Pistol rounds: why round 16 is worth (almost) double round 1

 in Categories CSGO, Metagame tags Tags ,

What I’m examining here is the importance of the pistol round to the final result. The way Counter Strike competitive games are structured implies that the first and second pistol round have the same value, both teams take turns playing either side and both teams take turns with the pistol rounds. The question is whether this is true, and how we can test for it.

Examining the first pistol round results is easy. The scores are always 0-0 and no team should have an advantage going into it. Winning the first pistol should usually set a team up for an eco advantage and a decent shot at a couple of additional rounds, so there’s should be a clear relationship to winning the first pistol and winning the overall map. There is, 60.2% of the time the team that wins the first pistol round wins the map, so all things being equal the first pistol round give a team an extra +10.2% chance of winning the game.

Moving onto the 2nd pistol round if the value is equal as implied by the structure of competitive Counter Strike, then the overall value should be the same. If we repeat the same analysis as above then we should get a similar answer… but we don’t. Teams that win the 2nd pistol round win 62.5% of the time. That’s +12.5% chance of an overall map win just from winning that one round.

This of course ignores the fact that the 2nd pistol round is a considerably more complicated animal. There have been numerous rounds already, the score could be lopsided, one team already psychologically broken. Surely by that stage the better team, the one that is most likely to win the 2nd pistol, will already have a comfortable lead and that is distorting the results?

For anyone following this line of reasoning the results make uncomfortable viewing. If we restrict the result set to only include maps with 1 round difference between the teams at the point of the pistol round – that is very close games – the win percentage of the 2nd pistol round winner rises to 68.9%. That’s +18.9% win rate just from winning that one round.

This includes scenarios where the team winning the round is ahead as well as behind, so being more specific we can break it down

Pre-round score differenceWin rate %
-330.5%
-163.75%
+173.9%
+382.4%

There is an obvious difference here between round 16 and 1, in round 1 the score is 0-0 and before round 16 the score must have a difference of at least 1. If we combine the numbers for the winner of the second pistol round being a round ahead and a round behind we get 69.2% win rate (note that this isn’t exactly the same as the simply averaging the two win rates because the number of times the scenarios occur isn’t exactly the same)

This is +19.2%, nearly double the +10.2% win chance bonus you get from winning the first pistol round. We can say that with confidence because we’re looking at, on average, two teams with no difference in round score and no difference in economy as the mid game reset has happened.

So why is this round so much more valuable? Each pistol round just adds one onto the overall score, and they don’t provide any additional eco bonus over the other. The answer is in how you model the win rates of rounds in the game.

The goal of Counter Strike can appear to be a linear target, get to 16 rounds and you win the game. But that ignores the fact that it’s a race. The two real variables to consider are how many rounds is one team ahead or behind (and their eco difference), and how many rounds remain to turn it around. Both pistol rounds only secure a single round on the scoreboard, but they also provide a significant guaranteed eco advantage.

When a team loses the first pistol round there are still potentially 29 rounds in which to turn it around. After the 2nd pistol round there are only 14 rounds maximum left to reverse the advantage, and because there’s a nice eco advantage guaranteed with it, it usually amounts to more than one round advantage.