From today to polling day in Italy
Abstract. We test the British pull-back towards the previous result on Italian polls at six distances up to a year before the Camera elections of 2013, 2018 and 2022. The best-fitting curve was no pull-back at all: the previous result missed each list by about eight points, twice as much as a year-old poll. The polling-day forecast keeps today's corrected shares and widens the ranges by a drift fitted to how far polls have missed at each distance.
Summary
Everything else in the Italy section answers the question "what if the election were held today?". The next Camera election is due by 22 December 2027 at the latest, 396 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 Italian polls, at six distances from one month to one year before the Camera elections of 2013, 2018 and 2022. The best-fitting curve was no pull-back at all, and the test confirmed it: in Italy the previous result has been a very poor guide, missing each list by about eight points, because the vote has moved so far between elections. So our polling-day forecast keeps today's shares, corrected as in Reading Italian polls: the correction and its test, and adds only uncertainty, which grows with the time left: polls this far out have missed each list by about 4.8 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 Italy the tests were already on elections since 2001, so nothing changed.
The evidence: polls at fixed distances before three elections
The evidence is every national voting-intention poll in our series for the twelve months before each of the Camera elections of 2013, 2018 and 2022, taken from Wikipedia's tables with some checked against the pollsters' or outlets' pages (The Italian data: results, maps, statistics, deputies and polls). Earlier elections are not in the series.
For each election we average the polls at 30, 61, 91, 182, 273 and 365 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 list, the poll average, the list's result at the previous election and its result at this one, with the same lineages as the rest of the Italian record. A list with no previous result, such as the M5S in 2013, is left out. That gives 84 horizon points.
The window closest to polling day, centred 30 days out, reaches into the fifteen days in which polls may not be published, so in practice its polls are those published before the ban.
What the data show
| Days before polling day | Elections | Polls' miss | Previous result's miss | Pull-back's miss | RMSE |
|---|---|---|---|---|---|
| 30 | 3 | 2.71 | 7.67 | 2.74 | 3.01 |
| 61 | 3 | 3.35 | 7.67 | 3.37 | 3.73 |
| 91 | 3 | 3.73 | 8.66 | 3.75 | 4.16 |
| 182 | 3 | 3.31 | 8.66 | 3.32 | 4.27 |
| 273 | 3 | 3.35 | 8.08 | 3.37 | 4.64 |
| 365 | 3 | 3.64 | 8.08 | 3.68 | 4.66 |
The previous result is a very poor guide. It missed each list by between 7.7 and 8.7 points, about twice as much as a year-old poll. Italian elections have moved further than any others on the site: the Lega went from 4.1% in 2013 to 17.4% in 2018 and back to 8.8% in 2022; the M5S from 25.6% to 32.7% and then 15.4%; FdI from 4.3% in 2018 to 26.0% in 2022. Pulling a list back towards a result like that would move it the wrong way.
The polls are only a moderately good guide. A month out they missed by 2.7 points per list, and a year out by 3.6, with little change in between. Italian polls are less accurate than German or Spanish ones at every distance, as they are at the final poll (Measuring Italian polling error, 2008–2022).
Election by election:
| Days out | 2013 | 2018 | 2022 |
|---|---|---|---|
| 30 | 2.85 | 3.42 | 1.87 |
| 61 | 3.75 | 3.61 | 2.69 |
| 91 | 4.17 | 3.65 | 3.36 |
| 182 | 2.22 | 4.14 | 3.56 |
| 273 | 2.32 | 4.43 | 3.31 |
| 365 | 3.52 | 4.10 | 3.30 |
Table: Polls' average miss per list at each distance and election, points.
2013 has an odd shape: the polls six to nine months out were closer than those one to three months out. The M5S, the list the polls missed most that year, is not in these figures, because it had no previous result; the pattern belongs to the other lists.
The pull-back curve and its test
We tested the British model. 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 \(f = 1\): all of the change since the last election counts, at every distance. That is no pull-back.
The test, each election predicted from a curve fitted on the other two, agrees: the pulled-back shares were very slightly further from the result than the polls at every distance. The site keeps today's shares for polling day, and shows the fitted curve on the election-day page, labelled as not used.
Drift
The polls' miss grows with the time left, and the polling-day forecast keeps that uncertainty. The root-mean-square miss of the polls at each distance is fitted by
over the six distances. At 396 days it is 4.8 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. Beyond a year, the longest distance tested, the curve is an extrapolation.
In the simulation the drift is scaled by the square root of each list's share relative to 15%, so that large lists get more and small ones less, and combined with the list's final-poll spread:
Polling-day ranges
The polling-day forecast runs the simulation (Simulating Italian polling error: where the chances come from) with 1,000 runs and equation (3), under the default line-up. Its centre is today's corrected shares; the difference from the "today" forecast is the width.
| Coalition | Seats today | 80% range today | 80% range, polling day | Majority today | Majority, polling day |
|---|---|---|---|---|---|
| Centre-left | 224 | 138–224 | 133–225 | 63% | 58% |
| Centre-right | 143 | 140–224 | 130–224 | 37% | 40% |
| FN | 23 | 18–30 | 0–41 | – | – |
| Azione | 10 | 0–13 | 0–23 | – | – |
| No majority | 0% | 1% |
At more than a year out, the drift is larger than any list's final-poll spread except the M5S's. The coalition ranges become several dozen seats wider, the chances of the lists near 3% move towards a toss of a coin, and the chance that no coalition reaches 201 changes, since a wider spread gives both more lopsided and more even outcomes.
The date
The Chambers first met on 13 October 2022 and last five years; the election must be held within seventy days of their end (Constitution, arts. 60 and 61). The site uses the last date that allows, 22 December 2027. That is the cautious choice, giving the widest ranges. The President can dissolve the Chambers earlier (art. 88), as happened in 2022, when the election came about six months before the end of the term. An earlier election narrows the drift: at six months it would be about 4.3 points, at three months about 3.9.
Limitations
Three elections. The curve and the test rest on three elections and 84 horizon points. That is thinner than in any other country on the site.
A year at most. The tested distances end at a year; the next election may be further off, and beyond a year the drift is extrapolated.
The drift includes the final-poll error, which the simulation also adds, so the ranges err on the wide side, as elsewhere.
Independent drifts. As with the final-poll error, drift is drawn independently for each list, which in Italy matters most for the PD and the M5S (Simulating Italian polling error: where the chances come from).
The line-up. The polling-day forecast holds the default line-up fixed. By polling day the line-up will be known; today it is the largest uncertainty of all (Who stands with whom: the coalition line-up assumption).
New lists. Italian politics has produced a new national force at most recent elections, from the M5S in 2013 to Futuro Nazionale since 2022. A list that forms after today's polls cannot be anticipated.
References
Costituzione della Repubblica italiana, articles 60, 61 and 88.
Erikson, R. S. and Wlezien, C. (2012). The Timeline of Presidential Elections: How Campaigns Do (and Do Not) Matter. Chicago: University of Chicago Press.
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 Italy. Retrieved 21 November 2026 from https://www.pollingforecast.com/research/italy-election-day. Live figures change as new polls arrive; the date above is when the text was last revised.



