Thread by @karthiks

A whopping 332 out of 542 constituencies in the just-concluded General Elections saw a two-way contest. Another 184 saw three-way contests.

In contrast, in the 2014 elections only 169 two-way contests, 278 three-cornered contests and 90 four-cornered contests

How do you determine the number of corners of a contest? It is using a formula called ENPV (Effective number of parties - votes. Thanks @vdehejia ). It is the inverse of the sum of squares of the vote shares of each candidate in each constituency (en.wikipedia.org/wiki/Effective…)

Once you calculate this for each constituency, you round the number (for the sake of convenience), and you get the effective number of parties.

For example, if two parties get exaclty 50% vote each, then ENPV = 1/(0.5^2 + 0.5^2) = 2.

If one party gets 90% of the vote and the other gets 10%, then ENPV = 1/(0.9^2+0.1^2) = 1.22. So it's barely more than one party!

If one party gets 70%, and 10 other parties get 3% each, then the ENPV will come out to 1/(0.49 + 10 * 0.0009) = 2

So here is the number of corners of contest in these elections and the elections the last time round. Clearly a lot more straight fights this time. In that sense, the opposition coalition sort of worked.

So limiting to parties that "effectively contested the seats" (I know it's not mathematically accurate but it makes intuitive sense), here are the 10 most common contests in this year's elections.

The parties have been ordered in descending order of votes.

In other words, 146 constituencies saw a two-way fight in which the BJP beat the Congress. The next most common contest was BJP beating SP, in 26 constituencies. BJP beat BSP in 21 constituencies. YSR Congress beat TDP in 14. 12 BJP-Trinamool_CPIM contests. Then single digits

Another way of looking at the above table - in 8 constituencies, Congress beat BJP in a straight fight. So out of 154 straight fights between BJP and Congress, BJP won 146. Take a moment to digest that.

In 2014, the score in BJP-INC bilaterals was 98-8 in BJP's favour.

A larger number of two-cornered contests meant that the BJP needed many more votes to win the same number of seats. And maybe they recognised this in the campaign stage. And they got the additional votes. Overall vote share for BJP went from 31% in 2014 to 37.5%.

Among the seats that BJP contested, its vote share was 45.3%. Again the large number of bilateral contests meant that the BJP needed lots of votes. And that was delivered.

This makes 2019 a far tougher victory than 2014. While the party was the challenger in 2014, opposition disunity meant it needed far less votes for a majority. Not the case in 2019.

This was partly balanced by the fact that the BJP was in power this time, and had the ability to influence the Election Commission's schedule. And did a masterful job of fixing the schedule to its own advantage.

OK I don't know how to finish the thread in a nice way so I'll stop now. Basic takeaway is that opposition alliances worked in some way, but that the BJP recognised this and possibly worked extra hard to get extra votes and won the election.