatchable hydroelectric resources are not equitable, especially if such rules treat dispatchable hydroelectric facilities on an equal footing as non-dispatchable solar and wind resources.  In practice, dispatchable hydroelectric resources will be required to provide peaking service that is comparable to dispatchable thermal resources. CAISO’s default accreditation rules should recognize that capability, consistent with the principle that accreditation rules should accurately reflect the value that resources provide to the grid.

NCPA strongly believes that a unit’s accredited QC should be based on what it can actually do. As such, as part of the default rules design, to the extent a resource can perform at an output level that is greater than the assumed default value, a resource must have the ability to provide evidence of its true operational capacity and update its unit-specific QC values based on its actual operational capability.  This could be accomplished via a formal attestation submitted by the resource owner in which evidence of its true operational capability can be provided to the CAISO, or by requesting an operational test (similar to an ancillary services test that can be requested and performed to demonstrate the resources operational capability). The CAISO’s proposal for resource-specific adjustments to ELCC resources based on historical performance relative to the class does not address this concern, since it may still result in an individual resource being accredited far below its actual RA capability. Moreover, because the CAISO’s proposed unit-specific adjustment is solely based on historical data, it will not reflect when a generator owner has performed a significant overhaul or repair to address the cause or source of a past performance issue. In these instances, past performance (i.e., before the overhaul or repair) does not reflect expected future performance (i.e., with the overhaul or repair now in place). NCPA reiterates its request that the CAISO modify and expand upon its proposal for an historic “adjustment factor” to include an option whereby a resource owner can submit a request to the CAISO to reset the default QC value to reflect a resource’s actual operational capability using one of these proposed methods (attestation or operational test).

For the same reasons, NCPA requests that CAISO develop a process for an SC to restore UCAP accredited capacity that was lost to due to a major outage or other event that has since been addressed with a major equipment overhaul and is unlikely to recur.

PRM

NCPA requests CAISO include in the next iteration of its proposal an estimate of the incremental cost the market will incur to achieve the draft proposed PRM targets reflected in CAISO’s Track 1 Draft Final Proposal presentation (specifically the example PRM targets reflected on Slide 41 of CAISO’s presentation discussed during the August 28, 2025 stakeholder meeting).  It is important to consider what additional cost the market may incur to achieve PRM targets that could be as high as 30% in certain months as compared to the current default PRM of 15% as reflected in CAISO’s current Tariff. Deference to LRAs to set the PRMs and counting rules is important because it allows LRAs (including the CPUC) to take factors such as affordability, market power and effectiveness into account when establishing the PRMs.

Next Steps

NCPA believes that elements of the CAISO’s proposal would benefit from additional stakeholder input prior to inclusion in a Final Proposal. NCPA proposes that CAISO organize more workshops to more thoroughly define UCAP rules and publish a Revised Draft Final Proposal addressing outstanding stakeholder suggestions and concerns before proceeding to the Final Proposal stage. These workshops should address, at a minimum, the application of UCAP rules to dispatchable hydro; what conditions truly demonstrate tightness or resource scarcity; the appropriate nature of work classifications to be included/excluded in the UCAP calculation; and the addition of the unit-specific QC adjustment processes described above.

2. Please provide your organization’s feedback on the proposed UCAP default QC methodology, specifically regarding the selection of supply cushion hours and the inclusion of natures of work in unit-specific forced outage rates.

Please see comments above. 

1. Please provide your organization's overall feedback regarding the Resource Adequacy Modeling and Program Design Track 1 Draft Final Proposal, which was posted on August 25, 2025.

PG&E remains concerned about the outstanding issues and unresolved items within Track 1 and would like to request a series of actions to take to resolve these concerns before getting to a final proposal. 

PG&E is concerned that Track 1’s UCAP methods will be misaligned with the CPUC’s UCAP methods, once finalized.   

PG&E emphasizes the need for further alignment and detailed analysis across the three tracks, particularly regarding UCAP methodology, supply cushion definition, planned and forced outage definitions, the forced outage natures of work used in UCAP calculation and the Availability and Incentive Mechanism (RAAIM) Reform consistent with UCAP implementation.  

Action requested before CAISO issues the final proposal: 

• PG&E requests that CAISO schedule a workshop to provide a deeper understanding of the UCAP assessment hours methodology, the supply cushion, the supply need, and other approaches (e.g., EFORd methodology or using the critical hours from CAISO's modeling LOLE analysis).  

• PG&E requests a better alignment between Tracks 1 and 2, particularly regarding the clarification of the urgent and forced outage definition; and the Availability and Incentive Mechanism (RAAIM) reform using any adopted UCAP methodology. As noted in previous comments, availability and incentive mechanism (RAAIM) is an additional tool that is not performing as intended and should be revised to create a greater incentive for resources to be available when needed. CAISO should provide clarity on the future of RAAIM before UCAP is implemented (link to comments submitted in July 2024  California ISO - All comments). 

• PG&E also recommends strong alignment between the CAISO and the CPUC UCAP methodologies before tariff development, particularly regarding UCAP assessment hours, forced outage types, the number of years and year weighting factors considered in the UCAP calculation, the treatment of new resources and the resources type subject to UCAP calculation. On the resources type, CAISO proposes an average ELCC counting rules for storage resources, however the CPUC is considering a UCAP calculation. Considering the high penetration of storage in the CAISO market, alignment of resources counting rules should be addressed. 

• Consequently, PG&E suggests moving elements of the UCAP methodology portion of Track 1 to Track 2. 

2. Please provide your organization’s feedback on the proposed UCAP default QC methodology, specifically regarding the selection of supply cushion hours and the inclusion of natures of work in unit-specific forced outage rates.

Further discussion and specifics on the outstanding concerns and reasons for requested action items listed in question 1: 

In the comments below, PG&E highlights the elements of the Track 1 proposal that require additional discussion (UCAP supply cushion definition, UCAP natures of work and ambient derates) and requests CAISO organize a workshop to address these issues. 

1) UCAP supply cushion: 

CAISO’s UCAP methodology relies on a “supply cushion” which represents how much shown capacity (in megawatts) remains available after taking into account outages, serving net demand, and covering contingency reserves. This value represents the relative level of supply scarcity across the CAISO BAA in each hour.  

Many stakeholders have expressed concerns about this definition, particularly the inclusion of outages in the determination of the supply cushion hours, which can have a circular reference effect where CAISO approves an outage because grid conditions are sound, but thereby creates a supply cushion hour because the outage capacity is removed from the cushion calculation. This ironically punishes the resources that took the outage because grid conditions were not tight. 

At the stakeholder meeting, the CAISO proposed a new alternative UCAP assessment hour methodology using a “supply need” approach (where the Supply Need is calculated based on Net Load + Planned Outages + Opportunity Outages + Urgent Outages + Forced Outages + Contingency Reserves). This alternative does not address the circular reference effect mentioned above and was not developed in the Draft Final Proposal. It requires additional discussion between CAISO and stakeholders. 

In previous comments (link California ISO - All comments - question 7 of the template), PG&E highlighted the key component of the UCAP methodology. A well-designed UCAP would reduce costs; better reward performance; improve reliability by creating long-term incentives for a resource to be available, lead to better maintenance practices, and better match planning and operations.  In the draft Final proposal, there is a lack of clarity on the UCAP assessment hours methodology, outage definitions and Natures of Work that are considered for the UCAP calculation which are key elements of the UCAP methodology. 

2) UCAP natures of work: 

PG&E requests additional discussion with CAISO and stakeholders on the forced outage natures of work to be included in the resource UCAP calculation. The objective is to clarify the different Natures of work and to identify which ones are relevant to assess the performance and availability of a resource. 

As previously commented (link to previous PG&E’s comments California ISO - All comments), PG&E notes the following observations on the Natures of work presented at the CAISO RA working group meeting (slide 23): 

• Situational derates: this nature of work should be used for the ambient derates calculation, not the UCAP calculation.  

• Metering/ Telemetry and RTU/RIG: these Natures of work are not true forced outages. The unit is available and can fully operate, they should not be counted in the UCAP calculation. 

3) Ambient derates due to temperature: 

The draft final proposal does not include the “Accounting for RA Resource Capabilities during Peak Conditions” element of the straw proposal. CAISO is considering scoping this proposal into future RA initiatives later. CAISO indicates stakeholders highlighted key design change recommendations regarding this proposal and LRAs also raised jurisdictional concerns in their comments.  

PG&E encourages CAISO to revise the Straw Proposal Accounting for RA Resource Capabilities during Peak Conditions element of the straw proposal including the following elements raised in stakeholders' comments such as using the seven-step CAISO verification process on a monthly basis. 

1. Please provide your organization's overall feedback regarding the Resource Adequacy Modeling and Program Design Track 1 Draft Final Proposal, which was posted on August 25, 2025.

Renew Home appreciates the opportunity to comment on the Track 1 Draft Final Proposal (Proposal) in the Resource Adequacy Modeling and Program Design (RAMPD) stakeholder initiative. Renew Home responds only to the elements of the proposal that directly pertain to demand response (DR) in question 2 below.

2. Please provide your organization’s feedback on the proposed UCAP default QC methodology, specifically regarding the selection of supply cushion hours and the inclusion of natures of work in unit-specific forced outage rates.

The Proposal would create a default Qualifying Capacity (QC) methodology for application when a Local Regulatory Authority (LRA) does not have its own DR QC methodology. Because the California Public Utilities Commission (CPUC), which has jurisdiction over DR QC values of the investor-owned utility (IOU) service areas, it appears most likely that this default QC methodology would apply primarily to DR providers (DRPs) operating under the jurisdiction of other LRAs, including municipal and public-owned utilities. However, Renew Home views the RAMPD process as a significant opportunity to develop and demonstrate a workable alternative to the QC process under the CPUC LRA. 

The proposed default QC methodology would create a performance factor for each DRP that would be multiplied by the DR provider’s claimed capacity value for its DR resources. Renew Home emphasizes that this approach shares the logic behind the Performance Assessment Metric (PAM) component of the methodology proposed by the CPUC and California Energy Commission (CEC). Renew Home has generally supported such performance-based approaches to QC that allow DR providers more agency to claim capacity. However, we have concerns about the performance factor as defined in the proposal and how it will interact with other aspects of market participation, such as the must-offer obligation and associated availability and incentive mechanisms.

As mentioned above, the performance factor generally reflects the same intuition as PAM that DRPs must be able to deliver on dispatched quantities. However, Renew Home observes that the proposed definition treats all DR events equivalently even though reliability needs are likely to be concentrated in a relatively small number of hours typically associated with high temperatures, high loads, and high wholesale prices. Accordingly, Renew Home recommends that the performance factor includes the same weights as proposed for PAM: dispatch quantity (that is, total expected energy) and day-ahead subLAP LMP. Additionally, Renew Home recommends the ratio be constrained to one at the aggregate level, rather than the event-hour level. With these proposed changes applied, the DRP performance factor would be written:

DRP Performance Factor = MIN( SUM(meter*dispatch*LMP[resource,hour]) / SUM((dispatch^2*LMP[resource,hour]), 1)

In the proposal, CAISO contemplates whether the performance factor should apply at the annual or sub-annual level, and DRP or subLAP level. Renew Home strongly recommends applying the performance factor at the annual and DRP level, except for local capacity contracts (which would be calculated separately at the level of the local reliability area.) The weighting scheme introduced above appropriately values variable performance across seasons. Furthermore, annual weighting is still compatible with resources that are seasonally available or come online or offline throughout the year for other reasons. The annual calculation in these cases will simply reflect only the months in which they are available. Furthermore, a monthly construct presupposes all resources will be dispatched within every month, which may not be the case, particularly in the shoulder seasons.

Together, these modifications will ensure that unavoidable random variability in performance can be canceled out while still giving more weight to larger dispatches (including test events), larger resources, dispatches during more critical times, and dispatches in subLAPs downstream from grid constraints to mitigate line losses and congestion, which are also reflected in LMPs.

While Renew Home is supportive of this modified performance factor, we emphasize that it is only one component of a comprehensive and cohesive QC system. First, the proposal does not include an equivalent of the Bid Assessment Mechanism (BAM) of the correspondingly named “BAM-PAM” methodology. While RAMPD Track 2 is focused on availability and incentive mechanisms, Renew Home stresses that it is critical that the requirements to bid (Track 2) and deliver (Track 1) capacity be compatible and incentivize the desired behavior. 

Second, this proposal could create a disconnect between QC and resource availability. Renew Home supports a framework that encourages DRPs to improve performance over time. If this proposal successfully does so, applying past performance to future capacity will undervalue resources that have improved performance or amended claimed QC to better reflect actual capabilities. As a result CAISO would not have visibility into actual available capacity and DRPs cannot contract for that capacity. The BAM-PAM proposal, which has already received significant stakeholder input, may be a strong starting point to address these concerns. 

The BAM-PAM proposal was intended to reflect DR resources’ obligation to both bid and deliver on dispatched capacity. In addition, it would create a “closed loop” incentive system, in which underperformance in a given RA year impacts revenue in that year through explicit financial penalties, as opposed to a future year. Track 2 contemplates whether “a new mechanism is needed to incent availability and/or performance” (emphasis added), suggesting the performance factor proposed here could be integrated into that effort. Renew Home suggests CAISO explore integrating this effort with RAMPD Track 2, using the BAM-PAM proposal as a starting point. 

1. Please provide your organization's overall feedback regarding the Resource Adequacy Modeling and Program Design Track 1 Draft Final Proposal, which was posted on August 25, 2025.

REV Renewables (REV) appreciates CAISO’s desire to move swiftly to resolve Track 1 of the RAMPD initiative, but suggests CAISO slow down to provide at least one more iteration of the proposal and stakeholder discussion. This additional time is needed to fully understand the details in the Draft Final Proposal, as CAISO jumped from a first Straw Proposal directly to the Draft Final Proposal. The new information, draft values, and details shared in the Draft Final Proposal are important for stakeholders to be able to vet and fully understand, and other key details such as definitions are not yet included.

REV supports CAISO’s proposed approach to establish the Planning Reserve Margin (PRM) to achieve a 0.1 Loss of Load Expectation (LOLE). While REV has comments on the details of the resource accreditation determination, as discussed below, the overall PRM approach seems reasonable.

2. Please provide your organization’s feedback on the proposed UCAP default QC methodology, specifically regarding the selection of supply cushion hours and the inclusion of natures of work in unit-specific forced outage rates.

REV has the following comments and questions on the proposed default Qualifying Capacity (QC) rules and would request more discussion on these items before CAISO moves the proposal to the Board.

• Supply Cushion approach needs more discussion.
  • Based on the discussion in the 8/28 meeting, it is unclear whether supply cushion is the appropriate methodology to use to determine UCAP. The heat map on slide 28 of the presentation illustrates the potential disconnect of this approach, where many of the tightest supply cushion hours occur in March and very few occur in September. This intuitively seems at odds with the previous analysis from CPUC and CAISO showing September of the highest concern for reliability, and low concern in March. CAISO indicated that the “red” boxes didn’t mean the system was at risk, just that supply was tight. However, if system reliability is not a concern, then it seems a resource should not receive an accreditation hit during those hours. CAISO could consider adding additional information and not just the number of hours with tightest supply cushion. For example, Expected Unserved Energy (EUE) and LOLE events would show the scale of impact better than “tight” supply hours. At a minimum, it should be clear that a metric such as supply cushion hours should align with when supply is needed to incentivize RA availability.
  • A definition of UCAP could help resolve whether supply cushion is the appropriate methodology or should be adjusted. For example, is UCAP the expected capacity contribution of a resource during periods of system stress? Does (or should) expected system risk tie into UCAP calculation? Is supply cushion related to system risk?
• Forced Outage Natures of Work used in UCAP and unit-specific adjustments needs more discussion.
  • REV requests more discussion to define forced outages and what natures of work should be included in UCAP or unit-specific performance adjustments for storage. In the 8/28 meeting, CAISO verbally shared that forced outages included those that are outside of management control. CAISO should document that in the next iteration of the proposal. REV also strongly suggests aligning forced outages with NERC GADS Appendix K guidelines for outage codes whose causes are outside of management control.  
  • REV recommends, at a minimum, that market model-related outages, including “Technical Limitations Not in the Market Model” should not be counted as a Forced Outage.
  • REV also suggests that CAISO and stakeholders discuss forced outages used in UCAP and RAAIM. Both mechanisms are used to incentivize availability, and CAISO should avoid double penalties if an outage counts in UCAP and in RAAIM, unless there are clear reasons to count in both.
• Energy Storage QC should be calculated consistently regardless of technology type and should account for different durations.
  • CAISO proposes to use two different methodologies for storage – pumped storage hydro will use supply cushion UCAP while battery storage will use average ELCC. This inconsistency needs to be further explained and justified.
  • REV recommends one QC method with sub-resource types for different durations.
    • CAISO’s current proposal would have the same ELCC value for a 4-hour battery storage resource as an 8-hour battery storage resource. The pumped hydro resource type does not specify duration, but could have similar reliability value as an 8-hour battery despite that pumped hydro would likely have a higher QC.
    • CAISO must be consistent in how it treats storage, regardless of whether it is battery, compressed air, pumped hydro, etc. REV suggests that breaking out storage by duration, e.g. 4-hr, 8-hr, 12-hr to start would be more appropriate. If CAISO chooses to continue to use supply cushion UCAP for pumped hydro, it should specify the duration, and allow 8-hr or more battery storage to also be calculated under supply cushion UCAP.
      • REV emphasizes this point because it has an 8-hr battery coming online next year, and therefore needs a QC rule that appropriately accounts for its reliability value.
  • REV also requests more information on why storage ELCC is lower in the winter months, particularly in December in the 8/28 presentation page 38.

1. Please provide your organization's overall feedback regarding the Resource Adequacy Modeling and Program Design Track 1 Draft Final Proposal, which was posted on August 25, 2025.

SDG&E appreciates CAISO’s efforts to develop the default values scoped in Track 1 but believes that additional work is required. At the August 28 workshop, there was a robust discussion regarding how operating within a resource’s design specifications should be treated with respect to UCAP. CESA expressed concern that under CAISO’s latest proposal, if a resource is operating within its design specifications and requires an outage card to represent its operations, although this scenario would not be an equipment failure, the resulting outage card could affect the resource’s UCAP. CESA requested that CAISO adopt a principle to prevent such an outcome. Similarly, Vistra noted that CAISO’s policy standards should be clear (a statement also supported by CESA), particularly given that CAISO’s UCAP efforts may inform the PUC’s UCAP efforts. SDG&E shares CESA’s and Vistra’s concerns and recommends that CAISO, in the next iteration of its proposal, clarify the criteria for selecting any natures of work to be used in its UCAP methodologies. These criteria should include the principle that operating within a resource’s design specifications will not affect a resource’s UCAP, and CAISO should ensure that its UCAP design reflects this principle.

The issue of operating within a resource’s design specifications is also a topic under CAISO’s Storage design and modeling initiative. As explained in SDG&E’s September 5 comments under that initiative, CAISO proposes to use the “Plant Trouble” nature of work to represent nonlinearity for BESS. This is problematic in that it: 1) would subject resources to the risk of RAAIM penalties for the periods in which the outage card is used due to CAISO not accounting for technical design specifications in the market model, 2) could ultimately drive resource owners to understate their availability in the Master File and artificially constrain the capacity CAISO can rely on for reliability, and 3) could make it nearly impossible to separate out outages due to nonlinearity vs. Plant Trouble, which could affect UCAP calculations depending on the inclusion of the Plant Trouble nature of work outage card in a future UCAP calculation. SDG&E’s recommended solution under the Storage design and modeling initiative is that, in the interim and until a long-term solution is developed, CAISO represent outages due to nonlinearity via the “Technical Limitations not in Market Model” nature of work. This solution would address the immediate need for increased tracking and visibility into these outages until a long-term solution is implemented, preferably via a comprehensive representation of nonlinearity as an operating characteristic of storage resources into the market model. This recommended approach is also responsive to the concerns raised within the RAMPD initiative. Rather than using the more generic “Plant Trouble,” implementing the “Technical Limitations not in Market Model” nature of work will enable clear identification of instances when a resource is operating within its design specifications. To the extent natures of work are used in UCAP, CAISO can then ensure that these instances do not affect a resource’s UCAP.

CAISO states that it plans to update its tariff to include a process that would give stakeholders the opportunity to provide regular feedback on its development of the default values. SDG&E supports this tariff modification and recommends that CAISO establish an annual process, so stakeholders have certainty regarding the timeline for developing the default values and opportunities to provide input.

2. Please provide your organization’s feedback on the proposed UCAP default QC methodology, specifically regarding the selection of supply cushion hours and the inclusion of natures of work in unit-specific forced outage rates.

No. comment.

1. Please provide your organization's overall feedback regarding the Resource Adequacy Modeling and Program Design Track 1 Draft Final Proposal, which was posted on August 25, 2025.

The Six Cities provide comments on the topics addressed during the August 28^th stakeholder meeting in their comments on the Draft Final Proposal below. 

2. Please provide your organization’s feedback on the proposed UCAP default QC methodology, specifically regarding the selection of supply cushion hours and the inclusion of natures of work in unit-specific forced outage rates.

The Six Cities remain supportive of the CAISO’s proposal to preserve the current structure of including default—rather than required minimum RA program elements—in the CAISO Tariff and the CAISO’s continued acknowledgement that determining planning reserve margins (“PRMs”) and qualifying capacity (“QC”) counting rules is the responsibility of local regulatory authorities (“LRAs”). 

However, the Six Cities have remaining concerns regarding the CAISO’s proposed approach to assessing the resource adequacy of the CAISO balancing authority area (“BAA”) through periodic loss of load expectation (“LOLE”) assessments.  The Six Cities have no conceptual concern with the CAISO conducting such informational assessments, and the CAISO’s proposal to do so through an iterative stakeholder process, as the Draft Final Proposal suggests, is a step in the right direction.  However, there still has been no discussion about whether use of a 0.1 LOLE to set the informational PRM is appropriate or, more fundamentally, whether the outcome of these studies will be PRM values that are reasonably achievable in the current procurement environment.  The Six Cities request that the CAISO evaluate if there is likely to be adequate capacity online and deliverable for RA purposes to and within the CAISO to meet expected PRM values starting in the RA year that this new methodology is proposed for implementation.  In connection with this evaluation, the Six Cities stress that RA cannot simply be assumed as available at any price.  As the Six Cities have explained throughout this initiative, RA prices have reflected significant increases in recent years relative to past periods, and raising target procurement requirements without a corresponding increase in available supply will not improve market conditions and will likely exacerbate scarcity and pricing impacts.  Although some LRAs and their LSEs may use the default PRM and QC values as baseline metrics, broad adoption and implementation of the new default methodologies depends on whether the resulting PRM values can be realistically achieved and whether the QC values are perceived as accurate and reasonable.  What are the implications for the CAISO if it annually establishes updated PRM and QC values, but these are not adopted in LRA RA programs?  It may be preferable for the CAISO use default methodologies that are more readily able to be implemented by most, if not all, LSEs.  The Six Cities also urge the CAISO to consider structural market constraints as well as costs that may result in limited use of the CAISO’s default values.  

To enable stakeholders to make a fully informed decision about the likely impacts of the approaches discussed in the Draft Final Proposal, the Six Cities recommend that the CAISO publish an LOLE PRM study showing the PRMs and QC values that would have resulted in 2024 and 2025 had the default methodology been in place for these years.  This would facilitate a cost-benefit assessment of moving to the updated default values produced by the proposed studies. 

The Six Cities also remain concerned that frequent updating of PRM values, as is implied by the proposed annual process, may risk establishing a “moving target” that will undermine LSEs’ resource planning and procurement.  How will the CAISO determine whether the annual study process should result in an increase in the PRM?  When will increased values take effect?  If immediately effective for the next RA year, the closure of the CAISO’s study window in August (as explained in the Annual Review Process Timeline in the Draft Final Proposal), may result in an unreasonably narrow window available to LSEs to meet any increased requirements (again, assuming that an LSE elects to use the CAISO’s default methodology for setting its PRM). 

With respect to the proposal for adoption of a supply cushion-based unforced capacity (“UCAP”) methodology to establish the default QC values for thermal and pumped hydro resources, the Six Cities remain concerned that this approach appears to result in double counting, or at least overlapping, of the impacts of stressed system conditions.  Moreover, combining such an approach with a rigid 0.1 LOLE target may result in default procurement targets that are overly conservative and unduly expensive, especially in view of the current RA market conditions noted above. 

In terms of the use of UCAP for individual resources, the Six Cities agree with the CAISO’s proposal to reference resource-specific outage information as the basis for calculating the UCAP values, but the Draft Final Proposal does not include any provisions for adjustments to the outcomes of the CAISO’s calculation methodology based on the completion of major maintenance for resources.  If a generating resource undergoes significant maintenance work that makes the CAISO’s proposed three-year look-back of outage data not an accurate representation of the resource’s forward looking capabilities, the Scheduling Coordinator for the resource should be able to provide the CAISO with updated information regarding the resource’s anticipated performance that would form the basis for an adjustment to the resource’s UCAP value.  Alternatively, the CAISO could consider an approach that would remove the worst performing year within the historical period (although this would assume that the effects of any major maintenance would be shown through a year with poor performance, which may not necessarily be the case).  The CAISO’s methodology should reflect that there are benefits to incenting resource owners to perform significant and costly (but also necessary) maintenance, while also recognizing that the timing and duration of maintenance activities may often be affected by circumstances outside of the resource’s control, such as supply chain impacts or the availability of contract personnel. 

Relatedly, the Six Cities request that the CAISO address conditions whereby a forced outage of a resource is transitioned to a planned outage status, in the event that an extended maintenance period becomes necessary following a forced outage. 

Finally, the Six Cities also support the CAISO’s determination not to move forward with the inclusion of an adjustment for ambient derates in the Net QC values for RA resources. 

1. Please provide your organization's overall feedback regarding the Resource Adequacy Modeling and Program Design Track 1 Draft Final Proposal, which was posted on August 25, 2025.

SCE appreciates the hard work that the CAISO has undertaken in constructing the draft QC and PRM values. SCE believes it is important for both the CAISO and stakeholders to understand what these are and how they may relate to other QCs and PRMs, especially those used by the CPUC and CPUC-jurisdictional entities in the CPUC’s Slice-of-Day (SoD) RA mechanism. The estimates in the CAISO’s analysis, presented in the Draft Final Proposal, are based on an RA model that looks at the peak hourly load each month. This is different from the SoD mechanism used by the CPUC. One should not expect that the QC values or the PRM values constructed under the CAISO’s single peak hour construct are comparable to those in use for the SoD mechanism.

There are additional differences between the CAISO’s modeling and the CPUC’s modeling that make direct comparisons challenging. These differences include how the stochastic load shapes and the renewable generation profiles are developed, resource outage assumptions and how storage is modeled. These differences, along with the fundamental structural differences between the CPUC’s SoD and the CAISO’s peak hour RA programs, result in the two sets of numbers not being directly comparable.

Given potential differences in reliability modeling, entities should not be able to pick and choose various QC values or PRMs from unrelated mechanisms or studies. The CAISO RA rules allow the Local Regulatory Authority (LRA) to determine the RA rules for LSEs, and the CAISO rules only provide a default set of values to be used if the LRA hasn’t determined its own RA rules. The CAISO rules should also require LSEs to use consistent modeling in determining the PRM and counting rules. Entities that choose to use the CAISO’s default counting rules should also be required to use the default PRM from those same studies. If the LSE chooses to use a PRM constructed in another manner, it should be required to use an appropriate counting methodology. Specifically, the CAISO should indicate that entities are not permitted to use the CPUC’s SoD counting rules unless they are using a SoD mechanism to determine if they have sufficient resources.

SCE also provides the following comments on the Performance-Based UCAP for Demand Response (DR) resources. Performance-Based UCAP for DR resources should not be set at the DR provider (DRP) level. The CAISO would apply a performance-based default accreditation (e.g., performance factor) to supply-side demand response resources that are listed on RA and supply plans to ensure the capacity values reflect their load curtailment capabilities (i.e., de-rate DR resources based upon their historical performance). The CAISO’s proposal states that “The calculation will be done at the DR provider level, rather than at the individual resource-aggregation’s qualifying capacity calculation.”[1]

While SCE’s supply-side DR resources are not listed on its (CAISO) RA and supply plans, and the rules would not apply to SCE’s DR resources since SCE is CPUC jurisdictional, SCE is concerned that this proposal would de-value a diverse DR portfolio (e.g., DR from controllable devices, battery energy storage, electric vehicles, behavioral DR, etc.). Because DR load reduction can be achieved when customers reduce or shift their electricity consumption and/or increase the output of their behind-the-meter (BTM) generation (e.g., battery energy storage), a DRP that has a diverse DR portfolio would receive a reduced capacity value of its more reliable DR resources, such as battery energy storage, because of other more variable DR resources, such as behavioral or weather-sensitive DR load reduction. Since a DRP can separate the types of DR into different Resource IDs, CAISO should consider applying the performance-based UCAP by Resource Level, not at the DRP level.

[1]       “California ISO Resource Adequacy Modeling and Program Design (RAMPD) Modeling and Default Rules (Track 1) Draft Final Proposal,” dated August 25, 2025, at p. 20.

2. Please provide your organization’s feedback on the proposed UCAP default QC methodology, specifically regarding the selection of supply cushion hours and the inclusion of natures of work in unit-specific forced outage rates.

SCE has no comments at this time.