Thread by @shrikanth_krish

Uttar Pradesh, the 8th largest country in the world, clearly decided the course of the 2019 elections

A province where Modi's BJP won 60 seats despite having a united opposition of SP and BSP arrayed against him!

A wonder of a result

Let's study this more closely

First the basics - let's look at the seat tally

The BJP seat tally fell by just 11, despite the consolidation of the two largest opposition parties

Interestingly while BSP went up from 0 to 10, SP stayed put at 5, the same as in 2014 when it went alone

Then let's move to vote share - which tell us the bigger story

A 7% jump for the BJP.

And a 5% drop for SP + BSP relative to 2014

Now back in February I had undertaken a study projecting the winner for each constituency assuming -

a) Perfect vote transfer within SP-BSP combine
b) Exactly the same vote shares as 2014

This had given me the following tally -

BJP : 35
SP+BSP : 41
INC : 2
OTH : 2

So the 5% swing away from SP / BSP and the 7% swing in favor of BJP caused the BJP tally to increase from 35 to 60!

This clearly shows -

A) Vote transfer was far from perfect. Many SP / BSP voters from 2014 did vote BJP

B) The first time voter factor - UP clearly had at least 10MM first time voters if not much more

And this segment arguably leaned BJP strongly

While B. is a hunch from my end, A is something we can study more objectively

So let's take a constituency to show how MahAgaThabandhan failed.

We will also pick a constituency to show where MGB worked.

Let's pick Jhansi. A town in south western extremity of Uttar Pradesh. A town made famous by its famous queen Lakshmi Bai

This is a classic constituency where MGB failed.
Failed royally

But with the assumptions I stated in my February thread, it should have worked.

Jhansi 2014 result :
Victor : Uma Bharti

Vote shares :
BJP : 43.6%
SP : 29.2%
BSP : 16.2%
INC : 6.4%

Now with perfect vote transfer and exact 2014 vote shares -

SP+BSP should have won 45.4% of the votes (29.2 + 16.2), and won the seat this year

How did 2019 pan out?

Here you go!

Jhansi 2019 result:

Uma Bharti did not contest the seat. Instead you had Anurag Sharma, a less pedigreed candidate

Vote share:
BJP : 58.6%
SP : 32.1%
INC : 6.2%

So BJP increased its vote share from 43.6% to 58.6%!!!

BSP did not contest thanks to the alliance

But the 16% voters who voted for it in 2014 did not shift allegiance to SP at all

In fact, SP's vote share barely increased despite BSP support. From 29% to 32%

No transfer

Whereas BJP's vote share increased by 15%

Where did this increase come from? Did the BSP voters now vote for BJP? Instead of the SP candidate?

Or did the BSP voter abstain?

Is it the new voters in 2019 who contributed to BJP's increase?

If we look at the total # of voters -

It was 13.2 lacs in 2014.
13.8 lacs in 2019

Some increase. Though not a lot

So clearly in this case at least one can safely say the BSP vote moved to the BJP!

Despite Uma Bharti abstaining!

Let's pick another constituency - also beginning with J

This is Jaunpur. A town that lies to the north west of Varanasi. Not far from Modi's own constituency

This is a fine example of a seat where MGB worked. One of the few seats where it did.

2014 election result:

Victor : BJP's Krishna Pratap

Vote share:
BJP : 36.5%
SP : 17.9%
BSP : 21.9%

Comfortable win for BJP

But with perfect vote transfer, you'd expect SP+BSP to win here. 17.9%+21.9% = 39.8% > 36.5%

And this happened

2019 election result:

Victor: BSP's Shyam Singh Yadav

Vote share :
BJP : 42.3%
BSP : 50.1%

Here we see SP vote nearly transferring to BSP

Why?

My take - Yadav candidate fielded by BSP

In 2014, BSP fielded a Pandey.
While in 2019, the Yadav candidate did the trick, with a neat transfer of the Yadav vote to BSP

Also interestingly the votes to parties other than BSP and BJP totally evaporated in 2019, resulting in a bipolar contest

So in much of UP, the scenario described in Jhansi played out.

With much of the vote transfer happening to BJP and not the alliance partner.

Jaunpur is a contrary example where MGB worked as intended

In part because of the judicious choice of a Yadav candidate by BSP

UP remains fascinating

A deeper study is required of several more constituencies for which I don't have time

Under what circumstances did the vote transfer happen

Under what situations it failed?

Worth studying

Post-script : My earlier thread from February where I had projected for each seat assuming "perfect" vote transfer is here for reference

https://twitter.com/shrikanth_krish/status/1095895548249468928

Post-script2 : Some typos

In 1st tweet, I mentioned BJP tally as 60. It is correctly stated as 62 in the 2nd tweet. So the decline w.r.t 2014 is 9

Also mentioned UP as 8th largest country. Actually it might be 5th. Was comparing 2011 estimates for UP with 2019 est for countries