Why Exit Polls Could Be Wrong: A Tale of Two States – UP & Bihar

Despite exit polls predicting a saffron wave, like Morpheus says in The Matrix, “It ain’t over till it’s over.”

Is it another clean sweep for the BJP-led NDA or will 2019 throw up an upset for the saffron giant?

(Illustration: The Quint)

Till Sunday, 19 May, the question about who would form India’s next government was still up for debate. But the speculation was swiftly shut down by not one or two, but nearly every exit poll that came out. And they’ve all given the BJP-led NDA alliance a clean sweep victory in 2019. But will 2019 witness another Modi wave AKA ‘TsuNamo’ Part 2 AKA The Return of Modi Sarkar?

We look closely at current exit polls, past predictions and ground realities, which could VERY well tell a different story about the 2019 election results.

Now, we already covered Part 1 of this story on our previous podcast - Exit Polls: To Trust or Not to Trust. So today, we’re looking at two states that hold 120 out of the 543 seats that went to polls – Bihar and Uttar Pradesh.

Bihar: NDA Sweep or Mahagathbandhan’s Resurgence?

First, let’s look at Bihar, NDA-ally Nitish Kumar’s home ground. Exit polls have predicted a landslide win for the BJP-led alliance in Bihar.

The NDA is projected to win most of Bihar’s 40 seats, with India Today-Axis My India estimating a near-total shutdown of the Lalu Prasad-led coalition, which also includes the Congress.

But Bihar has had a history of exit polls going horribly wrong. Take the case of the 2015 Assembly elections. Of the seven exit polls, as many as four projected a win for the BJP in the state. One of them, in fact, gave as many as 155 out of the total of 243 seats to the BJP and its allies. But, like Morpheus says in The Matrix, “It ain’t over till it’s over.”

The actual verdict turned out to be the complete opposite of what was projected. The Lalu-led Mahagathbandhan bagged nearly 180 seats. Lalu’s Rashtriya Janata Dal alone won 80 seats and had a vote share of 44 percent in the seats it contested.

Back then, the Mahagathbandhan included the RJD, Nitish Kumar’s JD(U) and the Congress.

Nitish Kumar switched sides shortly after and is now part of the NDA. The Mahagathbandhan, however, has new allies like Upendra Kushwaha’s RLSP, Jitanram Manjhi’s HAM and Mukesh Sahani’s VIP.

These new allies are no slouches. They’re seen to represent sections of what are known as ‘extreme backward classes’ and ‘mahadalits’.

The seats where the Mahagathbandhan is seen to have advantage include Kishanganj, Katihar, Purnia, Araria, Bhagalpur, Banka, Buxar, Sitamarhi, Patliputra, Siwan and Nawada.

Incidentally, many of these seats bucked the trend even in the ‘Modi wave’ elections of 2014.

That aside, the Muslim-Yadav votebank, which forms more than 40 percent of the voting population, is almost unshakeable in its support of the Mahagathbandhan.

To top this off, the RJD-led Gathbandhan in Bihar has tried to expand its voter base by sharing seats with the Congress, the Hindustani Awam Morcha, the RLSP and a newly emerged party called the VIP 3. The party thinks that this seat-sharing will draw a section of the upper castes, Dalits and the lower OBCs in addition to the Muslims and the Yadavs.

So, despite the prediction of an NDA landslide victory, Bihar could very well be an upset.

Now let’s head on over to UP.

UP: Can BJP Really Topple the Mahagathbandhan?

Different exit polls have predicted three potential outcomes in India’s heartland: An NDA sweep AKA the return of the BJP. A Mahagathbandhan sweep, AKA the others sweep the polls, ie., neither BJP nor Congress. And finally, the third outcome, which is that the NDA and the Gathbandhan evenly split the seats.

Now first let’s examine the prediction that the NDA and the Gathbandhan will be evenly split. Three surveys predict this result. India-TV CNX, ABP-Nielsen and CVoter.

The CNX survey indicates a less than perfect vote transfer in the alliance and a slight positive swing in favour of the NDA. The Nielsen survey is the reverse, reflecting a perfect transfer of Gathbandhan votes and a slight swing away from the NDA. Curiously, Nielsen’s initial prediction on the evening of 19 May was 22 seats for the NDA but it later adjusted it to 33.

There are three surveys which give the NDA a sweeping victory in UP. The India Today-Axis poll predicts 62-68 seats for the NDA and 10-16 for the Mahagathbandhan.

The Today’s Chanakya poll predicts 65 for the NDA and 13 for the Mahagathbandhan. And the News18-Ipsos poll predicts 60 for the NDA and 18 for the Mahagathbandhan.

These exit polls would mean that Uttar Pradesh has seen a Modi wave that cuts across communities, age groups and other social categories.

Finally, the surveys that give the Gathbandhan a sweeping victory are News Channel National Voice, which says the Gathbandhan could get 48 seats and Anthro.ai which uses artificial intelligence to make predictions. Anthro has given the Gathbandhan 54 seats, and the NDA, just 22.

The Axis My India exit poll which predicted 62-68 seats for the NDA, was taken down from Axis’ website later, because of several discrepancies, including, predicting a congress win in a seat in Tamil Nadu, where the party hadn’t even put up a candidate from! It was later published on the India Today website, with some alter

Out of all these surveys, the only ones that give seat predictions along with seat-by-seat vote share predictions, were Axis, National Voice and Anthro.ai.

Once we remove Axis from the mix, which seems like the logical thing to do, we can examine the other two. National Voice and Anthro both predict a decisive lead for the Mahagathbandhan, with 48 and 54 seats respectively. Both, also give detailed seat-by-seat predictions in UP.

Clearly, the more detailed seat-by-seat predictions add considerable strength to these two polls, meaning that they’re presumably more reliable than the other polls, meaning that a Mahagathbandhan victory in UP might be, in all likelihood, what we’re looking at.