The future of chess and elo ratings

[Patrice Duhamel]

In Xiangqi there is no draw, why it's not possible to use the same rules for chess ?

[syzygy]

I am a total chess nitwit but isn't the degree of sharpness measured by some perceived advantage from that opening for one side or the other?

I mean, as against "dull and drawish" which implies that neither side has an advantage.

I would see "sharp" as the opposite of "dull", where I would consider positions that are predictably won to be "dull" as well.

Juan uses the term "complex". Dynamic also seems to be a good term.

Further, how is dynamism to be measured? As in, what evaluation?

Evaluation is not a very a good measure for sharpness/complexity/dynamism. Except that high computer evaluations will usually mean the position is a dull win for one side.

I think absence of relevant conventional opening knowledge together with a neutral low-depth computer evaluation would be two things to look for.

Finally, human chess was thought to have been on its last legs too, partially as a result of advances in opening theory going deep into the game. Hasn't happened yet...

In that case there is no real problem that needs to be solved. But if there is one, then I think it is caused more by opening knowledge that enables players to avoid most pitfalls than by the initial position having a computer evaluation not far enough from 0.

[mvk]

I think it depends on the objective.

If you want the highest quality chess, include books, as excluding them throws away knowledge. The best chess is played in the correspondence chess championships.

If you want to gauge which program is best for analysis, play matches from set positions. But then also play at a set thinking time per move (no rollover of unused time to the next move) and disable pondering and contempt. After all, time control and opponent modelling has no place in analysis.

If you want the know who is the best engineering team, don't restrict yourself to software, certain platforms and/or operating systems, but allow any hardware and books. If you want to prevent a money race in open hardware tournaments, add a cost, weight or power consumption constraint.

If you want the highest entertainment value, do what TCEC does.

My idea has always been to select openings based on the result distribution in trial games. For example: min(W, D, L) * 2.5 >= max(W, D, L). This avoids staring at the evaluation too much, because that is a one dimensional entity and we are interested in an orthogonal concept: sharpness / drawness.

Fisher random is also interesting, although that is perfectly bookable with modern methods (960 is not a big number).

[JoshPettus]

In Xiangqi there is no draw, why it's not possible to use the same rules for chess ?

Actually not true. It doesn't come up as often with the complicated chasing rules and the fact that stalemate is counted as a loss to the stalemating side, but it does happen. Usually it happens when neither side can get any lead against the other, and the position keeps returning, and none of the chasing rules have been violated.

[syzygy]

In Xiangqi there is no draw, why it's not possible to use the same rules for chess ?

Actually not true. It doesn't come up as often with the complicated chasing rules and the fact that stalemate is counted as a loss to the stalemating side, but it does happen. Usually it happens when neither side can get any lead against the other, and the position keeps returning, and none of the chasing rules have been violated.

It is possible to replace the rules of chess with some other set of rules that exclude draws and still continue to call the game chess. For example, there could be a rule stating that the last person to make a move leading to a position that is either a draw by FIDE rules or a position that would entitle one to claim a draw loses.

However, I believe it is a mistake to think that the problem (if there is one) is too many draws. Connect Four can always be won by the first player to move, but that does not make the game more interesting than chess.

I guess what makes a game interesting is the likelihood that players make errors (in the sense that they play moves that change the game-theoretical outcome of the game).

[hgm]

Actually Xiangqi is just as drawish as Chess. Not because of perpetuals, but because of the defensive pieces. These make the draw margin really large.

It is Shogi that has the large draw rate, because the captured material is dropped back, so that the game basically never has to finish before it is decided. In mini-Shogi the rules do not allow for a draw at all.

[mvk]

You can also change the rules (or actually the tournament format) and say that drawing resets the board to the starting position, but leaves the clocks as is. The whole thing then counts as one game. It can be debated what the color assignment should be. I would say unchanged.

[lkaufman]

Here are comments about several above posts.
1. My simplifications overstated the case, but the 35% benefit Kai mentions for using positions near the win/draw line is still pretty significant. I would conclude from his analysis that there is no need to go right up to the win/draw line, just approach it. Maybe +50 centipawn positions would be about ideal, probably still theoretically drawn in general but big chances of losing errors.
2. I have long been an advocate of banning perpetual check, as done in Chinese Chess and Japansese chess. I am also a fan of chess 960, having been US open champion of it, and I also like Seirawan chess, for which Don and I made a Komodo version. But I'm trying to focus on ideas in this thread that don't change the rules of chess, other than by mandating certain openings, for which there is ample historical precedent.
3. Choosing "sharp" openings is another way to increase resolution. The simplest way to do this is simply to select openings from a database of decisive GM games. But I think that pretty soon it will be difficult to find openings popular in GM praxis that offer Black any significant chance to win in a match between top engines. Going for the win/draw threshold is a solution that should last for centuries.
4. I have myself proposed (for human play) the idea of replaying games without resetting clocks until someone wins. But if you are testing two super strong engines at long tc that might take forever. Of course you can avoid draws by playing super-fast games, but that has obvious drawbacks. Note that this idea has very different consequences if you reverse colors than if you don't. If you reverse, play will be more like it is normally. If not, White will aim for just slightly better endgames, because even if they are drawn 95% of the time, as long as only White can win, he eventually will do so.

[Uri Blass]

Imbalanced opening positions will result in (predictable) 1-0, 0-1 outcomes but are not a better way to measure the relative strength of two players.
/quote]

Only if the position has enough imbalance.

We talk about positions that are imbalanced enough to produce 1-0 half of the times but 1/2-1/2 in another half of the times.

[syzygy]

3. Choosing "sharp" openings is another way to increase resolution. The simplest way to do this is simply to select openings from a database of decisive GM games. But I think that pretty soon it will be difficult to find openings popular in GM praxis that offer Black any significant chance to win in a match between top engines. Going for the win/draw threshold is a solution that should last for centuries.

You think going for the win/draw threshold is going to give Black any chance to win??