Chess Futures - Hyper Accelerated Live Chess Ratings Speculation
Chess Futures is a rating system based on the Live ratings by 2700chess.com which in turn is based on FIDE ratings. What is so interesting about Chess Futures is it calculates its rating using part historic, part predictive and part speculative elements. The idea came to me when I was looking at a FIDE rating list and as I looked up and down the list I could see both male and female GMs, IMs and randoms playing out of their skin and winning events like stars some were shooting stars, rose suddenly and disappeared just as quickly, while others stuck around like the Carlsens, Wei Yi's, Polgar's and Hou Yifans of the chess world. I looked at the 2700chess.com rating list to see if it could make sense of the phenomenon. It couldn't, the Live chess ratings are primarily interested in joining up the monthly FIDE ratings. This means that is not really an instantaneous rating system as many believe, it takes the current FIDE rating and calculates new ratings based on your results which are in turn based on the win-loss-draw results against your opponents and adds the change. It does this on a daily basis, which sounds instant enough right? Wrong. They recalculate the same win-loss-draw result based on the FIDE rating that is sometimes a month old. They do this so that their results always tie back to the monthly FIDE ratings. There is nothing wrong with this, in fact when I spoke to 2700chess.com about this they were pretty open and honest "We don't want to cause disparity between our ratings which are calculated on a continuously daily basis and the FIDE monthly ratings."
Ok so I sat down with a blank sheet paper and pen, old school, I thought how I could create a rating system that was the following.
1. Dynamic - Responsive to recognisable elements that constituted measuring chess play and chess playing potential. Chess Futures looks at chess stock in motion, so players who are not making any rating changes are not even considered. Think about it if you just maintain your chess rating never losing rating points or gaining them then you might as well not be playing chess. Far too much information on existing rating lists is 20% about summing to zero the zero sum changes and 80% the immovable restatement of previous ratings. Chess Futures is about the changeable moving and dynamic 20%.
2. Gender neutral - Men and women play chess and a large number of the 740,000 FIDE registered play in a single pool but women only events is a reality that FIDE created and it distorts the performance of women players. About 100,000 chess players registered to FIDE are female, the other 640,000 are men. In women only events rating changes limit the progress of women since the upper limits of the women only event pool are the highest rated women players, while in the open events their is no limit since the highest rating player, 2nd highest player and so on are available to win rating points from. The rating system had to counter this two population mean distribution skew by being gender neutral.
3. Credible - I've seen a lot of weird and wonderful rating systems that are touted as next best thing, chess.com ratings, Grand Chess Tour Universal Chess Ratings, MSA Chess Ratings. I prefer to talk about measures of strength, performance and effort, than just rating. The Chess Futures combines all these.
4. Allow for financial speculation on chess rating elos, and tournament predicted and actual outcomes.
I built the Chess Futures rating system as a dashboard for predictive performances, so the current index of players combines those elements of strength, performance and improvement (effort).
What was pretty cool about this rating system is Carlsen was #6 on the list and Giri #1, but the most pleasing part was yet to come, 3 days into going live it had predicted a high performing Grandmaster Lei Tingjie was going to do really well in her Sevilla Open tournament in Spain. This was at the same time as Vishy and Giri were slaying it at Tata Steel. Well as the Chess Futures index indicated GM Tingjei got to #2 on the index and Shakhriyar Mamedyarov shot out of nowhere and Vishy started waning. This was all predicted on the index, the whole time despite Carlsen's performances and massive rating he sat at #6.
Chess Futures can be used as a trading index on speculations on the future Elo of any chess elite player. I posted some futures assets for bidding on www.chessbidder.com to prove the point. The reason conventional betting companies do not take daily speculations on chess is there has been no predictive index until Chess Futures.
For those who are interested, the Chess Futures index is live here -> https://twitter.com/ChessFutures
I'll take the bow now if this particular lunchtime project kicks off big style.