The Le Pen 2027 polls all tell the same broad story: Marine Le Pen is currently projected to win a runoff against the opponents being tested. But the useful question is not simply who is ahead. It is how large the gap is that needs to be closed. And on that point, the percentages already allow a very concrete calculation, even though no polling institute provides an absolute number of votes for 2027 here.

To read the Le Pen 2027 polls correctly, the right approach is therefore to reason on a base of 100 valid votes cast. In a Le Pen–Mélenchon matchup, the three July 2026 measurements give 70-30 at Ifop-Fiducial, 67.5-32.5 at Elabe and 67-33 at Toluna Harris. Those are leads of 40, 35 and 34 points. But if one vote moves directly from one candidate to the other, the gap shrinks by two votes: one fewer for Le Pen and one more for her opponent.

The real calculation: 17 to 20 points of votes need to switch sides

Out of 100 valid votes cast, a 70-30 score does not mean 40 votes have to be “won back” to reach a tie. What is needed is to shift 20 points of votes from the first camp to the second: 70 becomes 50 and 30 becomes 50. To actually move ahead of Le Pen, slightly more than 20 points would therefore have to switch on this theoretical basis.

  • Ifop-Fiducial, July 7-8, 2026: 70-30, meaning 20 points of votes must be transferred to reach 50-50.
  • Elabe, July 9-10, 2026: 67.5-32.5, meaning 17.5 points must be transferred.
  • Toluna Harris, July 7-8, 2026: 67-33, meaning 17 points must be transferred.
The right way to read it: a 70-30 result is a 40-point gap, but 20 points of votes switching directly from one side to the other are enough to produce a 50-50 tie.

By contrast, if we imagine that Le Pen loses no votes and her opponent gains only through additional former abstainers, the requirement is higher. On the same base of 100 initial votes, the candidate on 30 would need the equivalent of 40 additional votes to catch Le Pen's 70. In the 67.5-32.5 and 67-33 scenarios, the equivalent would be 35 and 34 extra votes respectively. This calculation does not predict turnout in 2027: it simply shows why mobilizing abstainers and switching sides do not have the same mathematical effect.

Which voter blocs could actually move the result?

There is no detailed projection of vote transfers for 2027 in the available figures. The 2022 runoff can nevertheless be used as a comparison point, provided it is not turned into a forecast.

Mélenchon voters: many votes stayed outside the final matchup

According to Elabe's April 24, 2022 study, out of 100 Mélenchon voters in the first round, 38 voted for Macron, 18 for Le Pen and 44 abstained in the runoff. The striking point is therefore not just the 38-18 split. It is above all the size of the group that chose neither finalist.

Looking bloc by bloc, Mélenchon voters in 2022 produced a net advantage of 20 votes for Macron per 100 first-round voters: 38 versus 18. But 44 remained outside the duel. That is a much larger reservoir of possible change than the direct transfer between the two finalists alone. Le Monde's polling-station analysis also confirms that a non-negligible share of areas that were very favorable to Mélenchon in the first round also fed votes to Le Pen in the runoff.

The traditional right: a split bloc, not an automatic transfer

Valérie Pécresse's voters were also highly dispersed: 51% Macron, 22% Le Pen and 27% abstention. Out of 100 voters in this bloc, Macron's net advantage was therefore 29 votes. Here again, more than one voter in four backed neither finalist.

This 2022 snapshot rules out an overly simple conclusion about “the right.” It mainly shows three possible destinations: the candidate opposing the RN, Le Pen, or abstention. For 2027, there is no precise figure showing how this split would evolve. But that is exactly why this is one of the blocs where a few points of movement could matter, because it was already far from homogeneous in 2022.

The Zemmour bloc: much less apparent room for movement

Éric Zemmour's voters were far more polarized: 80% voted for Le Pen, 11% for Macron and 9% abstained. Out of 100 voters in this bloc, Le Pen's net advantage reached 69 votes. In scenario terms, this bloc therefore mechanically offers less reserve through abstention than the Mélenchon or Pécresse electorates of 2022.

  • Mélenchon 2022: 44 voters out of 100 abstained in the runoff.
  • Pécresse 2022: 27 out of 100 abstained.
  • Jadot 2022: 24 out of 100 abstained, compared with 68 for Macron and 8 for Le Pen.
  • Zemmour 2022: only 9 out of 100 abstained, with 80 voting for Le Pen.

This table does not say who will win in 2027. It only shows where the largest areas of voter mobility were three years earlier: more among left-wing and traditional-right electorates than in the Zemmour bloc.

Why polls one year out show a direction, not a final score

The 2022 case is a useful warning. In April 2021, Mélenchon was measured at 11% by Ifop-Fiducial and peaked at 13.5% in a Le Monde compilation. One year later, he won 21.95%, less than 500,000 votes short of qualifying for the runoff. The gap between expectations and the actual result therefore ranged from 8.45 to 10.95 points depending on the comparison point used.

Conversely, Valérie Pécresse was polling between 9% and 13.5% one year before the election, then finished on 4.8%. The overestimation therefore ranged from 4.2 to 8.7 points. Those two examples are enough to show that an individual polling score can still move substantially a year before the vote.

That does not make long-range polls entirely useless. Since 2007, 2017 is the only election in which the eventual winner had not been identified as the favorite one year beforehand. In 2007, 2012 and 2022, the two candidates expected to reach the runoff at that stage did in fact make it through. Another compilation nevertheless indicates that the favorite identified more than a year before the election has ultimately won in only 50% of cases since 1995.

What the 2027 figures really say

Le Pen currently has a very high base in first-round voting intentions: 35% to 38% in the Harris Interactive poll of August 18 and 19, 2026, while the RN level remains around 34% to 36% regardless of which candidate is tested. And in the most documented runoff scenario, against Mélenchon, her lead ranges from 34 to 40 points depending on the July polling institute.

But converting that lead into an “exact number of votes to flip” would be misleading without knowing the number of valid votes cast and the abstention rate in the 2027 runoff. The cleanest measure is therefore relative: 17 to 20 points of votes would have to switch sides directly to bring these scenarios back to an even split. Growth based only on new voters would require an even larger volume.

The comparison with 2022 adds a second lesson: the outcome does not depend only on the RN's core support. It also depends on the proportion of left-wing and traditional-right voters who transfer, abstain or switch sides. That is where today's polls are still incomplete. They describe the current lead very well; they do not yet allow that lead to be converted into a definitive verdict.