Did the NTK particularly affect/impact DMK's chances? Let's take a look at the data.
For starters, here's an obvious metric:

NTK's average votes polled in constituencies where DMK+ won: 11,165

NTK's average votes polled in constituencies where ADMK+ won: 9,702

That is, the NTK, on average, polled ~1500 votes higher in constituencies where the DMK won.
This is not something that one can go by. It's possible that the DMK and NTK both maximize their votes in certain constituencies.

So a better metric on top of this would be to look at how the NTK's performance is as a function of the margin of victory or defeat for each party.
Correlation coefficient of margin of victory and NTK's votes polled in constituencies where the DMK won is -ve. That is, as the margin goes lower in places where the DMK won, NTK polled higher.
But here is the kicker. Consider the reverse scenario. It's jaw dropping.

Correlation coefficient of margin of victory and NTK's votes polled in constituencies where the ADMK won is 0.9!

I just couldn't believe my own eyes!
That is, in constituencies where the ADMK won, the NTK performed poorly when the margin was small and performed well when the margin was large.
In other words, the NTK was a significant player in constituencies where the DMK won by small margins. It was a non-entity in constituencies where the ADMK won in tight contests.
To the extent that the correlation coefficient for the latter is 0.9! Like this is as if the NTK were made in an ADMK lab!
Now consider the MNM.

The average votes polled by MNM in constituencies that the DMK won: 4877.

The average votes polled by MNM in constituencies that the ADMK won: 1639.

That is, the MNM, on average, polled 3,238 votes more in constituencies that the DMK won.
And if we look at the correlation coefficients for margin of victory in constituencies where DMK won with MNM votes polled, it's again slightly negative. Just like NTK.

And if we flip that and measure the correlation for margin of victory with seats where ADMK won, it's 0.77!!
Again, it's like the MNM was a non entity in seats where ADMK had a chance and won. But the MNM performed better than its own average in places where the DMK was in a tight race!

Again, it's like the MNM were created in an ADMK lab!
Both NTK and MNM are behaving almost identically. They have a slight -ve correlation in places where the DMK won with the margin. And they have an almost perfect correlation with the margin in places where the ADMK won!
This is something your party should take a look at, @ptrmadurai, @Udhaystalin @ksivasenapathy @skpkaruna
Everyone who asked for the data, find it here: drive.google.com/file/d/1tkoxJt…

• • •

Missing some Tweet in this thread? You can try to force a refresh

Keep Current with Puram

Puram Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!


Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @puram_politics

3 May
One guy I know is from a place called Meerut. In western UP. He was the median BJP voter in 2019.

He’s now posting memes about how the people have rejected BJP in Assembly Elections. Narrative is everything, I guess.
Another friend is from a place called Hathras. Also in western UP. Same story.

I just hope my random anecdotal things end up being representative.
People I know from UP are from obscure places: Meerut, Hathras, Jhansi, Moradabad, Chandauli etc.

I don’t think I can even locate these places on a map.
Read 5 tweets
2 May
In the United States partisans work hard to achieve gerrymandering. In India, the EC does it.

There’s no other way to explain the boundaries of Manamadurai constituency.
Or try Andipatti.
Or try Sankari.
Read 4 tweets
2 May
English News TV simply hasn't learnt anything from @PrannoyRoyNDTV in all these years.

Can't watch election coverage on NDTV. Without Roy NDTV is unwatchable.
@PrannoyRoyNDTV Anchors like Jain don't have an implicit understanding of basic statistical distributions. That's a shame after all these years.
Postal ballots aren’t independent and identically distributed. How difficult is that to understand, Jain? High school students know that.
Read 4 tweets
1 May
Blaming the executive in a Westminster System is an amusingly pointless exercise. But then we no longer have a Westminster System.

So, I guess it’s par for the course that Modi/Shah get blamed now given they have captured government.
But while we blame Modi/Shah, do not forget the GST cheerleaders.

They have blood on their hands in terms of limiting the amount of money and fiscal room states have to combat this crisis.
I mean one could say UP would always have been disaster. But MH, KL, KA and TN would have saved more lives had they had more money.

And GST is directly to blame for that.
Read 6 tweets

Did Thread Reader help you today?

Support us! We are indie developers!

This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Too expensive? Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal Become our Patreon

Thank you for your support!

Follow Us on Twitter!