Gergely Gulyás, the outgoing minister heading the Prime Minister’s Office, said that internal polling—on which Fidesz–KDNP’s campaign measured support for Viktor Orbán’s governing alliance—was misleading and inaccurate. Speaking to opposition outlet 444, Gulyás admitted that polls commissioned by the campaign team, even right before the election, showed that they would win.
Fidesz–KDNP suffered a crushing defeat on 12 April, winning only 13 of the 106 single-member constituencies and securing 43 mandates on the national list. The opposition Tisza party won a constitutional supermajority with 137 seats in the 199-seat parliament, with a record of more than 3.1 million votes. Tisza’s advantage could increase by as many as three mandates when the final results are announced on Saturday, 18 April.
Throughout the campaign—especially in its final phase—there was a sharp contradiction between polling data published by pro-government pollsters and that of independent and opposition-aligned institutes. The first poll to show Tisza ahead of Fidesz–KDNP was published by the 21 Research Center in October 2024, and since then these pollsters consistently measured a Tisza lead.
‘Gulyás admitted that polls commissioned by the campaign team, even right before the election, showed that they would win’
Meanwhile, institutes such as Nézőpont Institute, the Center for Fundamental Rights, and Századvég presented entirely different conclusions, showing Orbán’s party ahead by a decisive margin, creating an alternative polling reality in which Fidesz–KDNP was portrayed as the leading force.
At the same time, media figures and politicians linked to Fidesz attacked pollsters suggesting that Tisza was leading. The most prominent example was Orbán posting a photo on social media calling Endre Hann, founder of Hungary’s first polling company Medián, the ‘best comedian’, reacting to a survey showing Tisza ahead of Fidesz by 20 percentage points in February 2026.
Az AtlasIntel volt a legpontosabb közvélemény-kutató intézet a magyar országgyűlési választáson.
— Andrei Roman (@andrei__roman) April 15, 2026
Április 5–10-i felmérése (Tisza 52,1%, Fidesz 39,3%, Mi Hazánk 5,1%) átlagos hibája mindössze 0,2 százalékpont volt – a legalacsonyabb a nagyobb intézetek közül. pic.twitter.com/87dwZgtMwf
Medián, Závecz Research, Publicus, the 21 Research Center, and other pollsters all projected the final result of the election with relative accuracy, with margins of error ranging from 0.4 to 5 percentage points. By contrast, government-aligned pollsters such as the 21st Century Institute, Nézőpont Institute, and the Center for Fundamental Rights missed the result by 17.6, 18.6, and 20.6 percentage points respectively.
Similar dynamics appeared in international polling. Atlas Intel, considered one of the most accurate pollsters in the United States, projected a Tisza victory with 52.1 per cent on the national list against Fidesz–KDNP’s 39.3 per cent, missing the margin by only 0.2 percentage points. Meanwhile, McLaughlin & Associates, a long-time pollster for US President Donald Trump, projected a Fidesz victory at 45.4 to 39.7, missing the result by 18.3 points. McLaughlin worked in cooperation with the Nézőpont Institute during the campaign.
Speaking in a podcast in January, Gulyás himself said it was not worth ‘attaching excessive importance’ to polling. At the same time, he added that those pollsters that ‘correctly predicted the last election results’ were now also measuring a ‘fairly confident lead’ for Fidesz. ‘Fidesz has a very good chance of securing a comfortable majority,’ he said, almost exactly three months before Tisza’s landslide victory. Gulyás also stated that he had not seen any poll in which ‘Fidesz was not in the lead’.
Related articles:
At a time when public debate is increasingly polarized and superficial, Hungarian Conservative remains committed to depth, intellectual honesty, and independent conservative thought.
Producing high-quality journalism requires resources. Your contribution helps us expand our coverage, reach new audiences, and keep our content accessible.
Please consider supporting our mission.
