Heading into the back half of the tournament, every player needs to ask themselves this question. Pawnalyze is here to help!

So far I’ve run over 500,000 simulations with results through Round 7, and we can inspect the simulations to figure out how likely each player is to win if they achieve a certain score.

Ok… now shut up and show us the data!

In order of most likely to win the event, to least:

How to read the graph: If Nepo scores 9 points, he wins just over 80% of the time. If he scores only 8.5 points he wins 50% of the time. It’s worth noting that some of the wins come from playing in tiebreaks, unfortunately, I didn’t split that out when making the graphs.

After every round, check out my simulations page for updated predictions. And, don’t worry about waiting until the end of the round! You can filter my simulations for the outcome of each game and get an instant updated predictions. You can even use my crystal ball to see the future. Pick the outcome of unplayed games to see the imapct on everyone’s winning chances.

]]>For the first time ever the Candidates Tournament and Women’s Candidates Tournament will be held at the same time and place. Toronto welcomes 16 chess stars vying for the right to challenge Ding Liren and Ju Wenjun for their respective World Chess Champion titles. I’ve simulated each tournament 1,000,000 times using players’ April 2024 Elo rankings and my machine learning game simulation model. I’ll share the predictions for each tournament below, including my thoughts about each event. Read about my prediction methodology, if you’re interested.

It’s no secret that Ding Liren has struggled in several recent tournaments, including Tata Steel (where he finished with a TPR of 2678) and the Grenke Chess Classic (where he finished tied for last place) I’m confident that betting markets will favor *any* Candidates Tournament winner to beat Ding Liren in the World Championship match. Hopefully Ding returns to fighting form to defend his crown. Why does any of this matter? Whoever wins the Candidate is the **likely** new World Champion. With such honors come money, prestige, and a pretty sweet title. In years past, the Candidate’s tournament was crowning a challenger, who would then go on to lose to Magnus… this year is different. The winner will be in good shape to take the title of World Chess Champion.

**The Candidates**

*And, their chances of winning!*

Name | Win % |
---|---|

Nakamura, Hikaru | 22.7 |

Caruana, Fabiano | 21.3 |

Firouzja, Alireza | 14.9 |

Nepomniachtchi, Ian | 14.9 |

Praggnanandhaa R | 8.9 |

Vidit, Santosh Gujrathi | 8.2 |

Gukesh D | 8.5 |

Abasov, Nijat | 0.2 |

Fabiano Caruana is the favorite to win the event if it doesn’t go to tiebreaks. However, the event is expected to go to tiebreaks 23% of the time where Hikaru has an edge in both Rapid and Blitz. This gives him a very slight overall edge based on strictly Elo simulations.

Fabi finished 2023 with the highest TPR of any player, even surpassing the TPR of Magnus for the year, at a whopping 2808. Hikaru was close behind him at 2804. The next highest Candidate is not even in the Top 5 for 2023 performances, behind Wesley So’s 5th place performance of 2756. Does recent performance matter? Who knows - in my opinion, Fabi and Nakamura are the real favorites to win this event.

It’s hard to count out Ian Nepomniachtchi who won previous two iterations of the Candidates Tournament, but the reality is he didn’t have a great 2023. Chess pundits like Daniel Naroditsky suggest that Nepo has fantastic prep for events like this and should not be counted out. Obviously, he has great winning chances.

What about Alireze Firouzja? He seems like a bit of a wild card to me. To my average joe chess eye, his rise has stagnated a bit over the last couple years and I think he’s likely to finish in the middle of the pack.

And that’s just half the pack! The three Indian stars could also do some damage. Gukesh, Pragg, and Vidit all have serious chances. I think Abasov is going to have a rough go - I imagine every single player will use their absolute best prep against him because that potential one point with white is so valuable.

*And, their chances of winning!*

Name | Win % |
---|---|

Lei, Tingjie | 20.2 |

Goryachkina, Aleksandra | 21.0 |

Lagno, Kateryna | 17.4 |

Koneru, Humpy | 17.4 |

Tan, Zhongyi | 10.7 |

Muzychuk, Anna | 9.1 |

Rameshbabu, Vaishali | 3.3 |

Salimova, Nurgyul | 0.5 |

Honestly, I don’t know a whole lot about the Women’s field. My predictions are purely Elo based so I don’t “need” more than that to make predictions. It will be exciting to learn more about all these players as the event unfolds.

One storyline that I am aware of: Praggnanandhaa and Vaishali are brother and sister! Both hailing from India, they’ll make history simply by being in the tournaments together, and they have an outside chance to become an unprecedented brother-and-sister duo of World Chess Champions. I’d love to see that!

After every round, check out my simulations page for updated predictions. And, don’t worry about waiting until the end of the round! You can filter my one million simulations for the outcome of each game and get an instant updated predictions. You can even use my crystal ball to see the future. Pick the outcome of unplayed games to see the imapct on everyone’s winning chances.

In the example below, you can see Fabi’s odds of winning go from 21% all the way to 35% if he beats Hikaru in round 1. A critical game!

]]>If you asked a Grandmaster what would they say? And is the position complex for humans, computers, or both? Do you know if a position is complex when you look at it?

Today, I am introducing an open-source tool that predicts the complexity of a chess position *for a human*. The tool is called Elocator, and I hope you find it *interesting*.

I chose to define complexity as the expected change in Win % after a move is made. Imagine a position where white has a +1 advantage from Stockfish. That implies a 59% win rate for white. Assuming Stockfish is perfect, a human can only play a move that is as good or worse than Stockfish (i.e., white can not play a move that does increases the win rate for white). We know that after the next move is played, white will have a 59% or lower chance of winning.

Depending on the position a grandmaster may find the best move, or maybe it’s a really difficult position to find the best move. Over a large enough dataset we can make correlations between the state of the board and how much we expect the win % to go down after a move is made. As an example, over 20,000 moves, my data shows that a GM is expected to lose 1.4% win rate after a move is made in a position with a queen on the board, compared to 1.3% if there is no queen. That seems small, but also implies positions are about 7% more complex when there is a queen on the board (1.4/1.3).

I created a dataset of FENs mapped to the loss in Win % from a GM that made a move in that position (classical OTB games only). Underlying this tool is a neural network (AI, deep learning, yada yada) that has been trained on 100,000 chess moves made by grandmasters. The model has learned to predict the complexity of a position by learning the expected change in Win % after a move is made, as measured by Stockfish 16 at depth 20.

The model is then used to predict the complexity of a given position, 1-10. The model is not perfect, but it is a good starting point for understanding the complexity of a position. I look forward to making it better over time.

When building a model like this it’s good practice to use some of the data to build the model, and then set some data to the side. This let’s us evaluate the model on data that wasn’t used to build the model.

With roughly 100k moves in my dataset, I decided to keep 20,000 off to the side for evaluation, and I’ve plotted the model performance below.

**Actual vs. Predicted by Complexity Score**