Phase V Predictions & Who wins Bihar!

The D-Day is about to arrive. The Phase V (Day 5) of the gripping Bihar Cricket Match is scheduled today. By 5 pm the fate of all the candidates in Bihar polls will be sealed. After 5 pm all channels will start with their exit polls and by 8 pm we should know who is slated to win Bihar (hopefully). The outcome will decide whether Manjhi is the “Mountain Man of BJP in Bihar” or his debut is a damp squib.

I still maintain that this is a damn close elections and every vote will count. This election has uncanny similarities to Delhi elections of 2015. In Dec. 2013 when the people of Delhi gave a hung verdict, 2 exit polls predicted simple majority for BJP, one predicted AAP victory while two others predicted a hung assembly. So exit polls can give misleading results even though they are considered to be better than opinion polls. Bihar EXIT POLLS may also show a similar picture – different polls may give different predictions / winners. So the suspense would continue till 8th of November for sure.

Phase V of the polls is very crucial for both players – Janata alliance (Maha Gathbandhan) and NDA. MGB is strong in this Phase leading in 44 seats in LS polls if you aggregate their vote shares together. MGB needs to sweep this Phase to trump NDA. NDA knows well it can’t lead in this phase, has to hang in there and push back MGB.

Many factors are in play in Phase V and I wrote a detailed article in Dailyo explaining the 11 reasons why the last phase of Bihar polls will change everything.

Now let’s see what’s working and what’s not working for both these alliances:

Working for MGB:

  • Nitish Kumar’s popularity – This may well be one of the biggest clues to Bihar assembly results. Read my c-authored piece in Hindustan Times with Subhash Chandra for more. If NDA is expected to win, how does he get such high ratings.
  • No CM candidate of BJP (ostensibly to prevent infighting among Bihar party leaders and not make Manjhi / Paswan / Kushwaha unhappy)
  • Muslim-Yadav strong votebank of Lalu giving the alliance a minimum 20% vote share (conservative estimate) to start with

Not Working for MGB

  • Lalu’s jungle raj regime horror stories – can it really negate Nitish popularity rating advantage
  • 41 seats on which Congress is contesting (in 28/41 seats its opponent is BJP)
  • High number of rebel candidates

Working for NDA

  • Modi factor – Record crowds at Modi rallies even in MGB bastions. If MGB is winning, why are so many people jostling to hear him speak. And is BJP infra able to convert these votes!
  • Upper caste (15%) afraid of MGB coming to power rallying behind NDA including Banias (7%)
  • Loss of votes in translation for MGB, 100% seamless transfer of votes not happening

Not Working for NDA

  • Absence of credible local face to take on Nitish
  • 80+ seats on which allies are contesting – smaller allies like HAM / RLSP could have been merged with BJP, then allotted lower seats, also solving recognition / recall of symbol problem – brilliant point by reader @Purple_Truth
  • Inability to manage and prop up Owaisi could harm especially in Phase V

God Knows Working in whose favour?

  • High percentage of women voters exercising their franchise
  • Comparatively higher turnout
  • Beef ban controversy

The trump card (paplu) – Youth & Most Backward Class votes.

Revised Phase 4 Predictions

I had carried a  post poll analysis & predictions after Phase 4 in which 2 scenarios emerged based on 43-47 seats for NDA in Phase IV.

Scenario 1: NDA 105, MGB 79, Others 2

Scenario 2: NDA 114, MGB 70, Others 2

Since then, estimates for Phase 4 have been revised downwards by 4 seats so now its 39-43 for NDA in Phase 4. See tweet link. So the revised scenarios now till Phase 4 are:

Scenario 1: NDA 101, MGB 83, Others 2

Scenario 2: NDA 110, MGB 74, Others 2

Phase 5 Analysis

An overview of Phase 5 shows NDA has 25 sitting MLAs in this region. However, it was leading in only 8 seats in LS polls, that too on 5 seats, the lead advantage was less than 5%. MGB was leading in 44 seats and its lead advantage was huge >10% on 40 seats. Can NDA sitting MLAs negate this huge MGB lead? It remains to be seen. Removing the 4 seats which Pappu Yadav was leading in Madhepura, MGB lead reduces to 40.

Rebel candidates from both sides have made the contest interesting. There are rebels in following seats which I could trace Araris – Congress rebel, Jokihat – BJP rebel, Trivenigaj – JDU, Chatapur – Jagannath Mishra relative from JAP, Alamnagar – JDU rebel, Madhepura – RJD rebel. Additionally, some candidates from JDU have got ticket from NDA and vice-a-versa. Both parties have also given tickets to some independents who did well last time in 2010.

NDA is relying on Pappu Yadav to split Yadav votes, but the chap has also put up other caste candidates and is seen hurting even NDA candidates in some seats. It remains to be seen how much of a damage he can do to MGB.

Tariq Anwar’s NCP has decent presence in Katihar and again can do some damage to MGB in these seats. But can he do so much damage that NDA candidates win from these seats. That’s a big question mark?

AIMIM has put up 6 candidates and is well placed to win 1 seat. High number of Muslim candidates per seat (approx 6) in Seemanchal makes the contest pretty interesting.

Back to my favourite test match funda

These 5 phases are like 5 days of a test match. In a test match pendulum swings from one team to another during the five days (one team has upper hand on some days while other team has upper hand on other days). The fortunes have swing in favour of both alliances during these various phases.

Phase 1 / Day 1: Janata alliance had a clear upper hand

Phase 2 / Day 2: NDA had a marginal upper hand

Phase 3 / Day 3: NDA had a very good upper hand

Phase 4 / Day 4: NDA had a clear upper hand

Phase 5 / Day 5: ?

NDA (18-26 seats) has taken a lead and it is now upto Janata alliance (Gathbandhan) to chase down this lead and win the match. Chasing down on the last day is always difficult in however good form your batsmen are. So there will be intense pressure on Janata alliance to perform and sweep Phase 5. That too when stakes are high. Its been a long campaign, volunteers / karyakartas have become tired and Janata alliance needs to keep the motivation levels high. In a way NDA will not be even contesting all the 57 seats, rather focus on 25-30 seats only. Polarization amongst Hindus could turn the tables in NDA favour (a lot of visible efforts were made on this aspect).

Phase V Predictions

By the time Phase V ends and I come up with post poll predictions, exit polls would have started. If I publish numbers after exit polls, I will be influenced by the same. So no post poll predictions on Phase V. Pre-poll predictions equals post poll predictions as well.

NDA is expected to win anywhere between 13-17 seats in this phase. MGB 37-41 seats and Others 3 seats. Based on this the scenario estimates are as follows:

Scenario 1: MGB wins – NDA 114, MGB 124, Others 5

Scenario 2: NDA wins – NDA 127, MGB 111, Others 5


Why is it difficult to predict this elections?

There are many inherent contradictions in these elections. The interplay of these complex & moving in opposite direction forces would determine the results of these polls.

Janata alliance was ahead since Day 1 of these elections based on arithmetic. They successfully deployed this arithmetic in state bye polls where they won 6/10 seats and exhibited that their votes can be transferred. However, elections not all based on arithmetic but also chemistry. NDA assumed poor chemistry in MGB, however, forgot to take care chemistry in its own venture. Could one-upmanship between Manhji and Paswan have hurt NDA? If yes by how much only time will tell.

So who will win Bihar?

Unpredictability of Elections

Elections are always unpredictable. Janata janardan hai! Yeh jo public hai yeh sab jaanti hai! While all pundits wasted tonnes of paper writing who will support NDA in 2014 as it was expected to fall short of majority and whether NDA could change PM under pressure from allies, BJP got an absolute majority on its own. Similarly, nobody predicted a mammoth victory for AAP in Delhi earlier this year. In 2004, Atalji lost despite India Shining. There are many such examples in history of elections in India.

Can there be a hung assembly?

Though there are significant number of rebels (1/3rd-1/4th of seats), chances of a severely hung assembly are remote. These candidates are not winning many seats, max 5 as per my estimates. They will though damage prospects of either alliance in many seats. In any case the largest party / alliance will not be short of majority by more than 5 seats.

Vote Shares

MGB started with 45%, NDA with 39% on an approximate basis. Mahadalits expected to add 3%-4% to NDA kitty. Yadavs who voted for NDA seem to have gone back to their godfather and take away 2%-3% vote share. The key is the MBC vote on which nobody has a clue. The clue here again is in the hypothetical aggregation of Janata alliance (MGB) votes. Can 1 (RJD/Congress) + 1 (JDU) become 2? Can this alliance seamlessly transfer votes from one seat to another across partners? Tricky! Are Yadavs voting for JDU candidates? Are Kurmis voting for Lalu? Are upper castes voting for Congress when they are pitched in battle with upper caste candidate of BJP? Are Mahadalits voting for Paswan? Are Dusadhs voting for Manjhi? Is Kushwaha able to garner majority of Koeri votes (seems to be struggling)?

Who do I put my money on?

The krackjack moment (50:50). It could go either ways. Very close call!

My final call – NDA has a 65% probability of winning the polls / emerging as the largest combination (scenario 2) while MGB has a 35% probability of winning the elections / emerging as the largest combination (scenario 1).

Thanks to all the supporters / followers for their support and encouragement. A long arduous three months covering the polls with almost 50 insightful articles and an E-Book with Subhash Chandra (who has been as source of inspiration). This has been a learning experience which I have thoroughly enjoyed.

From evening, back to exit polls, Part 2 of E-Book and results on 8th. Happy reading….


  1. Excellent report and analysis. Alls well that ends well and hopefully in the end India gets back on the development path. Every effort should be made so that Modi can lead India to a bright future and regressive attitude of Rahul Gandhi should be defeated at all costs


  2. Reading your articles on last couple of days on Bihar election and believe me its awesome.I would also like to see scenario 2 on 8th


    1. Media is full of prédictions including one with BJP/NDA getting 60 (none other than Surjit Bhalla in Indian Express). Its been a most fascinating contest and I enjoyed your analysis but the proof of the pudding is in eating.


  3. Fist post claiming that most of the bookies are worried who had bet on NDA and now doubts whether it can touch 100 or not? Any reports or news on exit polls?


  4. Now that we have heard all TV channels poll prediction, which is 2 vs 2 split (Todays Chankya and Ciecero for NDA and C voter, ABP for MGB), whats your prediction?


      1. I remember Chanakya is the one which predicted 320+ for NDA in LS 2014 polls…
        How reliable is Todays Chanakya? Why they are giving good upper hand for NDA whereas others are giving around 110?


      2. Chanakya usually gives the higher end of the range. If you are aware the polls have a +-3% margin of error. It usually goes upto the +3% and is fairly aggresive. That’s their style. Eg they forecasted 61 in Jharkhand NDA got 41.


    1. In phase 3 they are giving just 13 seats whereas patna has 14 seats out of which 9-10 seat has upper caste dominance where Nda easily win….they r manipulating very less


      1. Political Baaba what is your comment on phase 3 & 4, it seems like huge variation between different predictors??


      1. don’t be surprised if abp give 10 seats extra to bjp 2morrow and decrease 10 seats of mgb frm third phrase- make it even 118-120 , so they don’t feel embarrass on counting day. todays chankya stunned all the media. final result will be 140-145 nda mga- 80-90 other 10-15, diff will be 4-5% vote share between 2 parties and chankya till never get wrong in bi-polar contest, only time they got little wrong in maha coz multi-party contested-


      2. I agree with ur comment. If necessary correction is done in ABP Exit poll then this poll will be tilted towards NDA


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s