A utility can own 120 megawatts of installed nameplate capacity and still have much less capacity available under a stated outage condition. The short answer is to keep nameplate, seasonal rating, current availability, dependable or firm contribution, reserve and demand as separate quantities. Each depends on a time, system boundary and rule.
Consider three fictional units rated 50, 40 and 30 MW. Their installed nameplate sum is 120 MW. If the 50 MW unit is unavailable in the studied N-1 condition, the remaining nameplate sum is 70 MW. If a planned 20 MW derating simultaneously affects the 40 MW unit, the arithmetic available sum becomes 50 MW. Calling all three results ‘capacity’ hides the event and can make a reliable plan look larger than it is.
The examples teach ledger discipline. They are not operating limits, dispatch instructions, reliability certifications or investment advice. Terms such as firm capacity, accredited capacity and dependable capacity can have jurisdiction-specific definitions and resource-specific methods. Use the applicable utility, market operator, regulator and engineering standards for a real system. This article also distinguishes capacity availability from capacity factor, which measures energy produced over a period relative to a maximum and answers another question.
This guide develops a focused reader question within the Utilities subject guide. It adds an explicit evidence chain, original worked examples, changed-context checks and limits. Read it with The Gold Standard of Education | Learning That Transfers into Life and the How X Works.
Every dataset and calculation below is an invented teaching case. The article does not certify a field survey, laboratory result, website implementation, utility plan or operational decision. Use the current method, responsible specialist and applicable authority for a real decision.
eduKateSG · Utilities
Find your next learning step
Keep nameplate, seasonal capability, current availability, firm contribution, outages, deliverability, demand and reserve rules separate.
ROUTE 1 · CHAPTERS 1–6
Name the capability
Define capacity quantities, unit states, outages, deratings, demand and reserve.
ROUTE 2 · CHAPTERS 7–12
Protect adequacy logic
Separate reserve, capacity factor, firm rules, seasons, net output and deliverability.
ROUTE 3 · CHAPTERS 13–18
Change the system context
Test overlaps, dependencies, storage, variable supply, demand response and imports.
ROUTE 4 · CHAPTERS 19–22
Audit and continue
Use scenarios, repair misconceptions, preserve the worksheet and follow sources.
Chapter contents
Name the capability · Chapters 1–6
1. Name every capacity quantity
4. Add a planned derating without double counting
Protect adequacy logic · Chapters 7–12
7. Distinguish reserve margin from unused output
8. Do not confuse capacity factor with firm capacity
9. Treat firm contribution as method-dependent
Change the system context · Chapters 13–18
13. Changed context: overlapping outages
14. Changed context: shared dependencies
15. Changed context: storage duration
16. Changed context: variable resources
Audit and continue · Chapters 19–22
19. Quantify uncertainty and scenarios
20. Repair four common misconceptions
Capacity is incomplete without a qualifier. Nameplate is a registered or design rating; net seasonal capability can differ; available capacity reflects current conditions; firm or accredited contribution follows an applicable reliability rule.
Begin with a measurement ledger. Name the quantity, unit, observational unit, time window and transformation rule. Mark every input as observed, stipulated, estimated or assumed. Read the ledger as a sentence before calculating; if the sentence changes when the denominator changes, the arithmetic belongs to a different claim. Keep the raw rows beside every derived column so another reader can reproduce the step. Record exclusions and missing values rather than letting software silently shorten the denominator. If two sources use the same label differently, preserve both definitions and postpone the comparison until they can be reconciled.
Three units carry fictional nameplates of 50, 40 and 30 MW. Adding them gives 120 MW installed nameplate. This arithmetic does not establish that 120 MW is deliverable at the studied hour.
Before moving on, record the observation that would distinguish the two most plausible explanations. If no feasible observation exists, narrow the claim instead of pretending the ambiguity has disappeared.
A changed-context transfer check is to replace nameplate with net summer capability and identify which source record must change State what would change the result and what the example cannot establish. That limit belongs inside the answer.
Use one row per resource with rating basis, time, status, derating, fuel or energy constraint, network condition and counted contribution. A total is reproducible only when every row uses compatible definitions.
Keep the mechanism beside the number. Ask what physical, biological, statistical or protocol relationship connects the observation to the conclusion, and which conditions could break that link. Draw one arrow for each inferential step, then challenge every arrow with a plausible changed setting. A mechanism sketch should identify where selection, dependence, timing or an external constraint enters. It should also show the point at which the available evidence stops. This makes the next measurement purposeful: it tests one uncertain link instead of collecting more convenient data from a link that is already well described.
Unit A is 50 MW available; B is 40 MW available; C is 30 MW planned out. The status-aware available sum is 90 MW, while installed nameplate remains 120 MW.
Before moving on, record the observation that would distinguish the two most plausible explanations. If no feasible observation exists, narrow the claim instead of pretending the ambiguity has disappeared.
A changed-context transfer check is to return C to service at a 10 MW derating and recompute both totals State what would change the result and what the example cannot establish. That limit belongs inside the answer.
An N-1 study removes one relevant element under its defined contingency. The remaining capacity depends on which element is lost and whether other limits are active.
Use an independent check that follows a different route. Reconstruct totals from components, compare a bound with a point estimate, inspect a distribution behind an average, or solve a changed case from the original definitions. Disagreement is useful because it locates the definition, unit or boundary that needs repair. Preserve the first result while investigating rather than overwriting it. The audit trail should show which input or rule resolved the difference. Agreement between two calculations is strongest when they do not merely repeat the same assumption in different notation.
With 50, 40 and 30 MW units, loss of the largest unit leaves 70 MW of nameplate before other deratings. Loss of the 30 MW unit would leave 90 MW. Saying ‘N-1 leaves 70’ silently assumes the largest-unit case.
Before moving on, record the observation that would distinguish the two most plausible explanations. If no feasible observation exists, narrow the claim instead of pretending the ambiguity has disappeared.
A changed-context transfer check is to change the credible contingency set and state why the minimum remaining total can change State what would change the result and what the example cannot establish. That limit belongs inside the answer.
CHAPTER 4 OF 22 · Name the capability
4. Add a planned derating without double counting
An outage removes a unit or capability; a derating reduces it. Apply each event to the correct row and interval. Do not subtract the same unavailable megawatts again from a total already based on status-aware rows.
Match the claim to the evidence. A tidy decimal cannot repair selective observation, dependence or a missing protocol condition. Report the result, the conditions that make it interpretable and the next observation needed for the actual decision. Keep the numerical conclusion separate from the practical conclusion. Add a brief boundary statement naming the population, representation, system or interval to which the result applies. Then name one tempting stronger claim that the evidence does not support. This pair of sentences prevents a precise local answer from becoming an unsupported general rule.
After the 50 MW unit is out, the 40 MW unit is derated by 20 MW and the 30 MW unit remains available. The available sum is 0+20+30=50 MW, not 70-20-50.
Before moving on, record the observation that would distinguish the two most plausible explanations. If no feasible observation exists, narrow the claim instead of pretending the ambiguity has disappeared.
A changed-context transfer check is to restore the large unit while keeping the derating and recompute from rows State what would change the result and what the example cannot establish. That limit belongs inside the answer.
CHAPTER 5 OF 22 · Name the capability
5. Separate planned from forced unavailability
Planned outages are scheduled; forced outages arise unexpectedly under defined reporting rules. Both can affect availability, but their predictability and planning treatment differ.
Begin with a measurement ledger. Name the quantity, unit, observational unit, time window and transformation rule. Mark every input as observed, stipulated, estimated or assumed. Read the ledger as a sentence before calculating; if the sentence changes when the denominator changes, the arithmetic belongs to a different claim. Keep the raw rows beside every derived column so another reader can reproduce the step. Record exclusions and missing values rather than letting software silently shorten the denominator. If two sources use the same label differently, preserve both definitions and postpone the comparison until they can be reconciled.
A maintenance plan removes Unit C next Tuesday. A separate unexpected trip of Unit A occurs today. Combining them into one annual unavailable total hides timing and cannot answer either hour’s adequacy question.
Before moving on, record the observation that would distinguish the two most plausible explanations. If no feasible observation exists, narrow the claim instead of pretending the ambiguity has disappeared.
A changed-context transfer check is to move the planned outage away from peak demand and describe the changed planning consequence State what would change the result and what the example cannot establish. That limit belongs inside the answer.
Available capacity alone does not establish adequacy. Compare it with demand plus the reserve or reliability requirement defined for the same boundary and time.
Keep the mechanism beside the number. Ask what physical, biological, statistical or protocol relationship connects the observation to the conclusion, and which conditions could break that link. Draw one arrow for each inferential step, then challenge every arrow with a plausible changed setting. A mechanism sketch should identify where selection, dependence, timing or an external constraint enters. It should also show the point at which the available evidence stops. This makes the next measurement purposeful: it tests one uncertain link instead of collecting more convenient data from a link that is already well described.
At an hour with 45 MW demand and a stipulated 10 MW reserve requirement, 50 MW available meets demand but misses demand-plus-reserve by 5 MW. The system is not described fully by saying supply exceeds load.
Before moving on, record the observation that would distinguish the two most plausible explanations. If no feasible observation exists, narrow the claim instead of pretending the ambiguity has disappeared.
A changed-context transfer check is to lower demand to 35 MW while preserving the same reserve rule State what would change the result and what the example cannot establish. That limit belongs inside the answer.
CHAPTER 7 OF 22 · Protect adequacy logic
7. Distinguish reserve margin from unused output
Reserve margin can be a planning ratio based on capacity and peak demand, while operating reserve is a current capability meeting defined response requirements. Idle nameplate is not automatically usable reserve.
Use an independent check that follows a different route. Reconstruct totals from components, compare a bound with a point estimate, inspect a distribution behind an average, or solve a changed case from the original definitions. Disagreement is useful because it locates the definition, unit or boundary that needs repair. Preserve the first result while investigating rather than overwriting it. The audit trail should show which input or rule resolved the difference. Agreement between two calculations is strongest when they do not merely repeat the same assumption in different notation.
A 30 MW unit is offline for maintenance. Its unused 30 MW is not operating reserve merely because it is not generating. Status, start capability, response time and applicable rules determine counted reserve.
Before moving on, record the observation that would distinguish the two most plausible explanations. If no feasible observation exists, narrow the claim instead of pretending the ambiguity has disappeared.
A changed-context transfer check is to put the unit on a qualified standby state and identify the evidence still required State what would change the result and what the example cannot establish. That limit belongs inside the answer.
CHAPTER 8 OF 22 · Protect adequacy logic
8. Do not confuse capacity factor with firm capacity
The related article How Capacity Factor Works measures actual energy over a period relative to continuous output at the capacity denominator. It does not by itself provide a reliability contribution at a critical hour.
Match the claim to the evidence. A tidy decimal cannot repair selective observation, dependence or a missing protocol condition. Report the result, the conditions that make it interpretable and the next observation needed for the actual decision. Keep the numerical conclusion separate from the practical conclusion. Add a brief boundary statement naming the population, representation, system or interval to which the result applies. Then name one tempting stronger claim that the evidence does not support. This pair of sentences prevents a precise local answer from becoming an unsupported general rule.
A 50 MW unit generates 219,000 MWh in a 8,760-hour year, giving 50% capacity factor. That arithmetic does not state whether it is available during the system peak or under the studied contingency.
Before moving on, record the observation that would distinguish the two most plausible explanations. If no feasible observation exists, narrow the claim instead of pretending the ambiguity has disappeared.
A changed-context transfer check is to keep annual energy fixed while moving generation away from the peak period State what would change the result and what the example cannot establish. That limit belongs inside the answer.
CHAPTER 9 OF 22 · Protect adequacy logic
9. Treat firm contribution as method-dependent
Firm, dependable and accredited capacity can be defined through contractual, engineering or probabilistic rules. State the governing method instead of assigning a universal percentage by resource label.
Begin with a measurement ledger. Name the quantity, unit, observational unit, time window and transformation rule. Mark every input as observed, stipulated, estimated or assumed. Read the ledger as a sentence before calculating; if the sentence changes when the denominator changes, the arithmetic belongs to a different claim. Keep the raw rows beside every derived column so another reader can reproduce the step. Record exclusions and missing values rather than letting software silently shorten the denominator. If two sources use the same label differently, preserve both definitions and postpone the comparison until they can be reconciled.
A planning rule credits a fictional resource at 20 MW for one study. Its 50 MW nameplate remains a separate record. The 40% ratio is a result of that stipulated method, not an inherent constant.
Before moving on, record the observation that would distinguish the two most plausible explanations. If no feasible observation exists, narrow the claim instead of pretending the ambiguity has disappeared.
A changed-context transfer check is to change the reliability method and refuse to carry the old credit automatically State what would change the result and what the example cannot establish. That limit belongs inside the answer.
Temperature, cooling conditions, fuel availability and hydrology can change net capability. A winter rating should not be used for a summer adequacy claim without an applicable conversion.
Keep the mechanism beside the number. Ask what physical, biological, statistical or protocol relationship connects the observation to the conclusion, and which conditions could break that link. Draw one arrow for each inferential step, then challenge every arrow with a plausible changed setting. A mechanism sketch should identify where selection, dependence, timing or an external constraint enters. It should also show the point at which the available evidence stops. This makes the next measurement purposeful: it tests one uncertain link instead of collecting more convenient data from a link that is already well described.
Unit B is 40 MW in winter and 34 MW in summer. Summing winter ratings for a summer peak overstates the stated seasonal capability by 6 MW before outages.
Before moving on, record the observation that would distinguish the two most plausible explanations. If no feasible observation exists, narrow the claim instead of pretending the ambiguity has disappeared.
A changed-context transfer check is to compare two seasons while keeping installed nameplate unchanged State what would change the result and what the example cannot establish. That limit belongs inside the answer.
CHAPTER 11 OF 22 · Protect adequacy logic
11. Account for auxiliary loads and net versus gross output
Gross generator output includes electricity used by the plant; net output delivered to the grid can be lower. The system boundary must match demand and reserve calculations.
Use an independent check that follows a different route. Reconstruct totals from components, compare a bound with a point estimate, inspect a distribution behind an average, or solve a changed case from the original definitions. Disagreement is useful because it locates the definition, unit or boundary that needs repair. Preserve the first result while investigating rather than overwriting it. The audit trail should show which input or rule resolved the difference. Agreement between two calculations is strongest when they do not merely repeat the same assumption in different notation.
A unit produces 50 MW gross and consumes 4 MW in auxiliaries, leaving 46 MW net. Adding gross supply to net system demand mixes boundaries.
Before moving on, record the observation that would distinguish the two most plausible explanations. If no feasible observation exists, narrow the claim instead of pretending the ambiguity has disappeared.
A changed-context transfer check is to reduce auxiliary load to 2 MW and update net capability only State what would change the result and what the example cannot establish. That limit belongs inside the answer.
Generation capacity can exist but be constrained from reaching the relevant load by network limits or outages. Deliverable capacity needs a network boundary and transfer assumptions.
Match the claim to the evidence. A tidy decimal cannot repair selective observation, dependence or a missing protocol condition. Report the result, the conditions that make it interpretable and the next observation needed for the actual decision. Keep the numerical conclusion separate from the practical conclusion. Add a brief boundary statement naming the population, representation, system or interval to which the result applies. Then name one tempting stronger claim that the evidence does not support. This pair of sentences prevents a precise local answer from becoming an unsupported general rule.
A remote 30 MW unit is available, but only 10 MW can cross the constrained interface during the studied condition. Counting all 30 MW at the load pocket overstates deliverable supply by 20 MW.
Before moving on, record the observation that would distinguish the two most plausible explanations. If no feasible observation exists, narrow the claim instead of pretending the ambiguity has disappeared.
A changed-context transfer check is to repair the interface and identify which capacity quantity changes State what would change the result and what the example cannot establish. That limit belongs inside the answer.
CHAPTER 13 OF 22 · Change the system context
13. Changed context: overlapping outages
Outages that overlap in time affect the same capacity snapshot; outages in separate weeks should not be subtracted simultaneously. Use an interval timeline.
Begin with a measurement ledger. Name the quantity, unit, observational unit, time window and transformation rule. Mark every input as observed, stipulated, estimated or assumed. Read the ledger as a sentence before calculating; if the sentence changes when the denominator changes, the arithmetic belongs to a different claim. Keep the raw rows beside every derived column so another reader can reproduce the step. Record exclusions and missing values rather than letting software silently shorten the denominator. If two sources use the same label differently, preserve both definitions and postpone the comparison until they can be reconciled.
Unit A is out Monday–Wednesday and Unit B Thursday–Friday. A weekly list totals 90 MW of unit ratings touched by outages, but no hour loses both units. Snapshot availability follows the overlap, not the sum of event labels.
Before moving on, record the observation that would distinguish the two most plausible explanations. If no feasible observation exists, narrow the claim instead of pretending the ambiguity has disappeared.
A changed-context transfer check is to make the outages overlap on Wednesday and recompute that hour State what would change the result and what the example cannot establish. That limit belongs inside the answer.
CHAPTER 14 OF 22 · Change the system context
14. Changed context: shared dependencies
Two nominally independent units can share fuel, cooling, switchgear, software or transmission. An N-1 unit ledger can miss a common-mode event.
Keep the mechanism beside the number. Ask what physical, biological, statistical or protocol relationship connects the observation to the conclusion, and which conditions could break that link. Draw one arrow for each inferential step, then challenge every arrow with a plausible changed setting. A mechanism sketch should identify where selection, dependence, timing or an external constraint enters. It should also show the point at which the available evidence stops. This makes the next measurement purposeful: it tests one uncertain link instead of collecting more convenient data from a link that is already well described.
Units B and C together provide 70 MW but rely on one fuel line. A line outage can remove both. Their separate nameplates do not establish independent firm capability.
Before moving on, record the observation that would distinguish the two most plausible explanations. If no feasible observation exists, narrow the claim instead of pretending the ambiguity has disappeared.
A changed-context transfer check is to add a second truly independent supply path and document its tested capacity State what would change the result and what the example cannot establish. That limit belongs inside the answer.
Storage has a power rating and an energy inventory. A 30 MW power rating cannot be counted for an arbitrary duration without the required megawatt-hours and state of charge.
Use an independent check that follows a different route. Reconstruct totals from components, compare a bound with a point estimate, inspect a distribution behind an average, or solve a changed case from the original definitions. Disagreement is useful because it locates the definition, unit or boundary that needs repair. Preserve the first result while investigating rather than overwriting it. The audit trail should show which input or rule resolved the difference. Agreement between two calculations is strongest when they do not merely repeat the same assumption in different notation.
A battery rated 30 MW with 60 MWh usable energy can sustain 30 MW for two hours in the simplified lossless case. A four-hour firm claim needs another energy account.
Before moving on, record the observation that would distinguish the two most plausible explanations. If no feasible observation exists, narrow the claim instead of pretending the ambiguity has disappeared.
A changed-context transfer check is to start at half state of charge and recompute supported duration State what would change the result and what the example cannot establish. That limit belongs inside the answer.
CHAPTER 16 OF 22 · Change the system context
16. Changed context: variable resources
Wind and solar nameplate does not equal output at every hour. Their reliability contribution depends on correlated weather, demand and the applicable accreditation method.
Match the claim to the evidence. A tidy decimal cannot repair selective observation, dependence or a missing protocol condition. Report the result, the conditions that make it interpretable and the next observation needed for the actual decision. Keep the numerical conclusion separate from the practical conclusion. Add a brief boundary statement naming the population, representation, system or interval to which the result applies. Then name one tempting stronger claim that the evidence does not support. This pair of sentences prevents a precise local answer from becoming an unsupported general rule.
A 50 MW solar plant can reach nameplate at a bright time yet contribute zero after sunset. Neither a single maximum nor annual average alone determines critical-hour firm contribution.
Before moving on, record the observation that would distinguish the two most plausible explanations. If no feasible observation exists, narrow the claim instead of pretending the ambiguity has disappeared.
A changed-context transfer check is to move system peak from afternoon to evening and reassess the relevant evidence State what would change the result and what the example cannot establish. That limit belongs inside the answer.
A resource can support adequacy by reducing demand under verified conditions, but baseline, availability, notice and rebound rules matter. It should not be added as generation without a clear boundary.
Begin with a measurement ledger. Name the quantity, unit, observational unit, time window and transformation rule. Mark every input as observed, stipulated, estimated or assumed. Read the ledger as a sentence before calculating; if the sentence changes when the denominator changes, the arithmetic belongs to a different claim. Keep the raw rows beside every derived column so another reader can reproduce the step. Record exclusions and missing values rather than letting software silently shorten the denominator. If two sources use the same label differently, preserve both definitions and postpone the comparison until they can be reconciled.
A 10 MW demand-response programme is enrolled; only 7 MW is verified available for the event. Counting 10 MW firm ignores participation and measurement rules.
Before moving on, record the observation that would distinguish the two most plausible explanations. If no feasible observation exists, narrow the claim instead of pretending the ambiguity has disappeared.
A changed-context transfer check is to change the baseline method and state why the credited reduction can change State what would change the result and what the example cannot establish. That limit belongs inside the answer.
Imports can contribute when contracts, source availability and transfer capability align. A signed quantity is not automatically deliverable under every regional stress event.
Keep the mechanism beside the number. Ask what physical, biological, statistical or protocol relationship connects the observation to the conclusion, and which conditions could break that link. Draw one arrow for each inferential step, then challenge every arrow with a plausible changed setting. A mechanism sketch should identify where selection, dependence, timing or an external constraint enters. It should also show the point at which the available evidence stops. This makes the next measurement purposeful: it tests one uncertain link instead of collecting more convenient data from a link that is already well described.
A 20 MW firm purchase exists, but a simultaneous interface limit allows 12 MW. The study must use the applicable deliverable contribution, not the headline contract alone.
Before moving on, record the observation that would distinguish the two most plausible explanations. If no feasible observation exists, narrow the claim instead of pretending the ambiguity has disappeared.
A changed-context transfer check is to restore the interface while making source supply non-firm State what would change the result and what the example cannot establish. That limit belongs inside the answer.
Availability, load, renewable output and network conditions are uncertain and often dependent. Scenario tables or probabilistic studies can reveal combinations that one deterministic total misses.
Use an independent check that follows a different route. Reconstruct totals from components, compare a bound with a point estimate, inspect a distribution behind an average, or solve a changed case from the original definitions. Disagreement is useful because it locates the definition, unit or boundary that needs repair. Preserve the first result while investigating rather than overwriting it. The audit trail should show which input or rule resolved the difference. Agreement between two calculations is strongest when they do not merely repeat the same assumption in different notation.
Three scenarios show available capacity of 90, 70 and 50 MW against demands of 60, 65 and 45 MW. Only the last misses a stipulated 10 MW reserve despite lower demand, because overlapping derating dominates.
Before moving on, record the observation that would distinguish the two most plausible explanations. If no feasible observation exists, narrow the claim instead of pretending the ambiguity has disappeared.
A changed-context transfer check is to correlate high demand with one outage and explain why independent averaging can mislead State what would change the result and what the example cannot establish. That limit belongs inside the answer.
Installed capacity is not current availability; capacity factor is not firm capacity; unused nameplate is not automatically reserve; and non-overlapping outages should not be stacked into one snapshot. Restore the missing time and rule.
Match the claim to the evidence. A tidy decimal cannot repair selective observation, dependence or a missing protocol condition. Report the result, the conditions that make it interpretable and the next observation needed for the actual decision. Keep the numerical conclusion separate from the practical conclusion. Add a brief boundary statement naming the population, representation, system or interval to which the result applies. Then name one tempting stronger claim that the evidence does not support. This pair of sentences prevents a precise local answer from becoming an unsupported general rule.
Audit every total with columns for unit, rating basis, interval, status, derating, deliverability and counted contribution. If the rule is unknown, label the total as an arithmetic scenario instead of certified firm capacity.
Before moving on, record the observation that would distinguish the two most plausible explanations. If no feasible observation exists, narrow the claim instead of pretending the ambiguity has disappeared.
A changed-context transfer check is to delete the interval column and show how double subtraction becomes possible State what would change the result and what the example cannot establish. That limit belongs inside the answer.
Preserve source ratings, unit states, outage intervals, network constraints, demand, reserve rule, resource accreditation and calculation version. Recompute totals from rows rather than editing headline numbers.
Begin with a measurement ledger. Name the quantity, unit, observational unit, time window and transformation rule. Mark every input as observed, stipulated, estimated or assumed. Read the ledger as a sentence before calculating; if the sentence changes when the denominator changes, the arithmetic belongs to a different claim. Keep the raw rows beside every derived column so another reader can reproduce the step. Record exclusions and missing values rather than letting software silently shorten the denominator. If two sources use the same label differently, preserve both definitions and postpone the comparison until they can be reconciled.
A reviewer reproduces 120 MW installed, 70 MW after the largest-unit loss and 50 MW after the overlapping 20 MW derating. Each transition names the event and avoids a hidden denominator.
Before moving on, record the observation that would distinguish the two most plausible explanations. If no feasible observation exists, narrow the claim instead of pretending the ambiguity has disappeared.
A changed-context transfer check is to change one unit’s seasonal rating and propagate it through every scenario State what would change the result and what the example cannot establish. That limit belongs inside the answer.
The US Energy Information Administration distinguishes nameplate, net summer and net winter capacity. The NERC glossary provides reliability-standard definitions, while the US Department of Energy capacity explainer introduces generation capacity. Return to the Utilities root for delivery, metering, maintenance and recovery.
Keep the mechanism beside the number. Ask what physical, biological, statistical or protocol relationship connects the observation to the conclusion, and which conditions could break that link. Draw one arrow for each inferential step, then challenge every arrow with a plausible changed setting. A mechanism sketch should identify where selection, dependence, timing or an external constraint enters. It should also show the point at which the available evidence stops. This makes the next measurement purposeful: it tests one uncertain link instead of collecting more convenient data from a link that is already well described.
The transferable habit is to keep a capability rating separate from availability under stated conditions. It applies to water pumps, hospital beds, computing clusters and transport fleets.
Before moving on, record the observation that would distinguish the two most plausible explanations. If no feasible observation exists, narrow the claim instead of pretending the ambiguity has disappeared.
A changed-context transfer check is to apply the ledger to a pumping station without importing electric-market definitions State what would change the result and what the example cannot establish. That limit belongs inside the answer.