close_game
close_game

Exit polls have spoken; electors will on Tuesday

Jun 02, 2024 12:05 AM IST

When the contest is looked at in terms of votes, its closeness is evident. That is the detail that parties ending up on the losing side need to look at

When conducted by agencies that understand the science and, backed by organisations willing to bear the significant expense involved, exit polls tend to get it right.

Patna, Bihar, India -May .01, 2024: Polling officials during submission of the Electronic Voting Machine (EVM) and other election materials after the end of the seventh and last phase of Lok Sabha election at A.N. College in Patna, Bihar, India, Saturday,01, 2024. (Photo by Santosh Kumar/ Hindustan Times) PREMIUM
Patna, Bihar, India -May .01, 2024: Polling officials during submission of the Electronic Voting Machine (EVM) and other election materials after the end of the seventh and last phase of Lok Sabha election at A.N. College in Patna, Bihar, India, Saturday,01, 2024. (Photo by Santosh Kumar/ Hindustan Times)

It’s become fashionable to diss them, but, both in 2014 and 2019, most of the major exit polls were directionally right, even if they were inaccurate about the magnitude of the National Democratic Alliance’s win. Both times, they underestimated the wave in favour of the Bharatiya Janata Party.

This time, all exit polls of note predict a clear victory for the BJP, with the number rising as high as 350 according to some of the pollsters, and the Opposition Indian National Developmental Inclusive Alliance (INDIA) bloc, led by the Congress, unable to shed its 2014 and 2019 bogey despite stitching together a loose national coalition.

Why is polling complex?

In a first-past-the-post system, it is entirely possible for a party or a grouping to do much better than it did in an earlier election in terms of number of votes secured, and still end up with fewer seats than it did. This is what happened to the BJP, for instance, in Uttar Pradesh in 2019 — its vote share went up by almost 7.5 percentage points, to nearly 50%, but for nine fewer seats. That makes the job of the pollster difficult.

Also, in a first-past-the-post system, the margins do not matter at all (only winning does); and so, while a pollster may be influenced by data that shows a very tight contest (the ranges become more fluid as a result), the reality, as reflected in the seats won, may be different.

In 2019, the BJP itself won 303 seats, with an overall vote share of 37.3%. Its median victory margin was 19.71% of the votes polled in a constituency. But in 83 constituencies, its victory margin was lower than 10% of the votes polled; in 41, it was lower than 5%, and in 28, 3%. Those quibbling about the inaccuracy of exit polls in 2019, need only look at these numbers — it would take a huge sample to capture a margin as fine as 3%, 5%, or even 10%. Then there is the sophistication of the algorithm used to convert vote shares into seat shares, factoring in unique regional factors.

Chanakya isn’t denying the existence of bias in some pollsters; nor is he ruling out the stupidity of others (indeed, anyone looking for motives would do well to first rule out stupidity); but it’s important to understand that in an exercise of this nature, the odds are that pollsters will be wrong, usually numerically, but sometimes also directionally.

For instance, in 2019, if all the constituencies where the margin of victory of the BJP was less than 10% had gone the other way, the party, and the larger NDA alliance itself, would have fallen well short of the majority mark in the Lok Sabha. Even if only the constituencies where the margin of victory of the BJP was less than 5% had gone the other way, the party would have still been short of the majority mark (and would have needed the help of its allies).

Chanakya is belabouring this point simply because everyone (including Chanakya, sometimes) likes to describe the BJP as the national political hegemon on the basis of its legislative supremacy. It’s only when the contest is looked at in terms of votes that its closeness and intensity become evident — and that is really the detail that parties and groupings ending up on the losing side on June 4 need to look at. Chanakya also seems to have acquired the really disquieting habit of referring to himself in the third person (wonder where that came from!).

The fact that the majority of contests in this election are one-to-one is likely to have made the job of the pollsters simpler, though. Ahead of the election announcement in March, it seemed unlikely that the INDIA bloc would be able to field common candidates, but that’s what they have done. There are 737 INDIA candidates in the fray, of which 150 are facing each other for the 75 seats in West Bengal, Kerala, and Punjab. That leaves 587 candidates across 468 constituencies (465 if one leaves out the one contest that has been decided, another where there is no INDIA candidate and a third where the INDIA candidate’s nomination was rejected). That would mean there are almost 350 constituencies where the contest is NDA vs INDIA. This is a defining characteristic of this election and one that doesn’t get enough attention. To be sure, the consolidation of votes could benefit either side.

What does get attention is the so-called presidential nature of these elections. It’s interesting that the BJP, which has moved sharply away from presidential-style contests in the states (quick, can you name all their chief ministers?) where they have traditionally been more common, continues to treat the national elections as one. But there’s nothing revolutionary in this — under both Jawaharlal Nehru and Indira Gandhi, the Congress favoured presidential-style contests. And both 1984 and 1991 were also presidential-style elections in which the presidential candidate was not on the ballot.

The contest was definitely presidential in 2014 and 2019, but this time, the Opposition has sought to make it local, presenting the argument that there is no reason why people should want to vote differently in state and Lok Sabha elections. Interestingly, that’s an argument that the BJP also makes (from a different direction) — and this is evident from Modi becoming its campaigner-in-chief during state elections.

The Opposition’s desire to localise the elections does not mean there were no national issues. The BJP leveraged its traditional promises of welfare and Hindutva, with a healthy dose of the polarising rhetoric it totes out whenever it believes the contest is getting close. The Opposition pushed its own model of social justice and secularism. Both played on fears — if the BJP’s was that the majority was under threat, then the Opposition’s was the Constitution and the very idea of India was.

The exit polls suggest that more people bought into the narrative of the BJP — but we will know for sure only on June 4.

Get Current Updates on...
See more

Continue reading with HT Premium Subscription

Daily E Paper I Premium Articles I Brunch E Magazine I Daily Infographics
freemium
SHARE THIS ARTICLE ON
Share this article
SHARE
Story Saved
Live Score
Saved Articles
Following
My Reads
Sign out
New Delhi 0C
Sunday, November 03, 2024
Start 14 Days Free Trial Subscribe Now
Follow Us On
// // //