Rasat
Does this electricity price forecast beat a simple rule of thumb?
For people who work in the sectorIn plain languageFor machines (JSON)
What is being measured
Electricity is bought and sold a day in advance. A price for every hour of tomorrow is set today. A forecast tries to work those prices out in advance.
The comparison uses two simple rules anybody could apply. The first takes tomorrow’s hourly prices to be the same as the last day whose prices are known. The second uses the price at the same hour a week earlier. Each day, the forecast has to beat whichever of those two came closer to the prices that actually happened.
Each forecast goes on the record first, and the outcome becomes known afterwards. The record keeps the forecast as it was first written.
The answer so far
So far, yes. The forecast came closer on 19 of the 27 days scored than the simple rule did, and its average error across the whole period is lower too.
These figures are preliminary for now. The record needs 60 days before it counts as evidence. It is 33 days short of that.
The days, sorted by the difference
This month
- September 2026
- 20 more days went on the record this month. On 14 of them the forecast came closer to the prices that actually happened than the simple rule did.
What this record covers, and what it does not
- The record covers days that have already happened. For those days it shows which of the two came closer to the prices that actually happened.
- Why the price moved is outside what this record covers.
- Only days with every hourly price published are counted. The same condition applies to days on which the forecasts were very close to what happened.
- This record covers the price only. The other quantities are measured separately and a separate record is kept for each.
Where this comes from
- Source
- EPİAŞ Şeffaflık Platformu (Piyasa Takas Fiyatı). The forecasts and their scores are recorded by Rasat.
- These figures were read at
- 21.09.2026 02:09 TRT
The same record, in more detail
The method, the figures for every day and the limits of what can be concluded from this record are on the other version of this page.