Nearly a quarter of voters planned to reject both candidates entirely
Three weeks before Peru's presidential runoff, a nation historically divided between its capital and its interior finds that division rendered in polling numbers: Pedro Castillo, the rural schoolteacher from Cajamarca, holds a meaningful lead over Keiko Fujimori, whose strength is concentrated in Lima while the rest of the country tilts away from her. The contest is not yet decided — nearly a third of voters remain uncommitted or plan to reject both candidates — but the geography of support speaks to something older and deeper than any single election cycle.
- Castillo's lead has quietly widened to 36.5% against Fujimori's stalling 29.6%, a gap that feels modest on paper but carries real structural weight.
- Nearly one in four Peruvians plans to cast a blank or null ballot — a protest vote that has grown by 2.3 points in a single week and could scramble any projection.
- The map tells a stark story: Castillo commands the south, the center, and the east by wide margins, while Fujimori's only decisive advantage is Lima Metropolitan.
- Fujimori's campaign faces a structural trap — winning the capital loudly while losing the country quietly, with no clear path to closing the gap in the interior.
- With roughly one-third of the electorate still unanchored, the final three weeks remain genuinely open, and both campaigns know the undecided mass could rewrite the result.
Three weeks out from Peru's runoff, Pedro Castillo held a real but not commanding lead over Keiko Fujimori. A mid-May poll by the Instituto de Estudios Peruanos placed the Perú Libre candidate at 36.5 percent — up a modest 0.3 points — while Fujimori had slipped to 29.6 percent, losing 0.4 points from the prior week. The numbers moved slowly, but they moved in one direction.
What complicated any confident forecast was the sheer size of the electorate still in motion. Nearly 23.6 percent of voters said they intended to cast blank or null ballots, a figure that had grown by 2.3 points in a single week. Another 7.8 percent remained genuinely undecided. Together, these voters amounted to roughly one-third of the electorate — enough to overturn any apparent advantage.
The geographic breakdown was where the race became most legible. Castillo, a teacher from Cajamarca, dominated Peru's central, southern, and eastern regions by wide margins. Fujimori's only clear advantage was Lima Metropolitan, where she led 38.4 percent to his 25.4 percent. Strip Lima from the count entirely, and Castillo's lead expanded to 42.6 percent against Fujimori's 24.9 percent.
The structural problem for Fujimori was plain: she was winning the capital decisively and losing the country decisively. Unless she could extend her appeal into the interior or find some other way to shift the terrain, the poll's margin of error of 2.8 points offered her little comfort. The momentum, such as it was, belonged to Castillo heading into the final stretch.
Three weeks before Peru's runoff election, the race remained stubbornly close but tilted toward Pedro Castillo. A poll by the Instituto de Estudios Peruanos, conducted May 13-15, showed the Perú Libre candidate holding 36.5 percent support—a modest gain of 0.3 percentage points from the previous week. Keiko Fujimori, the Fuerza Popular standard-bearer, had stalled at 29.6 percent, actually losing 0.4 points. The gap between them was real but not commanding, and the volatility in the numbers suggested the race could still shift.
What made the polling landscape genuinely uncertain was the size of the undecided bloc. Nearly a quarter of voters—23.6 percent—said they planned to cast blank or null ballots, a figure that had grown 2.3 points from the previous survey. Another 7.8 percent remained genuinely undecided. Together, these groups represented roughly one-third of the electorate, a mass of voters large enough to reshape the outcome entirely.
The geography of support told a clearer story. Castillo, a teacher from Cajamarca, dominated everywhere outside the capital. In Peru's central regions, he claimed 39.9 percent to Fujimori's 20 percent. In the south, he reached 49.6 percent against her 19.9 percent. In the eastern regions, 45.6 percent backed him compared to her 30.1 percent. Only in Lima Metropolitan did the picture reverse: Fujimori led there with 38.4 percent while Castillo managed 25.4 percent. When the pollsters excluded Lima's votes entirely, Castillo's national advantage became stark—42.6 percent to Fujimori's 24.9 percent.
Fujimori's campaign faced a structural problem. She was winning decisively in the capital, Peru's largest media market and political center, but losing decisively everywhere else. Castillo's base was distributed across the country's interior and southern regions, areas with deep reservoirs of rural and working-class voters. The poll, conducted by telephone with a margin of error of 2.8 points, suggested that unless Fujimori could either expand her appeal beyond Lima or suppress turnout in Castillo's strongholds, the momentum favored the Perú Libre candidate heading into the final stretch.
Notable Quotes
The candidate of Peru Libre maintained his advantage while the leader of Fuerza Popular failed to advance in voter support— Instituto de Estudios Peruanos poll summary
The Hearth Conversation Another angle on the story
Why does a 6.9-point lead feel fragile when Castillo is clearly ahead?
Because nearly a quarter of voters are still saying no to both candidates. That's not noise—that's a real rejection of the choice itself. If even half of those blank votes break toward Fujimori, the race tightens dramatically.
And the regional split—is that as decisive as it looks?
It's decisive in a mathematical sense, but it also reveals something about Peru itself. Castillo owns the countryside and the south. Fujimori owns the capital. The question is whether Lima's media dominance can translate into national victory when she's losing 2-to-1 everywhere else.
What about the 0.3 percent gain for Castillo—is that momentum or statistical noise?
At this point, with the margin of error at 2.8 points, it's noise. What matters is that Fujimori isn't gaining. She's been stuck for weeks. That's the real story.
So Castillo just has to hold?
Mostly, yes. But he also has to hope the blank vote doesn't explode further. If it keeps growing, it could indicate a broader rejection that hurts him as much as her.