​ISO-NE's 2026 Long-Term Load Forecast projects net load rising 36.8% from 117 TWh in 2026 to 160 TWh in 2046. This 43 TWh increase is the smallest absolute addition on the Eastern Interconnect, but the additions will fundamentally change New England’s system shape.

Winter peak overtakes summer in 2036 under the central-peak forecast (50/50 forecast) and in 2035 under high-peak forecast (90/10). Heat pumps (HPs) add 12.4 GW to the winter peak by 2046 while electric vehicles (EVs) add another 5.9 GW. Large loads, including data centers, add just 132 MW.

For BESS, the load forecast opens two winter opportunities: intraday volatility driven by HPs and amplified cold snaps. HPs cycle hardest in the coldest hours (generally morning) and reduce output when houses are warm. EVs concentrate charging in defined overnight windows. The combination sharpens winter days into a two-hump shape with steeper ramps into morning and evening than today's relatively flatter winter profile. A winter-peaking system makes cold snaps more stressful on the grid. Both favor short-to-medium duration batteries dispatching into winter price signals.

Key takeaways

• Winter peak load grows 84% to 38 GW by 2046, a 3.1% CAGR. This is a higher growth rate than that of PJM, MISO, or NYISO. HPs deliver 72% of net winter peak growth.
• Large loads contribute just 132 MW to ISO-NE peak load in 2046, versus roughly 35 GW in PJM and 32 GW in MISO.
• The 2026 Capacity, Energy, Loads, and Transmission (CELT) Report marks the second consecutive cut to the load forecast: 2033 net load is down 11.5% across two cycles.
• Massachusetts adds 7.5 GW to winter peak by 2046, 44% of ISO-NE's total. However, northern states have the fastest growth rates, led by Vermont with 114% growth in winter peak by 2046.
• Winter load growth will provide stronger winter BESS earning window from greater intraday volatility and amplified cold snaps.

ISO-NE introduces the least new load of any Eastern ISO, but projects the most stark seasonal transformation

Net annual load in ISO-NE rises 36.8% from 117 TWh in 2026 to 160 TWh in 2046, a 1.58% CAGR. Every neighboring ISO adds more load: NYISO adds 45 TWh, PJM adds 811 TWh, and MISO adds 426 TWh. PJM and MISO growth comes from large-load adjustments. NYISO’s growth is attributed to a mix of electrification and large loads. Heating and transportation electrification accounts for 107% of ISO-NE load growth (after behind-the-meter solar reductions to gross load), bolstered by state policy goals.

By 2046, ISO-NE adds only 132 MW of large loads. In comparison, PJM adds 75 GW of large-load peak by 2046, MISO adds 32 GW, and NYISO adds 2.3 GW. ISO-NE's large load additions are a fraction of those in other Eastern ISOs. New England has seen little interest from large-scale data center developers due to high electricity costs and restrictive local environments. Several state legislatures have introduced bills limiting their development (Maine LD 307, New Hampshire HB 1265, Vermont S.205).

Winter peak overtakes summer in 2036 under the central-peak forecast, 2035 under high-peak forecast

Although ISO-NE adds the least load of any Eastern ISO, the system becomes winter peaking the fastest. The winter net peak rises 84% to 37.6 GW by 2046, an 3.1% CAGR. By 2046, the winter peak exceeds summer by 21% under central-peak scenario and 28% under high-peak scenario, reaching over 42 GW—nearly double the 2026 winter peak.

Heat pumps deliver 72% of net winter peak growth while data centers add 0.8%

ISO-NE's net winter peak grows 17.1 GW between 2026 and 2046. Heat pumps add 12.4 GW and EVs add 5.9 GW; base demand and behind-the-meter offsets shave the gross back by only 0.7 GW due to reduced PV performance in winter. In winter, base demand declines, primarily due to energy efficiency gains.

Summer peak growth is not attributed to heating electrification. Base demand (including cooling) drives most of the growth and heat pumps barely contribute, because they serve heating loads, not cooling.

The New England load forecast is built bottom-up from six components (Base growth, EVs, HPs, Large Loads, behind-the-meter (BTM) photovoltaics (PV), and BTM BESS), extending the 2026 CELT report's 10-year horizon out to 2046.

Downward revisions to ISO-NE’s electrification adoption lowers load forecast, but the implied pace remains ambitious

Since heat pumps and EVs carry projected demand, the forecast relies on adoption materializing at scale. ISO-NE has revised both in successive cycles, but the decreased projections still imply an aggressive ramp.

EV annual load contribution in 2033 fell nearly 70% across the three most recent load forecasts. The federal EV tax credit expiration in September 2025 and falling EV sales trends underpin this trend.

Heat-pump revisions show a delayed adoption projection with modest decreases in the outlook long term. HP annual load contribution fell 37% from the CELT 2024 to 2026. Meanwhile the HP contribution to winter peak has a lesser decline. The 2030 winter HP peak fell 20% across three cycles, while the late horizon extended by one year. Overall, the HP buildout is trimmed near term and pushed out long term.

Massachusetts raises winter peak the most by 2046, but Vermont and Maine grow fastest

Massachusetts adds 7.5 GW of net winter peak by 2046, accounting for 44% of the 17.1 GW added across ISO-NE. Connecticut ranks second at 3.3 GW; the remaining states split the rest.

By growth rate, the order inverts. Vermont nearly doubles its winter peak at 114%. Maine reaches 108%. The two largest zones, Connecticut and Northeast Massachusetts, grow just 69%, the slowest in the system.

Heating geography contributes to the peak growth rate differences. Northernmost states (Vermont, Maine, and New Hampshire) have colder winters with more residential heating load left to electrify. Boston-area NEMA, which is denser and more commercial, starts from a higher peak and has less marginal heating headroom to capture.

The BESS opportunity: heat pumps create intraday winter volatility

ISO-NE's load forecast changes the winter dispatch window alongside growing the absolute peak due to the hourly load shapes from HPs and EVs. Winter days are projected to have a two-hump shape with increasingly steeper ramps in the morning and evening.

For storage, those steep ramps are a signal. A flatter winter load offers little intraday spread. The dual-peak shape presents more energy arbitrage opportunities for a fast-dispatching battery in contrast to ISO-NE’s historical winter load profile.

Extended cold snaps in a winter-peaking system further amplify price signals from steep ramps. These events drive most of today's winter BESS revenue. February 2026 was the clearest recent example: gas pipeline constraints pushed oil-fired generation up sharply. In turn, TB4 spreads at the Internal Hub averaged $404/MW-day across the cold stretch.

Capturing that volatility requires available capacity at the ramps. Assets that are fully charged ahead of the morning peak and recharged before the evening window are positioned to dispatch into the highest-value intervals.