Haryana elections: Pollsters lose again — why exit polls go so wrong

It was supposed to be a landslide for Congress. A TV commentator even remarked that it remained to be seen whether it would be a Congress wave or a tsunami in Haryana—but in the end, it wasn’t even a ripple. Once again, Congress managed to snatch defeat from the jaws of victory.

But it wasn't solely the defeat of the Congress party; the Haryana election results proved all exit polls wrong, without exception.

Also Read: Haryana Elections Highlights

Almost all exit polls had forecasted a clear majority for the Congress in Haryana. 'Dainik Bhaskar' predicted Congress would win 44-54 seats; it gave BJP 15-29 seats in the 90-member Haryana Assembly. C-Voter-India Today polls placed Congress ahead with 50-58 seats, predicting BJP to get around 20-28 seats, while Republic Bharat-Matrize polls gave Congress 55-62 seats against 18-24 for BJP. The Red Mike-Datansh exit poll projected Congress winning 50-55 seats and BJP 20-25, while Dhruv Research predicted 50-64 seats for Congress and 22-32 for BJP. Peoples' Pulse claimed that Congress to form govt with 49-60 seats.

But the predictions proved utterly wrong.

It wasn't for the first time though (and probably won't be the last).

In the run-up to the 2024 Lok Sabha elections, at least 12 exit polls predicted a landslide victory for the BJP-led National Democratic Alliance (NDA) with some even predicted '400-paar'. However, the actual results were strikingly different. The NDA finished with 293 seats, far below the predictions. The BJP, in particular, failed to secure a simple majority on its own, winning only 240 seats, a loss of 63 from their previous tally of 303 in 2019. Meanwhile, the Congress-led INDIA bloc claimed ..

In the 2023 Chhattisgarh Assembly Elections also the BJP regained control in Chhattisgarh in the 2023 Assembly elections, despite exit polls predicting an easy win for the Congress. The BJP secured over 50 seats, defying the projections. In Madhya Pradesh, only a few exit polls correctly predicted a decisive win for the BJP, which ousted the Congress after its five-year rule.

Same happened in 2017 Uttar Pradesh Assembly Elections. While pollsters then predicted a hung assembly, BJP emerged as the largest party winning an overwhelming majority of 325 seats.

Results of 2015 Bihar and Delhi Assembly Elections as well as Lok Sabha polls of 2004 and 2014 had been totally opposite to what pollsters have predicted then as well.

Why exit polls go wrong?
There could be several reasons why exit poll results may be inaccurate. These could include issues with accuracy, the methods used, the tightness of the contest, or even insufficient data.

Accuracy: Exit polls face several challenges that can impact their accuracy. One fundamental premise is the assumption that voters truthfully disclose their preferences during exit poll interviews. However, factors such as deliberate misinformation or fear, particularly among marginalised sections, can skew the results.

Tight Contest: Several agencies have predicted a tight contest in many states where the vote share difference is nearly 1-2%. The margin of error in exit polls, which typically ranges from 1-3 percentage points, also makes these exit poll predictions hard to believe.

Poll Methods: The pressure to swiftly deliver findings may lead to errors in methodology, with some agencies resorting to computer-assisted telephone interviews that might not accurately reflect the ground reality. Human errors, despite technological advancements, persist in sampling, with field resources often favouring convenient booths, potentially distorting results.

Insufficient data: The lack of socio-economic, caste, and gender data further challenges the accuracy of predictions. In a diverse country like India, the unavailability or overlooking of caste and socio-economic considerations can affect the accuracy of predictions. Further, the underrepresentation of women introduces potential errors in normalization processes.