Will World Chess Championship 2024 be a close match?
➕
Plus
20
Ṁ4668
resolved Nov 27
Resolved
YES

The match is considered close if any of these conditions is fulfilled:

  • each side wins at least one classical game

  • no side wins a classical game (and match thus goes to tiebreaks)

The match will be played between Ding Liren (current champion) and Gukesh D (challenger).

More info: https://en.wikipedia.org/wiki/World_Chess_Championship_2024

Get
Ṁ1,000
and
S1.00
Sort by:

i disagree with this definition of "close". https://en.wikipedia.org/wiki/World_Chess_Championship_2008 and https://en.wikipedia.org/wiki/World_Chess_Championship_2014 are just two examples of matches where both players won a game but the eventual winner had 80%+ odds of victory by game 6

@diadematus How would you define it in a way that is not overly complex but also objective?

Also my main objective was to check if the match will be one-sided or not. If each side wins a game, it is a strong argument that it wasnt. Maybe one-sided would be better term to use in the title.

@Weezing well there already exists an aggregate opinion tracker here - https://manifold.markets/Weezing/who-will-win-the-chess-world-champi - so something like: "resolves YES if one player has >80%/85% odds for at least 6 hours after the end of game 10/11 etc" is clearer imo. a bit subjective and manipulatable but with the volume on that market i doubt it. it also avoids e.g the situation in Vishy-Kramnik where Vishy races out to a dominant (one-sided) 3-0 lead...his victory is never in doubt but the losing player statpads a meaningless victory towards the end.

@diadematus Carlsen - Caruana was also very close for example but had 12 draws which would fail this market

@diadematus Carlsen/Caruana would be YES, because no one won a game.

Anyway I don't like linking resolution to another market. It makes it messy in my opinion and I personally usually avoid markets like that. Then it is no pure prediction, but you also have to account for chances of someone gaming that second market (which is not negligible). Also if someone is leading by game 10/11 they will be a big favorite even if the match before that was close. Maybe you could change the numbers a bit, but it's not easy and if you don't get the numbers right you can easily end up with way worse approximation.

I agree that there will be always some holes and for any set of rules you can find some exception, but I think this one approximates what I wanted to ask relatively ok (not perfectly though), while being simple and fully objective.

@Weezing I made a market that just goes by the margin of victory. https://manifold.markets/ChinmayTheMathGuy/world-chess-championship-2024-resul?r=Q2hpbm1heVRoZU1hdGhHdXk

In theory the best way to determine how close a match is to determine the match win probabilities for each state (assume the W/D/L loss probabilities are something like 20/70/10).

Then measure how much the win probability changed over the course of the match (it starts at 50/50 and ends in 100/0) and the more change the closer the match.

@ChinmayTheMathGuy Yea, that's more precise, but I wanted YES/NO market.

@Weezing you could ask is the match within 1 point (7.5 to 6.5 or tiebreaks)?