From today to polling day in Poland
Abstract. We test the British pull-back towards the previous result on Polish polls at ten distances up to three years before the Sejm elections of 2011 to 2023. It helps: from two years out the previous result was closer than the polls, and the fitted curve beat the polls at seven of ten distances. The polling-day forecast moves today's shares about half-way back towards the 2023 results of the lists they descend from, which lifts Trzecia Droga's parties and lowers the right's splinters, and widens the ranges for the time left.
Summary
Everything else in the Poland section answers the question "what if the election were held today?". The next Sejm election is due by 11 November 2027 at the latest, 355 days after the most recent poll. This paper describes how we get from today to polling day.
In Britain the older parties finish much closer to their previous result than their mid-term polls suggest, and we pull them part of the way back (From today to polling day: how far British polls move before a general election). We ran the same test on Polish polls, at ten distances from one month to three years before the Sejm elections of 2011, 2015, 2019 and 2023. The pull-back helps: two years or more out, the previous result was a better guide than the polls, and a weighted average of the two was closer than the polls at most distances. So the polling-day forecast moves today's shares about half-way back towards the 2023 results of the lists they descend from, and widens the ranges for the time left: polls this far out have missed each party by about 4.1 points by the time votes were cast.
What has changed since this paper was written
Updated 9 November 2026. The rest of the paper is as first published, except where a sentence below says it was corrected.
- Only elections from 2000 shape the forecast (28 September). Earlier elections stay in the long view and on the history pages but no longer feed any test, correction or error size. In Poland the 1993 to 1997 seat check is still shown but no longer counts.
The evidence: polls at fixed distances before four elections
The evidence is every national poll in our series for the three years before each of the Sejm elections of 2011, 2015, 2019 and 2023, from Wikipedia's poll tables, with the don't-knows taken out. The series for 2011 is thin, 64 polls; the others have 290 to 650.
For each election we average the polls at 30, 61, 91, 182, 273, 365, 548, 730, 913 and 1,095 days before polling day, over a 60-day window centred on each point, averaging within each pollster first and then across pollsters with equal weight. Each record holds the line, the poll average, the line's result at the previous election and its result at this one, with the same 2023 lines as the rest of the Polish record. That gives 190 horizon points.
What the data show
| Days before polling day | Elections | Polls' miss | Previous result's miss | Pull-back's miss | RMSE |
|---|---|---|---|---|---|
| 30 | 4 | 2.10 | 4.26 | 1.96 | 2.47 |
| 61 | 4 | 2.77 | 4.26 | 2.48 | 3.32 |
| 91 | 4 | 2.97 | 4.26 | 2.42 | 3.47 |
| 182 | 4 | 3.23 | 4.26 | 3.08 | 3.79 |
| 273 | 4 | 3.01 | 4.26 | 3.13 | 3.68 |
| 365 | 4 | 3.46 | 4.26 | 3.61 | 4.01 |
| 548 | 4 | 3.57 | 4.26 | 3.80 | 4.26 |
| 730 | 4 | 4.50 | 4.26 | 4.15 | 5.39 |
| 913 | 4 | 6.05 | 4.26 | 4.42 | 7.06 |
| 1095 | 4 | 5.35 | 4.26 | 4.63 | 6.50 |
The previous result is a moderate guide. It missed each line by about 4.3 points, the same at every distance.
The polls are good close in and poor far out. A month out they missed by 2.1 points per line, a year out by 3.5, and two to three years out by 4.5 to 6. From two years out the previous result was closer than the polls.
The pull-back beats the polls at most distances. A weighted average of the two was closer than the polls at seven of the ten distances, by the most far out; at nine months to a year and a half it was slightly further out.
Election by election:
| Days out | 2011 | 2015 | 2019 | 2023 |
|---|---|---|---|---|
| 30 | 3.47 | 1.56 | 1.31 | 2.06 |
| 61 | 4.52 | 1.90 | 2.24 | 2.40 |
| 91 | 3.89 | 2.58 | 2.61 | 2.79 |
| 182 | 4.32 | 4.21 | 2.73 | 1.68 |
| 273 | 3.65 | 3.66 | 3.12 | 1.60 |
| 365 | 4.96 | 3.28 | 3.56 | 2.04 |
| 548 | 4.78 | 4.18 | 3.36 | 1.96 |
| 730 | 6.58 | 5.06 | 3.96 | 2.40 |
| 913 | 6.71 | 5.78 | 6.14 | 5.56 |
| 1095 | 6.33 | 6.11 | 7.23 | 1.74 |
Table: Polls' average miss per line at each distance and election, points.
2011 was the hardest at every distance, partly because its series is thin. 2023 was the easiest from six months out; at 913 days, the autumn of 2021, the polls were far from the result.
The pull-back curve and its test
The polling-day share is a weighted average of today's share and the previous result,
with the floor \(f\) and time constant \(\tau\) chosen by least squares over all horizon points. The best fit is
With a low floor and a long time constant, the weight falls slowly: a month out almost all of each line's change since the last election is kept; at a year, a little over half; at three years, about a quarter. The test, each election predicted from a curve fitted on the other three, beat the polls as they stood on average, so the curve is used.
Split parties
The pull-back works on the 2023 lines, not on today's parties. Each line's parties are added together, the total is moved towards the line's 2023 result, and the parties share the moved total in their ratio today:
- PiS and Rozwój Plus move towards PiS's 35.4%;
- Konfederacja and KKP towards Konfederacja's 7.2%;
- Lewica and Razem towards Lewica's 8.6%;
- the PSL, Polska 2050 and Centrum towards Trzecia Droga's 14.4%;
- KO towards its 30.7%.
The effect today is large for two of them. The right-wing splinters' combined vote, far above Konfederacja's 2023 result, is pulled down. Trzecia Droga's parties, far below their 2023 result, are pulled up, enough to lift the PSL towards 5% on polling day. Whether that is right depends on whether the break-up of Trzecia Droga is a lasting change or a mid-term dip, and the record cannot say: no line in it has broken up and recovered.
Drift
The polls' miss grows with the time left, and the polling-day forecast keeps that uncertainty. The root-mean-square miss of the pulled-back shares at each distance is fitted by
over the ten distances. At 355 days it is 4.1 points. Under 30 days it is scaled down in proportion to the days left, reaching zero on polling day, and it is never less than half a point.
In the simulation the drift is scaled by the square root of each party's share relative to 15%, so that large parties get more and small ones less, and combined with the party's final-poll spread:
Polling-day ranges
The polling-day forecast runs the simulation (Simulating Polish polling error: where the chances come from) with 1,000 runs and equation (4), under the default line-up, centred on the pulled-back shares:
| List | Share today | Share, polling day | Seats today | Seats, polling day | 80% range, polling day | Clears threshold, polling day |
|---|---|---|---|---|---|---|
| KO | 32.6 | 32.2 | 191 | 190 | 141–228 | over 99% |
| PiS | 22.0 | 25.2 | 131 | 150 | 110–189 | over 99% |
| Konfederacja | 14.0 | 10.1 | 73 | 46 | 3–74 | 90% |
| KKP | 8.8 | 6.4 | 41 | 20 | 0–47 | 69% |
| Lewica | 7.4 | 6.6 | 24 | 20 | 0–47 | 67% |
| Rozwój Plus | 4.7 | 5.4 | 0 | 13 | 0–41 | 60% |
| Razem | 4.3 | 3.9 | 0 | 0 | 0–21 | 33% |
| PSL | 3.4 | 6.0 | 0 | 21 | 0–42 | 63% |
| Polska 2050 | 1.8 | 3.1 | 0 | 0 | 0–12 | 17% |
At more than a year out, the drift is larger than any party's final-poll spread. The ranges of the large lists become several dozen seats wide, and the chances of the lists near their thresholds move towards the middle. The chances of 231 and 276:
| Combination | 231 today | 231, polling day | 276 today | 276, polling day |
|---|---|---|---|---|
| The governing coalition | 10% | 44% | 0% | 8% |
| The governing coalition + Razem | 13% | 51% | 0% | 11% |
| PiS + Konfederacja | 2% | 14% | 0% | 1% |
| PiS + Rozwój Plus + Konfederacja | 5% | 26% | 0% | 3% |
| PiS + Rozwój Plus + Konfederacja + KKP | 86% | 47% | 2% | 11% |
The date
The Sejm's four-year term runs from its first sitting on 13 November 2023. The election must be held on a non-working day in the last 30 days of the term, ordered by the President at least 90 days before the term ends (Constitution, art. 98(2)): so between 14 October and 12 November 2027, and ordered by 15 August 2027. The latest possible date is Thursday 11 November 2027, a public holiday, which the site uses; the latest Sunday is 7 November. The Sejm can shorten its term with 307 votes (art. 98(3)), which no combination is near. An earlier date narrows the drift only a little: at a year out it would be about 4.1 points, at six months about 3.6, at three months about 3.1.
Limitations
Four elections. The curve and the test rest on four elections, one of them with a thin series.
Wikipedia's tables. The series is taken from Wikipedia's poll tables and not checked against releases.
Split lines. The pull-back moves today's parties by the 2023 result of the list they descend from. For Trzecia Droga's parties, which have fallen a long way since 2023, that is a strong assumption.
The drift includes the final-poll error, which the simulation also adds, so the ranges err on the wide side, as elsewhere.
The line-up. The polling-day forecast holds the default line-up fixed. By polling day the lists will be known.
References
Erikson, R. S. and Wlezien, C. (2012). The Timeline of Presidential Elections: How Campaigns Do (and Do Not) Matter. Chicago: University of Chicago Press.
Konstytucja Rzeczypospolitej Polskiej, article 98.
Stone, M. (1974). Cross-validatory choice and assessment of statistical predictions. Journal of the Royal Statistical Society: Series B, 36(2), 111–147.
How to cite
Polling Forecast (2026). From today to polling day in Poland. Retrieved 21 November 2026 from https://www.pollingforecast.com/research/poland-election-day. Live figures change as new polls arrive; the date above is when the text was last revised.


