Candidates 2026 Predictions

Now that the Candidates is right around the corner, I want to see what my tournament prediction model says about the eventual winners.

To recap the model briefly, I use only the FIDE ratings of the players as inputs and simulate the tournament 50000 times to calculate the win probability for each player.

If multiple players are tied for first, the tiebreak is simulated using the rapid and blitz ratings of the players.

This post is about the odds of the players before the start of the tournament. I’ll follow along with the tournament and update the probabilities for both sections after each round and you can follow along here.

Open section

I made one slight adjustment to the ratings in the open section, namely I randomised Nakamura’s rating before each simulation to be somewhere between 2775 and 2805. I did this because he has hardly played any classical games since the last Candidates tournament and I thought adding this uncertainty to his rating is justified.

So let’s see the winning chances of each player according to my model.

So my model has Caruana winning one third of the tournaments, Nakamura winning one third and the final 6 players winning the remaining third. Intuitively, I’d say that all players in the bottom 6 would have slightly higher chances than they get from my model, but it’s impossible to say what the correct probabilities should be.

My model also gives a 53% likelihood of a playoff for first place, which seems relatively high.

Given the high playoff probability, I thought it would be interesting to see how the likelihood of winning the tournament outright compares to the probability of winning in the tiebreaks.

One can see that Nakamura wins most tiebreaks, given his high rapid and blitz ratings. Wei Yi also has comparatively better chances in the tiebreaks compared to Giri.

Apart from the chances to finish in first place, I was also interested to see how likely each player is to finish in every position.

This graph has a nice symmetry, as there are two players rated much higher than the field and two players that are rated much lower than everyone else.

Women’s section

Now let’s also take a look at the winning chances in the women’s section.

Zhu Jiner is the clear favourite according to my model, but many players have decent winning chances. Again, my model gives a relatively high playoff percentage with 50%.

Here you can see how the chances to win split between winning outright and winning after tiebreaks for each player.

We can also take a look at the probabilities of finishing at different places.

Apart from Zhu and Vaishali, it looks like every player can end up in every position, which seems like an exciting prospect.

You can follow along with updates to the model after each round here and if you enjoyed this post, you should check out my Substack.