How Accurate Have Exit Poll Predictions Been in the past Lok Sabha Elections

Check what the prominent exit polls had predicted in the past Lok Sabha Polls - 2019, 2014, 2009, and 2004

As the final round of polling draws to a close, different pollsters and news channels are releasing their exit poll predictions for the Lok Sabha elections 2024.

While the final results of India's 18th General Election will only be declared on 4 June, these exit polls are viewed as indicators of voting trends. The exit polls can either correspond or contradict the final results.

So before looking at the exit poll figures of these elections, let's see how accurate have the exit polls been in past Lok Sabha elections.

Exit Polls 2019 Lok Sabha Elections: Which Pollster Got it Right, Who Missed the Mark?

India Today - Axis My India was right in forecasting the BJP-led NDA bloc emerging as the winner by placing it within a range of 339-365 and the Congress-led UPA coming in second place within a range of 77-108. However, they underestimated the seats won by Others, which in actuality stood at 100.  

News 18’s IPSOS got it wrong as they under-estimated BJP’s wins with an estimate of 276, that fell short of the party’s actual seat tally which stood at 303. Likewise, they predicted the Congress’ seatshare at a lower bid than their actual win on 52 seats. 

Times Now’s VMR was right in predicting the BJP emerging as the single largest party but underestimated their haul by placing it at 262 seats, when they actually won 303 seats. The exit poll also accorded Congress a victory on 78 seats, much higher than their actual result of winning 52 seats.  

Republic TV’s CVoter and ABP Nielson were accurate in predicting the BJP-led NDA emerging as the winning side, but underestimated their haul. In contrast, both outlets overestimated the seats won by the Congress-led UPA by a great measure. Similarly, they overestimated the number of seats won by Others which stood at 100 in reality.

News24’s Chanakya was spot on with their forecast, placing the NDA bloc at 350 seats, just two short of their actual victory. It was also right in predicting the UPA bloc at 95 seats, four higher than its actual tally of winning 91 seats. 

India TV’s CNX and NewsX’s Neta were both off the mark, grossly underestimating the seats won by the NDA bloc and awarding a higher range of seats won to the UPA bloc. Both organisations also overestimated the number of seats won by Others. 

So out of the eight exit polls we looked at, all were correct in their prediction of the winning side. But only two pollsters – News24’s Chanakya and India Today - Axis My India – predicted the range of the seats for BJP and Congress correctly.

Exit Polls 2014 Lok Sabha Elections: Which Pollster Got it Right, Who Missed the Mark?

CNN-IBN’s CSDS was right in forecasting BJP emerging as the largest party and Congress coming in second place. However, they underestimated BJP's haul by predicting 236 seats and the seats won by Others, while overestimating the seats won by Congress at 77.

India Today - Cicero and Times Now-ORG correctly predicted the winning alliance, but under-estimated the seats won by the NDA bloc by a big margin, instead assigning a higher-than-actual tally to the UPA bloc. However, both organisations closely estimated the number of seats won by Others, which stood at 148. 

News24’s Chanakya was the most accurate in predicting the number of seats won by BJP, which was only nine higher than their actual win of 282 seats. However, they overestimated the seats won by the grand old party Congress, placing it at a higher tally than its actual score of victories on 44 seats. 

ABP-Nielson and India TV-Cvoter both predicted the winning alliance correctly, but fell-off the mark in their predictions of their seat-count. They under-approximated the seats won by the NDA, and overestimated the seats won by the UPA bloc.

So all the six exit polls in 2014 we looked at got winning alliance right. However, only News25-Chanakya comes close to predicting the tallies of BJP and Congress.

Exit Polls 2009 Lok Sabha Elections: Which Pollster Got it Right, Who Missed the Mark?

CNN-IBN-Dainik Bhaskar was right in forecasting the ruling UPA emerging as the winner and the NDA bloc coming in second place. However, they underestimated UPA’s haul by predicting wins on a range of 185-205 seats, while overestimating the seats won by NDA within a range of 165-185.

Star-Nielsen and India TV-CVoter rightly predicted the winning alliance, but under-estimated the seats won by the UPA bloc by a great margin, instead assigning a higher-than-actual tally to the NDA bloc. Similarly, both pollsters were also off-the-mark in estimating the number of seats won by Others, which stood at 122. 

Overall, all the exit polls forecasted a victory for the Congress-led UPA bloc in the 2009 Lok Sabha Elections. However, their tallies were misplaced corresponding the actual result.

Exit Polls 2004 Lok Sabha Elections: Which Pollster Got it Right, Who Missed the Mark?

NDTV-Nielson was off the mark in forecasting the NDA bloc emerging as the winning side and the Congress coming in second place. Similarly, they underestimated the seats won by Others, which stood at 217.

AajTak-Marg and Star News’ CVoter also got it wrong as they exaggerated the seat tallies of the BJP-led NDA bloc and overestimated the seats won by the Congress.

So all the three exit polls for 2004 we looked at got it wrong, severely underestimating the Congress, which formed the UPA government as a post-poll alliance.

Finally, while exit polls have been in sync with the outcomes of General Elections in three out of the past four polls – in 2009, 2014, and 2019 – their accuracy can be called into question with regards to the tallies of the seats won.