On a napkin
$250 $200 $150 $100 $50 Peak: $238/MWh DR event window LMP = Energy + Congestion + Losses 00:00 06:00 12:00 18:00 24:00 $/MWhLMP is the price that determines what every wholesale settlement actually pays out. The grid operator calculates it every five minutes at every transmission node — thousands of separate prices across each ISO — and three things drive what it costs at any given point: the cheapest generator running that interval, whether congestion is forcing more expensive units to dispatch locally, and electrical losses on the wires between generation and load. For buyers on index contracts, those three components determine the monthly bill. For buyers on fixed-price contracts, they determine what the retailer earned or lost serving the load — and ultimately what the next renewal looks like.
LMP = Energy + Congestion + Losses. Congestion is the volatile component. ERCOT West Hub vs. Houston basis ran $300+/MWh during certain 2024 wind events. PJM Western Hub vs. Dominion zone basis has structurally widened as Northern Virginia data center load outgrew available transmission. The February 2021 Texas winter storm cleared the $9,000/MWh administrative cap (since lowered to $5,000) across multiple intervals — almost entirely a congestion and scarcity event, not a fuel-cost event.
Every ISO that uses nodal pricing — PJM, MISO, ERCOT, SPP, CAISO, ISO-NE, NYISO — solves the same problem on essentially the same cadence. The day-ahead market clears once per day at hourly intervals, typically by 1 PM the day before delivery. Generators submit price-quantity offers; load serves submit demand bids; the optimization solves for the lowest-cost dispatch that respects every transmission and reliability constraint. Day-ahead LMPs are the prices that flow through most financial hedges, FTR settlements, and forward contracts.
The real-time market reruns the same optimization every five minutes against actual physical conditions — a generator that tripped, a cloud over a solar farm, a heat wave spiking residential AC load. Real-time LMPs can deviate sharply from day-ahead prices, particularly during stress events. The difference between day-ahead and real-time at the same node is called the real-time settlement adjustment, and it is the primary income or expense from virtual bidding strategies.
Nodal pricing calculates a unique LMP at every transmission bus. PJM alone has more than 10,000 priced nodes. Zonal pricing averages prices across broader geographic zones. Five of the seven US ISOs are nodal at settlement. ERCOT settles generation nodally but load zonally — a structural mismatch that creates persistent transfer risks.
For a buyer with a 5 MW facility on a single distribution feeder, the relevant question is not "what is the PJM Western Hub LMP" but "what is the LMP at the specific load aggregation point my retailer uses for settlement?" Those two numbers can differ by tens of dollars per MWh on a routine basis and by hundreds during constraint events. A retail contract benchmarked to the hub but settled at the node passes that difference through as a separate line item — or absorbs it into the next renewal pricing.
Basis is the price difference between two points on the grid — typically between a generator's location and a load center, or between a load zone and the regional hub. Persistent basis indicates persistent transmission constraints, and the trend in recent years has been worsening: data center load growth has concentrated geographically faster than transmission can be built to serve it, with the result that basis between generation-rich and load-rich zones has structurally widened.
The standard hedge is the Financial Transmission Right. FTRs are tradeable financial contracts that pay the holder the difference in LMP between two specified locations over a defined period. A C&I load with persistent exposure to a congested node can buy FTRs from a less constrained hub to the load's settlement point, neutralizing the basis exposure. FTR auctions clear monthly, seasonally, and annually depending on the ISO. The auction revenue is paid to LSE load through ISO surplus distribution; the FTR cash flows hedge against actual settlement congestion. For loads above a few megawatts in chronically constrained zones, FTRs are not optional sophistication — they are the difference between a predictable energy bill and one that surprises every congestion event.
The headline number most buyers track is not the average LMP but the tail. Day-ahead LMP averages across an ISO might be $35-50/MWh, but real-time can hit the administrative price cap ($5,000/MWh in ERCOT, $3,000/MWh in PJM, similar across the others) during scarcity. The February 2021 Texas storm cleared the cap repeatedly. The August 2020 CAISO rotating outages did the same. The Pacific Northwest June 2021 heat dome cleared the cap in CAISO. Each event was driven by a different combination of supply shortfall, demand surge, and transmission constraint — but the price signal in each case was unmistakable, and the bills paid by load to clear the market were enormous.
Most C&I customers are partially insulated from these spikes by fixed-price retail contracts. But the spikes flow into renewal pricing, capacity charges, and ancillary service uplift across the broader market. Buyers with real-time index contracts, demand response participation, or behind-the-meter generation see the spikes directly — and in DR especially, the spikes are the entire business model.
Three implications for buyers and developers. First, settlement risk is local. The hub price your retail contract benchmarks is not the price you pay for power at your specific point of withdrawal. Basis between hub and node can run hundreds of dollars per MWh during constraint events, and the trajectory of constraint frequency in most US ISOs is up, not down.
Second, congestion is now the dominant driver of LMP volatility, and it is structural rather than cyclical. The pattern is most acute in PJM Dominion, CAISO load pockets, and ERCOT load zones with rapid data center growth, but the underlying dynamic — load concentration outpacing transmission build — applies across the country.
Third, FTRs are the primary tool for managing the resulting basis exposure. For large industrial loads served at congested nodes, FTR strategy has shifted from optional sophistication to standard practice. The same applies, mirrored, on the generation side: developers signing PPAs at constrained generation nodes face symmetric basis risk that FTRs partially address.
Common questions
Related reading on The Outlet