Your submission is now in Draft mode.

Once it's ready, please submit your draft for review by our team of Community Moderators. Thank you!

Submit Essay

Once you submit your essay, you can no longer edit it.


This content now needs to be approved by community moderators.


This essay was submitted and is waiting for review.

Will LeelaChessZero beat Stockfish in the Jan/Feb 2019 Top Chess Engine Championship?


Another opportunity for a very short-term prediction.

LeelaChessZero is a community-built chess engine that attempts to replicate DeepMind's methodology and success. They don't have the budget of DeepMind to train LeelaZero the way that AlphaZero was trained; instead, they crowdsource the training process, getting online volunteers to run self-play games on their home computers.

It will soon face Stockfish, the current reigning champion chess engine, in the TCEC Championship. (AlphaZero, which can beat Stockfish, is not part of this tournament.)

According to this announcement blog post,

Now there will be a 100 games match between Leela and Stockfish with 120 minutes + 15 seconds per move time control. This will be held after the TCEC CUP that starts on 21 January 2019. Cup is expected to last 15-20 days. The Leela net that will be used there is yet to be decided but it would probably be a test 30 net and not a test 40 net since test 40 is yet too immature. Stockfish will probably use the latest development version at that time.

So we ask:

Will LeelaChessZero beat Stockfish in the ongoing Top Chess Engine Championship (TCEC)?


This question will resolve positive if TCEC declares LeelaChessZero to have beaten Stockfish in the aforementioned 100-game superfinal. In the event that TCEC does not declare a winner, this question will resolve positive if we can determine that LCZ won more games than Stockfish won, out of those 100. Otherwise, this question will resolve negative.

On the one hand, LeelaChessZero is supposed to be like AlphaZero and AlphaZero crushed Stockfish. On the other hand, LeelaChessZero may have less training time and a less expert team of creators. (Allegedly LCZ has roughly 23 million games of training, AZ has 44 million.) On the prehensile tail, maybe AlphaZero's success won't replicate for some scandalous reason.

Make a Prediction


Note: this question resolved before its original close time. All of your predictions came after the resolution, so you did not gain (or lose) any points for it.

Note: this question resolved before its original close time. You earned points up until the question resolution, but not afterwards.

Current points depend on your prediction, the community's prediction, and the result. Your total earned points are averaged over the lifetime of the question, so predict early to get as many points as possible! See the FAQ.