IM TAPANI SAMMALVUO
Chess betting is an almost completely unstudied branch of sports betting that offers a lot of possibilities for specialists. Few, if any, bookmakers have acquired the services of a chess specialist and consequently it is possible to find favorable odds rather frequently.
The most common types of bets on chess available are the result of an individual game, the outright winner of a tournament, and the final score of a match. In team matches handicap matches are sometimes offered.
On these pages, we will give our recommendations to the most important chess events on offer, but in case you are interested in making your own analysis, we recommend you read this article very carefully.
There are three possible results in a game of chess: White wins (1-0), Black wins (0-1) and a draw (½-½). What makes chess different from most of the other games is the possibility for the players to agree on a draw at any time they want, unless stated otherwise by the regulations of the tournament. This is a factor that makes the analysis more complicated, as the tournament situation dictates the degree of risk the player should take.
Chess has a rating system, called Elo, whose predictive power can be used as the starting point of the analysis. At the moment of writing, the Elo of the number one player of the world, Magnus Carlsen, is 2834 and, for example, the Elo of the number 100 in the world, Anton Kovalyov, is 2652. The difference between their respective ratings is 182 points, making the predicted score between them in a 100-game match 74-26 to Carlsen. However, you should note that these numbers include draws, so Elo in itself is not enough to give a good estimate of the result of an individual game. Of all the games in my database (see below) White has won 39%, 29% have been drawn and Black has won 31% (the numbers don’t add up to 100% because of rounding).
The international chess federation FIDE publishes new ratings in the beginning of every month at http://ratings.fide.com/ and almost always you can find the players’ ratings on the official web page of the tournament. If you wish to use the so-called “live ratings” (updated daily), we can recommend https://2700chess.com/.
It has actually been proved that the Elo is not the most accurate way to predict the results, but it is not far off and we tend to tweak them only slightly in our own analysis. You will find the Elo expectancy table below.
Another important factor to take into account is the advantage of the white pieces. White begins the game and thus has some theoretical advantage, sometimes compared to the advantage of the serve in tennis. The size of this advantage in practice depends on the level of the players and their playing styles, to mention just a couple of factors. In average, White scores about 54%.
Microsoft Excel or other spreadsheet is usually enough for plugging in the numbers, if you don’t want to write your own program.
ChessBase’s Mega Database 2018 or similar is very useful in finding the distribution of results at any level and between any level of opponents. Note that besides the database, you’ll need a chess database program such as ChessBase 14 to use it. We have a slightly expanded version of the Mega Database 2018 with a total of 9,3 million games.
Furthermore, if you wish to calculate the probabilities for a tournament winner, you will need to be able to run some Monte Carlo analysis. It is doable, if slightly cumbersome, in Excel. However, note that the probabilities of the games of different rounds are not completely independent from each other, and especially the probabilities of the last round are heavily influenced by the tournament situation.