Live market data
ISO-NE load forecast
ISO-NE's own five-minute load forecast against realized demand, split across the eight load zones — with behind-the-meter solar's fingerprint visible in the daytime dip.
ISO-NE's own five-minute load forecast against realized demand, split across the eight load zones — with behind-the-meter solar's fingerprint visible in the daytime dip.
How much power will New England need?
ISO-NE demand peaks at 18,581 MW on Fri, 6 PM, against a low of 11,584 MW — a swing of 6,996 MW across the window. Connecticut (the southwest corner of the pool) is the heaviest area at 4,489 MW, 24% of the system peak.
Every line above, and who produced it.
What settled — the realized values every forecast here is scored against.
The grid operator's own published forecast for these hours. Source: ISO New England.
The same window, split by load zone — each area’s own curve.
Every hour of the window, shaded by load.
| 0 | 6 | 12 | 18 | |||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Fri, 8/28 |
| Load zone | Peak (MW) | Peak hour | Low (MW) | Average (MW) |
|---|---|---|---|---|
| 4004 | 4,489 | Fri, 5 PM | 2,636 | 3,606 |
| 4008 | 4,041 | Fri, 4 PM | 2,559 | 3,395 |
| 4006 | 2,496 | Fri, 6 PM | 1,465 | 1,890 |
| 4007 | 2,275 | Fri, 6 PM | 1,492 | 1,788 |
| 4001 | 1,722 | Fri, 6 PM | 1,156 | 1,381 |
| Forecast | Avg miss (MW) | Avg miss % | Bias | Hours |
|---|---|---|---|---|
| ISO-NE load forecast (day-ahead) | 494 | 3.4% | −494 | 1 |
Each forecast as it stood 24 hours before the hour it predicted, against what settled. Smaller is better on every column. Bias is signed: positive means the forecast ran high.
| Forecast | Produced by | Method | Horizon |
|---|---|---|---|
| Forecast demand | ISO New England | ISO-NE five-minute zonal load forecast, summed to the system, and five-minute system load | Next 24 hours, five-minute |
ISO-NE publishes point forecasts with no confidence interval. Behind-the-meter solar makes cloudy-day errors systematically larger, and winter cold snaps move demand faster than the forecast cycle.
ISO New England runs the grid for the six New England states — about 15 million people. This page plots its published five-minute load forecast for the next 24 hours against realized system load, then splits the same forecast across the eight load zones: the three Massachusetts zones, Connecticut, Rhode Island, New Hampshire, Vermont, and Maine. The system line is the sum of those zones, so the chart and the table below it are one publication rather than two that disagree.
New England has one of the country's largest behind-the-meter solar fleets relative to its size. It never shows up as generation — it shows up as missing demand, carving a midday dip into the load curve that deepens every year. The forecast has to predict the weather twice: once for temperature, once for the rooftop solar the grid never meters.
Unlike most US grids, New England's tightest hours are winter evenings, when heating demand meets constrained gas supply. A cold snap moves both demand and the fuel available to serve it, which is why the market watches this forecast hardest in January.
curl "https://askthegrid.com/api/v1/grid/series?iso=ISONE&metric=load_forecast&zone=SYSTEM&hours=168"
Connect your agent to mcp.askthegrid.com and this forecast is a tool call away — the same host Claude and ChatGPT connectors already use.
It is ISO New England's hourly projection of electricity demand across the six-state grid, published up to three days ahead. The ISO schedules generation against it, and this page scores it against what actually settled.
System peaks run roughly 24-26 GW on the hottest summer afternoons — modest next to PJM or MISO, but served by one of the most import- and gas-dependent fleets in the country, which is what makes the forecast consequential.
Behind-the-meter solar. Tens of thousands of rooftop systems serve load the grid never sees, so metered demand sags in the middle of sunny days and rebounds as the sun sets. The effect is strong enough that ISO-NE forecasts it explicitly.
Plotted window Aug 28, 3:25 AM EDT → Aug 29, 3:20 AM EDT. Reissued through the operating day; the latest vintage for each hour is shown.
| Sat, 8/29 |
|---|
Darkest cells are the tightest hours. The window peaks at 18,581 MW on Fri, 6 PM, local to the market.
| 4002 |
|---|
| 1,710 |
| Fri, 5 PM |
| 997 |
| 1,395 |
| 4005 | 1,292 | Fri, 6 PM | 802 | 1,033 |
|---|
| 4003 | 672 | Fri, 7 PM | 413 | 535 |
|---|