See the miss. Before the market does.
SkyrisQ scores how likely your solar schedule is to be wrong, for every 15-minute interval. Weather models disagree, clouds move, prices spike. You get one number and the reason behind it, hours before gate closure.
plantsby risk
HU-PV-03risk 78 · cloud front HU-PV-01risk 46 · models split HU-PV-02risk 14 · clear skyECMWF and ICON disagree on cloud timing by ~1h. Cloud cover forecast rising in the west.
Illustrative data. The interface shows the intended product; live scoring ships with the beta.
34.6 billion HUF in two years, from solar schedules that missed by about 12%.
Every quarter-hour, Hungarian solar producers settle the gap between the schedule they sent the day before and what they actually produced. We priced that gap for the whole market, using MAVIR's single imbalance price against the day-ahead price.
- 63.8 bn HUF of adverse cost in the intervals where the miss lost money; 34.6 bn HUF net after the intervals where it happened to pay.
- A few extreme-price quarter-hours carry a large share: the top 1% of intervals hold about 29% of the adverse cost. That is why timing the risk matters.
- National totals from ENTSO-E day-ahead forecasts and actual generation, not any single plant. Latest months use MAVIR preliminary prices.
Sources: ENTSO-E Transparency Platform, MAVIR, HUPX. Own calculation, Jan 2025 to early Aug 2026.
Models disagree. That is the signal. →
When ECMWF, GFS and ICON agree, your forecast is safe. When they split, you carry risk. We turn that disagreement into a number, calibrated against what your plants actually produced.
Testing whether observed clouds help. →
Weather models predict clouds; satellites see them. We are testing whether European satellite cloud observations improve the score for the next few hours. It is added to the product only if our backtest against measured sunshine shows a gain.
One score per 15 minutes. →
Hungary settles imbalance every quarter hour. SkyrisQ scores every quarter hour, with the drivers attached, so trading desks act on it instead of arguing about it.
Backtested on real settlements. →
Every score is validated against actual generation and imbalance prices from the Hungarian market. You see forecast error and its cost in HUF, not a vendor accuracy claim.
Plugs into what you run. →
REST API and a dashboard. No new hardware, no site visit. Bring a plant location and an installed capacity, get a score.
Live today. In build. Next.
Live: multi-model forecast grid, calibration against actuals, market data pipeline. In build: risk score. Testing: satellite cloud layer. Next: API, alerts, more markets.
Join the SkyrisQ beta.
We are opening a closed beta to a small group of Hungarian solar producers and scheduling desks. You send your schedules and actuals, and we return the risk scores next to your real imbalance costs. It is a beta: we learn from your data, and your feedback shapes what we build. Access is by invitation.
Know your exposure before gate closure.
One command, one plant, one score. Tell us where your plants are and we set up your beta access.
ben@skyrisq.com →