Re: Hastings


"Chess One" <OneChess@xxxxxxxxxxx> wrote in message
news:fYCdnTh7a5XbHB7anZ2dnUVZ_tijnZ2d@xxxxxxxxxxxxxx

"David Kane" <davidekane@xxxxxxxxxxx> wrote in message
news:qPKdnS_7o9_MRh_anZ2dnUVZ_hadnZ2d@xxxxxxxxxxxxxx

"Chess One" <OneChess@xxxxxxxxxxx> wrote in message
news:wtWdnYnnnsMSPB_anZ2dnUVZ_r-vnZ2d@xxxxxxxxxxxxxx

"David Kane" <davidekane@xxxxxxxxxxx> wrote in message
news:Tdqdnbco87Hg9B_anZ2dnUVZ_sejnZ2d@xxxxxxxxxxxxxx

A real world illustration of why it is wrong to rate draws.

While I understand you general reservations on draws relating to Elo - I
would be a bit miffed if, in another example, I had struggled against a
much higher rated opponent, drew, but failed to achieve any rating benefit
from it.

You would be miffed only because of your expectations.

Not at all. If I should draw with a player 200 points above me, shall I not
enjoy the benefit of it -mathematically speaking- related to my rating?


That's exactly the point. You are trained to expect to "benefit" from the
draw
by being given a higher rating. If draws weren't rated then you would have
no such expectation. You would still benefit from the knowledge that
you played well.

Sure. But isn't the rating a record of my playing well? My expectation is that
the result of the game will be recorded. The game itself may have been
indifferent in terms of chess art - therefore the 'play well' only really
relates to acheiving some sort of score against the rating of the opponent.

I'd guess most people can remember the game in which
they drew their highest rated opponent, but I'd guess very few remember
(or care) how many rating points it was worth.

Really? In turn I would guess that /everyone/ would expect to get a ratings
increase for playing-up 200 points and scoring a draw.

I do not understand your objection to rating draws, which seem to be not
mathematical at all, [is this merely my poor understanding of your schema?]
and counter-intuitive to the entire idea of achieving a rating by actual
play against others.


Because not all draws are play. The example you quoted being a
perfect example. Having ratings and titles depend on those shams
is a mockery.

But that is to do with all the factors involved, and can be simply eliminated
by, for example, stating no draw offers before move 30. OR by changing the
score for a draw so that the strategy in our example would be a foolish one.

True, but the most straightforward way would be to not count draws as
rateable results. I see only two downsides: First, that it goes against
existing custom. Second, that there could be some minor unquantified loss of
accuracy to the rating system.

The 30 move rule is just a joke - nothing is going to stop people from
drawing when it is in their best interest. Changing the score for a draw might
help
but is even more radical. The point is that changing the rating algorithm is
a simple step which moves things in the right direction. Until you can
show that every draw is fully contested, why rate them?

I suppose you could always have a more complicated system which
counts draws for lower rated players, or for huge rating mismatches.

I /do/ agree with you that it seems indecent to achieve titles and awards from
these things - and when there is money involved then the rules become 'tested'
by, in this case a leading player simply guaranteeing himself some of it - but
I do not think this is typical of all chess, nor that some abuses by some
players should skew honestly obtained draws by all players.

In summary - there seem to be at least 4 factors at work here:

1) The efficacy of various rating systems
2) Scoring systems for WDL
3) Rules for short-draws
4) Awards of titles

All of which have differing bases of operation - and need independent
resolutions. But let me finish with what I think we both feel is wrong - that
this variety of systems as result of playing chess needs adjustment, have
needed them for a long time, and rules-makers seem deaf to the issue.

I will ask the ECF about this Hastings incident, especially items 3 & 4 above.
I have asked USCF about the Tanner situation, which involved items 1 [quality
control] & 4 [rating-floor awarded for a politico!] above.

But all 4 need to be taken into account so that the result is harmony among
them for most players, while ensuring that some players do not skew the system
for money or prestige titles.

Phil Innes








.

Relevant Pages

  • Re: Hastings
    ... enjoy the benefit of it -mathematically speaking- related to my rating? ... It is most unfortunate for the young player to score his third norm this ... pseudo-games has nothing to do with mathematics. ... not allowing draw offers at all - but I am not so convinced of that. ...
    (rec.games.chess.misc)
  • Free Gift from the Chess Players Guild
    ... Rating, Opponent's Rating, Difference between the 2 ratings, Selected ... Highest rated opponent draw. ... Selected players current unoffical Rating ...
    (rec.games.chess.politics)
  • Re: Hastings
    ... "Simon Williams has qualified as England's ... rating exceeded 2500 and as he has secured three qualifying results in ... rated opponent, drew, but failed to achieve any rating benefit from it. ... allowing draw offers at all - but I am not so convinced of that. ...
    (rec.games.chess.misc)
  • Re: Hastings
    ... a much higher rated opponent, drew, but failed to achieve any rating ... If I should draw with a player 200 points above me, ... I /do/ agree with you that it seems indecent to achieve titles and awards ... abuses by some players should skew honestly obtained draws by all players. ...
    (rec.games.chess.misc)
  • Re: Hastings
    ... eliminated by, for example, stating no draw offers before move 30. ... accuracy to the rating system. ... But what do the other mathematicos writing here think of your idea? ... I have this idea that the game belongs to its players, not administrators, ...
    (rec.games.chess.misc)