Every desk wants a proven track record. There is nowhere to build one.
Foretec runs live competitions on real European electricity market data. Participants forecast, bid across day-ahead and intraday, and carry the imbalance. Every entry is scored on the money it would actually have made.
96 settlement periods · one day of imbalance prices
You run a wind farm. How much of its output do you sell day-ahead?
where you end up, day-ahead only where you could end up if you also trade intraday
How this is scored
Sell more than you generate and you buy the shortfall back at the imbalance price; sell less and the surplus settles cheap. The optimal commitment therefore sits below your median forecast. The shaded area is what a second trade in intraday could still recover once the forecast has sharpened. It widens as your day-ahead commitment gets worse, because there is more error left to correct.
100% = selling actual generation at the day-ahead price. Day-ahead €80/MWh, intraday €72 sell / €90 buy, imbalance €108 short / €58 long with occasional spikes. Forecast median 55.5 % of capacity. Illustrative conditions, not a specific market or trading day.
The demo assumes a well-calibrated forecast and a known price asymmetry, so it isolates the commitment decision from the forecasting problem. A real edition is harder: participants supply the distribution as well as the position, and nobody knows in advance which way the system will be short.
The same shortlist circulates, and the evidence arrives after the hire.
Typical agency fee for one quantitative hire in European power trading, at 15–25 % of first-year salary, for a candidate whose models nobody has measured.
What a European transmission system operator already pays forecasters, per market, ranked on accuracy alone. Crowdsourced forecasting is operational, not hypothetical.
There is no recurring, multi-sponsor venue dedicated to European energy trading talent. Firms run their own, occasionally, behind closed doors.
Every job advert in this market asks for a proven track record, and there is no venue in which anyone can build one. Foretec is that venue: neutral, recurring, and open to the whole market rather than to one employer.
How an edition works
-
01
Join and warm up
Entry is open all year. New participants start with a warm-up month on live data: the same gates and settlement, but no prizes. There is no qualifying round on historical data, because every score on Foretec comes from forecasts committed before the outcome was known. We publish the settled data series, a baseline model and tutorials, and Energy Bench gives everyone strong open forecasts to build on.
-
02
Commit, every day
Before each gate, submit a probabilistic forecast and a position, then revise it intraday. Every submission is timestamped and locked on our server before the market it targets closes. Nothing can be changed or resubmitted after the fact.
-
03
Settle against the real market
Positions settle against real market outcomes:
- Day-ahead at the published clearing price.
- Intraday against real exchange prices.
- Any remaining imbalance at the imbalance price, including the effect your own imbalance would have had on it.
Your score is the share of perfect-foresight revenue you capture.
-
04
Monthly awards, annual title, meet the top
Each month has its own leaderboard and award. The annual table decides the main prize, so one lucky month or one season does not decide the winner. Sponsors see the ranked pool throughout the year and can contact anyone. Participants compete under pseudonyms and choose when to reveal who they are.
What a submission looks like
96 periods, one row each. A single column of positions is a valid entry; the optional columns are where the edition is won.
Read the full mechanics: gates, settlement formulas, the price-impact model, what the platform publishes and the edition calendar.
The aggregate
The crowd competes too
Every submission is aggregated into a single meta-forecast that enters the leaderboard under its own name, scored on both tracks like any other participant.
Live now
The forecasting half is already running, in public.
Energy Bench forecasts European power markets every day before the day-ahead gate. Every model gets the same frozen public inputs, and every forecast is scored against the grid operators’ actuals.
of forecasting models: time-series foundation models, gradient boosting, and the grid operators’ own forecasts.
Price, wind, solar and load across multiple European bidding zones, every quarter-hour.
Locked before the 12:00 gate since 2 October 2026. Every forecast and score is open data.
A standing benchmark on your market, and the people who top it.
You choose the problem: the bidding zone, the asset mix, the products. Several hundred modellers work on it all year, in public, scored on realised money. You keep the aggregate, the leaderboard, and first access to whoever is at the top of it.
Anchor partner · one per edition
Set the problem
- Choose the target market: bidding zone, asset mix and which products are traded
- Seat on the edition’s advisory board, chairing the rules and scoring decisions
- Platform access to the individual and collaborative input of the top ten, period by period
- Exclusive first window on the ranked pool, ahead of every other sponsor
- Title branding on the edition
Recruiting partner
Meet the board
- Seat on the edition’s advisory board
- Platform access to the individual and collaborative input of every participant above baseline
- Full leaderboard access once the anchor window closes
- Opt-in profiles and introductions
- Presence on the leaderboard and in participant mailings
Ecosystem partner
Reach the field
- Opt-in profile pool excluding the finalists
- Visibility with several hundred quantitative modellers
- Distribute API credits, data trials or tooling to participants
What sponsors see, and what they don't
Participants keep their code. Sponsors work with what is submitted to the platform: individual input, participant by participant, and the collaborative input aggregated from it. Never the models behind either. That is what makes strong people willing to enter, and it is the more useful material anyway.
The advisory board runs for one edition. Anchor and recruiting partners each hold a seat; the anchor partner chairs it. It meets before the first live month and at each scoring milestone, and it decides rules, gates and scoring for that edition only. It sees nothing the tier does not already grant, and never the identity of a participant who has not revealed themselves.
Packages are priced per edition against what a single unverified hire costs today. Ask and we will send the numbers.
Find out when the first edition opens.
Occasional updates on dates, the edition spec, sponsorship and results. No more than one email a month, and nothing else ever.
Stored on our own server in the EU, used only for these updates, never shared, and deleted when you ask.
Thank you, you are on the list.