Numerically, the draw-odds in a 24-game match depend on the frequency of draws.

It is actually not a bad idea that people who want to discuss this issue get some feel for the subject. For instance, it is fairly easy to set up these quick-and-dirty numerical simulations: (1) Ignore the difference between playing with White and Black pieces and assume independence between the games. Also assume that the two players are evenly matched and, therefore, a win can be assigned to either player with equal probability. (2) Generate random sequences of wins and draws; tabulate the empirical statistics.

Draws in chess these days are about 60 percent of games. So in a 24-game match, about 14 games can be expected to be draws (assuming your “evenly matched players”). And if the remaining 10 games can also be won by either player with *equal probability*, well then the match can be expected to end 12-12. What else would you expect?

The discussions regarding the two main match formats (24-game match, vs first player to reach 6 wins) had to do with theoretical attempts to discourage draws. It was said by some that the 24-game format encourages the player in the lead to try to coast by just playing for a draw, and it was assumed that playing to 6 wins, with draws counting for nothing, would alleviate this.

The exact opposite is what happened. The 1984 WC match between Karpov and Kasparov became a drawing nightmare. Draws constituted a full 83% of the matches played there.

For the first 10 games, the draw average was about normal (60%), and Karpov was up 4-0. Then it seemed that Kasparov decided his best strategy was to hunker down and turn it into a contest of long-term stamina by playing for draws (Had this been part of a classic 24 game match, even with that strategy the match would have ended after only 20 games with the score of 12-8.)

It seemed to work. 5 months and 40 games later Karpov looked like a ghost, had lost a lot of weight, and was described as being on the verge of a physical collapse.

If this is what happened in a match to 6 wins that went on from September to February, and then was cancelled, you can imagine what could happen in a match to 9 wins. It could maybe last a year.

Clearly not a very practical format. But what is important to remember is this. In a match to a limited number of games, where draws count as half points, a tied match is an inevitable possibility. It comes with the territory. For the match to be of limited duration, a drawn game must count for something, and so it counts half a point. It is simply the small price to pay for the well-defined time frame this format offers. So it makes sense that in the case of a tied match, the champion retains the title.

In a match to a limited number of wins, where draws don’t count, there is no way of limiting or predicting the total number of games it will take to achieve those wins. (The 9-9 clause does not help solve this problem.) Therefore, the notion of a "tie" when both players reach a given number of wins is a completely artificial construct totally unnecessary to this format, and introduced solely for the purpose of giving the champion an additional advantage (additional to the already great advantage of not having to play any earlier rounds like the others).

John Nunn made some comments on the “draw problem” a few years ago: http://tinyurl.com/m552bm

Draws in chess these days are about 60 percent of games. So in a 24-game match, about 14 games can be expected to be draws (assuming your “evenly matched players”). And if the remaining 10 games can also be won by either player with *equal probability*, well then the match can be expected to end 12-12. What else would you expect? ... >

Ok, we ought to be a little bit careful with statistical concepts: In the scenario you describe (14 draws, 10 decisive games, evenly matched players), the probability that the standing champion retains the title is <~62.3%> -- in other words, roughly <5:3> odds in favor of the standing champion.

(1) A match to 14-games with 'classical' time controls; it goes to a tie-breaker if tied.

(2) The tiebreaker:

(i) 1-3 games.

(ii) One side gets the draw odds, the other side gets the white pieces and/or extra time on the clock.

(iii) Lower limit on the time would be 1/2 hour a game and side. (Upper limit would be the classical time control.)

<<(The probability of a drawn match in the said case is ... (10 choose 5)/2^10 = 252/1024 = 24.31% ... ~1/4; the probability of a decisive match is ... ~3/4.)>> ------------

<<DWINS: <tedceldor>, Can you explain what you mean by <fischer's anti positional style>?>>

In attempts to classify players styles, Fischer was sometimes described as being prone to making “antipositional” moves in the sense they use this word today to describe some computer moves. These would be moves that seem to run counter to basic positional principles. In the recent book by Kasparov http://tinyurl.com/rcxvl8 where he analyzes and tells the story surrounding all his encounters with Karpov, there is a quote from Petrosian to the effect that each grandmaster is always an amalgam of styles. He says something like (I am quoting by memory): Tal is not only “sacrifice” Petrosian is not only "caution", Capablanca is not only "position" and Fischer is not only "electronic computer chesss".

So it seems that Fischer had some kind of reputation for making computer-looking moves, which today are sometimes described as "antipositional".

That Fischer was first and foremost a positional player is masked by his sharp openings and drive for the initiative. Unlike Capablanca and Rubinstein who essentially waited for a weakness to exploit, Fischer actively pursued creating them with tactics.

There are so many Fischer games where his opponent seems not to be in the game at all but just a foil for Fischer's genius.

Which player was it that played a famous game which was won without any piece exchange?

In the final position not a single piece had left the board - yet the opponent resigned due to the hopeless situation...

I have forgotten who it was that was able to win such a game... was it Karpov, Nimzowitsch, Reti...? Or somebody else?