Daily guide · EE · 2026-09-08
The number your Estonia price model has to beat this week
All numbers computed from the week's published auction results at publication and frozen — this article is an honest snapshot, not a page that rewrites itself. Reproduce via the API.
The number to beat this week is 32,47 EUR/MWh, the average persistence MAE across the five days. That is the naive benchmark your model has to clear: use yesterday's price as today's forecast, and you carry that error. Any model that cannot beat 32,47 EUR/MWh is not adding value this week.
The daily spread is what stands out. Persistence MAE ranges from 13,56 EUR/MWh on 6 September to 58,85 EUR/MWh on 5 September, and the day means swing just as hard, from 75 EUR/MWh on 4 September to 10,56 EUR/MWh on 6 September. The hardest day for persistence, 5 September, pairs a high error of 58,85 with a low day mean of 16,16 EUR/MWh, a sign of a sharp level shift that naive carry-forward misses.
For context, our own scorecard over the trailing window shows an MAE of 38,01 EUR/MWh against a baseline of 47,29 across 352 days. The takeaway: judge your model day by day this week, and treat 5 September as the test that separates real signal from carry-forward luck.
Before you train anything on EE prices, you need the number your model has to beat. In day-ahead forecasting that number is naive persistence: predict that tomorrow's curve equals today's. It costs nothing, needs no features, and it is embarrassingly hard to beat — because power prices are dominated by daily and weekly rhythms that persistence gets for free. We computed it on this exact week:
32.47 €/MWh
persistence MAE on this week's EE auctions — the bar. A model that can't get under this adds negative value.
38.01 vs 47.29
our live production model vs its persistence baseline, trailing 352 scored days — the same scoring, in public, on /accuracy
Why the bar moves: this week, day by day
| Target day | Day mean €/MWh | Persistence MAE |
|---|---|---|
| 2026-09-03 | 51.33 | 39.61 |
| 2026-09-04 | 75 | 26.71 |
| 2026-09-05 | 16.16 | 58.85 |
| 2026-09-06 | 10.56 | 13.56 |
| 2026-09-07 | 26.51 | 23.61 |
Calm days make persistence look unbeatable; transition days (weather swings, weekend boundaries) are where a real model earns its keep. Averages hide this — always look at the daily distribution.
The part everyone gets wrong: the features
The model is the easy half. The silent killer is lookahead: training on weather reanalysis instead of the forecast that existed at issue time, on revised load data, on same-day flow averages that hadn't finished flowing at the auction. Backtests built that way look brilliant and die in production. Our Quant feature matrices serve one row per (issue day, 15-minute slot) where every column is documented with when it became knowable — the same rows our own models train on:
curl -H "Authorization: Bearer YOUR_KEY" \
"https://voltcast.com/api/v1/features/EE?issue_from=2026-07-01&horizons=1"
# columns include: price lags · previous-run weather (as forecast AT issue)
# · grid drivers (lagged to last COMPLETE day) · day-ahead load/RES forecasts
# · ensemble spread · target_price label (null until matured, never imputed)
Start with Home: the persistence bar above needs nothing but GET /v1/prices/EE
(Home covers one zone of your choice). When your walk-forward backtest beats it honestly, prove it in public on the
Forecast Bench — forward-blind, open scoring code, and our
own model is auto-entered under identical rules. If you'd rather buy the plumbing than build it,
contact us about ML-grade data access.
Method & citation. Prices are hourly means of published day-ahead auction results (native 15-minute periods averaged; ENTSO-E/SMARD, attributed). Wholesale-price component only — grid fees and taxes come on top and vary by supplier. Cite as "Voltcast Research, voltcast.com/guides/quant-features-ee-2026-09-08".
Next issue 2 Oct
The European power roundup. Once a month. Real numbers.
Negative-price records, the month's widest spreads, the zones we failed to forecast — computed from the same production recorders as the API. No filler. Unsubscribe in one click.
- ✓ Negative hours and the cheapest zones
- ✓ The month's widest day-ahead spreads
- ✓ Forecastability ranking — where we missed
Get the next issue in your inbox.
No account. Double opt-in. Nothing else, ever. Privacy.