The much-awaited UP exit polls are out and almost all of them have predicted that BJP will emerge as the single-largest party. In fact, Axis and Today’s Chanakya predict bumper victories for the party. Axis got it right in Bihar in November 2015.
Today’s Chanakya, which shot to fame by predicting accurately the Lok Sabha tally for BJP in 2014, got it horribly wrong in Bihar.
Axis’ recent success in predicting accurately a hung verdict in the BMC elections should bring cheer to BJP supporters.
Both CSDS and CVoter, which are predicting a hung Assembly, have been far more conservative and both have had their share of successes in the past. Sensing trouble, Akhilesh has opened channels of communication with the BSP and is open to the idea of forming a mahagathbandhan to stop the BJP from coming to power.
Before we discuss some key findings, though, we must note that the level of data disclosures by top polling agencies has been appalling, to say the least.
Agencies which were dishing out caste-wise data, leadership ratings, key issues, satisfaction level with the government in each edition of their opinion poll have not published any data except for vote share and seat tally. Only Today’s Chanakya and VDPA Associates have come up with some of this data.
1) Who did the OBCs Vote For?
With an average vote share of above 30 percent across all polls, it is clear that the BJP has retained much of what it won in 2014. The non-Yadav OBCs are the biggest voting block (31 percent) and both the SP and BSP have often struggled with this segment.
In 2014, the BJP won about 61 percent of this segment according to CSDS. According to Today’s Chanakya, the BJP has retained almost its entire non-Yadav OBC vote base in 2017. The VDPA survey also suggests that BJP managed to woo 54 percent of this vote bank.
It is clear from the data that the BJP did extremely well among the largest voting segment. To win UP, the BJP needs to get a minimum of 50 percent of non-Yadav OBC votes to win.
Though the data from other agencies is not available, we can bet that those showing a hung Assembly (CSDS, CVoter) will have 30-40 percent of OBC support for BJP in their data set.
2) Parties Hold on to Core Voters
The attempts by all the parties to consolidate their core voters seem to have been successful.
Today’s Chanakya suggests the BJP and SP held on to their core voters. BJP captured 63 percent of the ‘forward caste’ votes, while SP got 70 percent of Yadav votes and 67 percent of Muslim votes. The VDPA survey suggests that Mayawati performed well amongst the Jatav Dalits (81 percent against 68 percent in the Lok Sabha polls).
But BJP’s gamble of not to having a CM candidate worked in its favour as it was the only party that seemed to have done well outside its core vote base (32 percent of SC votes and 25 percent of Yadav votes as per the Today’s Chanakya poll). This implies it held onto its Lok Sabha vote base among these communities, especially young Yadavs and non-Jatav Dalits.
3) Preferred Choice as the CM Candidate
Most of the surveys did not publish this data. Only VDPA did and it appears that Mayawati performed the best at 26 percent, Akhilesh next at 23 percent. This is contradictory as the BSP is expected to finish third as per the survey. However, if we add all the BJP names together, they account for up to about 40 percent for the most preferred CM candidate.
In the past, this question by itself was a good predictor of election results. In most states, the party with the most preferred CM candidate has won in recent times.
4) Akhilesh’s Performance
Most of the surveys did not publish this data, which is again very crucial. Only Today’s Chanakya did and it appears that 27 percent rated Akhilesh’s performance as poor.
However, only 29 percent rated his performance as good. The low ratings are an indication of a poor voter response to the alliance.
5) Disparity in Vote Share and Seat Tally
The predicted vote share and seat tallies are difficult to reconcile though in the cases of Axis and Today’s Chanakya. Axis predicts a 36 percent vote share and 265 seats for BJP. Today’s Chanakya predicts a much higher vote share at 42 percent, while predicting only 20 additional seats at 285.
In eight out of 10 states which went to the polls after the general elections in 2014, BJP lost vote share, while only in Maharashtra and Kerala did it gain vote share.
Axis’ vote share tally is in line with the approximately 20 percent vote share loss which the BJP has suffered across state polls vis a vis the Lok Sabha polls, and hence believable.
The vote share prediction of 42 percent made by Today’s Chanakya means the BJP has not lost any votes compared to the Lok Sabha polls. This is a bit difficult to digest.
This would mean a Modi wave across the state which nobody could pick up, like in 2014. Additionally, with a 42 percent vote share, BJP’s projected seats tally should easily be 300+ (it was 337 in Lok Sabha with a similar vote share). The vote share prediction by Today’s Chanakya in our opinion is higher by 3-4 percent for such a seat tally.
VMR predicts 1 percent lower vote share compared to Axis, but 65 less seats. Also, in its last opinion poll, Axis predicted 34.8 percent vote share and 185 seats for BJP in January-end. However, now, with a 36 percent vote share prediction (+1.2 percent gain in votes), it is giving an additional 80 seats to the BJP.
This can be partly explained by the fact that once a party crosses the 35-percent threshold which was required to win in a truly triangular contest, the conversion of votes to seat share increases manifolds.
Reliance on Pollsters
The most accurate and aggressive pollsters of recent times appear to be giving a landslide win to the BJP. It needs to be seen on March 11 if they maintain their levels of accuracy.
In the past, when election results have surprised us, often at least one polling firm (sometimes obscure) got the outcome right. Can it be National Dastak this time, which is predicting a BSP win with 225 seats?
None of the pollsters are predicting a SP-Congress win and that is very bad news for the alliance. Their best case scenario is a hung Assembly.
But given the recent experience with polling firms (Brexit, Trump’s win, and the Delhi, Bihar and Tamil Nadu elections), it’s best to wait till March 11 for the actual results.
(This article has been co-authored by Amitabh Tiwari and Subhash Chandra. They are independent political commentators and can be reached at @politicalbaaba and @schandra_100 respectively. This is an opinion piece and the views expressed above are the authors’ own. The Quint neither endorses nor is responsible for the same.)