Exit Polls: ‘Bungling’ Pollsters or Too Close a Fight?

Different results are due to close nature of contests as well as the variance in the quality of data collection.

The exit poll results are out, and as has been the trend for the past few years, they haven’t presented a clear picture.

This, despite the fact that they are generally considered a better barometer of results than opinion polls. Though they may not be able to predict accurately seats and vote shares, they do generally indicate the direction of the trend correctly.

Also Read

Will Favourable Exit Polls’ Predictions Help Cong Contain Allies?

The best of the pollsters by track record ‘Axis’ and ‘Today’s Chanakya’ also get it spot on only half the time. As usual, on TV channels the adrenal rush of anchors, politicians and political commentators was for all to see.

Barring Mizoram and Rajasthan, where there is unanimity among pollsters, for the other three states – Telangana, Madhya Pradesh and Chhattisgarh – there is a marked difference in views.

You don’t need to be a pollster to predict the Rajasthan results. The poll of polls predicts hung assemblies in MP and CG, where the role of smaller parties gain importance in such a scenario.

Poll of Polls: 3 Out Of 5 states to Witness Hung Assembly

• Out of the five states which went to polls, only Rajasthan and Telangana are predicted to give a clear verdict, according to the median of polls
• MP, Chhattisgarh and Mizoram are expected to witness a close contest with no clear winner as per poll of polls.

(NOTE: MP - Madhya Pradesh; RAJ - Rajasthan; CG - Chhattisgarh; TG - Telangana; MIZ - Mizoram)

Some Bizarre Facts

• Out of 9 polls for Madhya Pradesh, 3 predict a BJP victory; 3 predict an INC victory – and 3 predict a hung assembly
• The above holds true for Chhattisgarh as well
• For Telangana, half of the polls predict a hung assembly, while the other half predict a TRS victory
• ‘Today’s Chanakya’ which is one of the premier agencies, was NOT hired by any channel to conduct an exit poll

Big No. Ranges Add to Confusion

1. In MP, taking all polls into account, BJP is slated to win anywhere between 90-130 seats, while Congress (INC) is likely to win 86-126
2. In Chhattisgarh, BJP is expected to win 21-52 seats, while INC is heading towards a 32-65, and Ajit Jogi-Congress combine, towards a 0-8
3. In Rajasthan, BJP is expected to win 52-103 seats, while INC looks like it might get anywhere between 81-145
4. In Telangana, TRS is expected to win 48-89 seats while Mahakutumi led by INC 19-59
5. What should one make of these big ranges? Which poll to trust and which to not?

Pollsters Playing it Safe Or Too Close a Contest?

Some predictions do not make sense; either the contest is too close or it gives a sense that the pollster is playing it safe. The pollster, in the end, can take credit for whoever wins. Below are the examples of some of such predictions; almost the same numbers for both the main players.

1. ‘Axis My India’ predictions for MP: BJP 102-120; INC 104-122
2. ‘NewsNation’ poll for Chhattisgarh: BJP 38-42; INC 40-44
3. ‘JanKiBaat’ poll for Rajasthan: BJP 83-103; INC 81-101
4. ‘C-Voter’ poll for Telangana: TRS 48-60, grand alliance with Congress (mahakutumi) 47-59

Rajasthan

The state is a no-brainer for political commentators and analysts because of the strong trend of anti-incumbency. In a way, it’s the Kerala of North India. ‘Jan Ki Baat’ is the only agency which states BJP could win, and break the trend. Its call is an outlier in this case.

Telangana & Mizoram

It seems that KCR’s gamble of calling for early polls is working.

Even the agencies which have predicted a hung assembly show that he has enough scope to form the government with BJP, AIMIM and Independents. None of the agencies show a ‘Congress + TDP + TJS + CPI’ grand alliance ahead of TRS.

In Mizoram, both polls predict a hung assembly, so the northeast’s fascination with close contests and ‘jod-tod’ politics continues.

Madhya Pradesh & Chhattisgarh

These two states have proved to be a nightmare for pollsters.

BJP is battling a 15-year anti-incumbency in these Hindi heartland states. However, its chief ministers Shivraj Singh Chouhan and Raman Singh remain highly popular. The Congress has suffered from factionalism in both the states.

Jan Ki Baat, CNX and C Cube (which the author is associated with, however not included in the poll of polls) predict a BJP victory, while CSDS and Pace Media predict an INC victory in Madhya Pradesh. CNX and CSDS predict a BJP victory in Chhattisgarh, while Axis My India and Pace Media predict an INC victory.

A Political ‘Khichdi’

Voters of these 5 states and viewers nationally were eagerly waiting for the exit polls, but rather than helping, they have added to the confusion. Many people planning to take market positions are baffled.

The different results are due to the close nature of contests as well as the variance in the quality of data collection. Either this, or the basic premise on which these polls are built – that the voter is speaking the truth – no longer holds true. Let’s wait and watch. Some reputations will be demolished while some new stars could emerge.

(For live updates on the exit polls, click here.)

(Amitabh Tiwari is a former corporate and investment banker turned political commentator, analyst and strategist. He tweets @politicalbaaba. He has worked in the states which went to polls advising some BJP candidates.)