Questioning reliability of Exit Polls: A closer look at their history of inaccuracy in elections

New Delhi: As the Haryana and J&K Assembly elections draw to a close, survey agencies are back in action with their Exit Polls. However, one must question why these pollsters should be trusted, especially given the numerous instances of significant inaccuracies in their projections in the past.

While Exit Polls claim to offer insights into voter sentiment, a closer look at past elections - Lok Sabha and Assembly - reveal a troubling trend: pollsters frequently misjudge actual outcomes. The consistent failures of Exit Polls to align with reality raise serious doubts about their reliability.

Based on some data related to past predictions, IANS has made an assessment of the history of overestimations by the Exit Polls.

2014 Lok Sabha elections

In the 2014 parliamentary elections, pollsters predicted a substantial win for the BJP-led National Democratic Alliance (NDA), with an average prediction of 283 seats. But the actual result was a massive 336 seats for the NDA. The BJP alone secured 282 seats. This suggests that the polls could not gauge the scale of the win. This underlined the disconnect between the data and voter mood.

2019 Lok Sabha elections

The prediction in 2019 polls was that the NDA would secure around 306 seats. On the contrary, the NDA bagged 353 seats. The BJP alone got 303. Once again, the Exit Polls failed to judge the mood of the electorate. Voters favoured the NDA by a larger margin than what was predicted.

2024 Lok Sabha elections

Exit Polls projected a massive victory for the NDA in 2024 parliamentary polls, with estimates ranging from 361 to 401 seats. However, the actual results were far more modest, with the NDA ending up with just 293 seats and the BJP alone winning 240. This was a loss of 63 seats from their previous tally. This marked a significant blunder in Exit Poll predictions.

The 2004 Lok Sabha elections were interesting in this context. The pollsters overestimated the BJP-led NDA’s chances. They had predicted a comfortable majority with seats ranging from 240 to 275 seats. The outcome, however, was a shock, with the NDA securing only 187 seats, whereas Congress and its allies emerged victorious with 216 seats.

Discrepancies in state polls

These inaccuracies extend beyond national elections to state assemblies as well. For example, in the 2019 Haryana Assembly elections, Exit Polls suggested that the BJP would secure around 61 seats, with some estimates soaring as high as 75-80 seats. The reality was completely different: the BJP won only 40 seats, while Congress managed to capture 31, contradicting the Exit Polls’ indications about a clear BJP majority. In this Haryana polls, C Voter had predicted 72 seats for BJP which turned out to be only 40.

Similarly, Exit Polls predicted a competitive fight in Chhattisgarh Assembly polls in 2023 but were again off the mark. The BJP won 54 seats compared to expectations of 34-42.

The 2023 Madhya Pradesh Assembly elections saw pollsters predicting a BJP majority, with estimates ranging from 140 to 162 seats. Ultimately, the BJP won 163 seats, which, while close, still reflected the unpredictability of voter behaviour.

In the 2021 Bengal Assembly elections, predictions of an overwhelming BJP victory (with estimates ranging from 138 to 185 seats) were smashed as the Trinamool Congress (TMC) emerged winner with 215 seats, leaving the BJP with just 77.