Wednesday, February 27, 2008

Mark Ginsburg's History of USATE Stacking

Dr. Mark Ginsburg offers a wonderful history of stacked teams at the U.S. Amateur Team East (see "The Fabulous 00s: Curing the GGGg Debacle at the US Amateur Team East 2008"). That history includes his own winning teams, one of which instigated the original anti-stacking rule in effect after 1994 for the four-team playoffs (that is, until the famous "Karpov exception"). Ginsburg writes of his 1994 team with Ilya Gurevich, Victor Frias, himself, and "the requisite low-rated kids": "What a trio of scum-sucking stacked opportunists. It was Karmic revenge that we forfeited in the playoffs after Frias pulled a no-show in our first match versus the South."

Though I'm not sure I like Ginsburg's "competency test" solution (to disqualify teams where the bottom board can't even put together one non-forfeit, non-bribed point), it's really wonderful to hear a famous reformed USATE team stacker discuss the GGGg case with the brilliant insight that only he can provide!

I always enjoy Ginsburg's nostalgic chess blog, which offers more proof that those who do not know chess history are doomed to repeat it. As for Steve Doyle: what's his excuse?

11 comments:

Anonymous said...

I don't think you've done a good job articulating what exactly your problem is here (despite going on at great length with respect to the issue). A multitude of teams with 2 GMs, an expert/master, and a weak 4th board have been fielded with ne'er an objection. Such a team is statistically going to perform just fine against "GGGg" since the first two boards are a wash and there's more uncertainty on board 3 than on board 4. An expert or weak master can sometimes chip a GM for a draw where the 2GM-Master-1500 team is always going to win on 4.

Your empty complaints about "What it means to be an amateur event" don't hold up to any kind of logical scrutiny. 2 grandmasters and a master are an "amateur team" where 3 GMs and a baby aren't? What logical standard are you using to divine that, precisely?

Since we've established that another acceptable way to stack the team is actually favored to beat GGGg, your objection then shifts to the lack of "fun" and "competition" which you, an expert, had on board 4. Well, duh. Plenty of teams consist of a GM and his 3 horrible little students or a group of friends where the 4th board is very weak. You're 2000-2100 playing the bottom board when people are forced to keep an average of 2200. Of course you're going to be much stronger than most 4th boards, and of course some are going to be horrible.

On 3rd board of a team that went 5-1 I only had 3 opponents who were able to play interesting games against me, and I'm nothing special myself. The bottom boards are going to tend to be fairly weak, and there's nothing to be done about it.

In the final analysis, your complaint really seems to be that this team plays better chess than yours, and in so doing impedes your chances at victory. Well, that's the nature of competition.

Your arguments about "ethics" are similarly incomprehensible to me.
Again, teams which would be favored over this one have been fielded in the past (including the Karpov team, which has a lock on 4 and a big edge on 1). So talk some specifics here if you can. What logical argument can you construct to support the contention that this team is so much worse than the other lineups I've described?

-Brian Meinders

Michael Goeller said...

Look, the real issue is that some of the organizers have been under the impression that doing anything to get GMs to the event -- even allowing them to form completely stacked teams -- outweighs the view among most players that stacked teams simply don't belong at the event. The Karpov team was certainly a terrible case of stacking at the time (though by today's standards they were hardly a complete lock since they risked loss on boards 3 and 4). The more important issue is that back in the "Karpov exception" year there were not many forums for amateur players to voice their concerns and so those concerns went unheard. Now we have a way of expressing opposition to stacking, and we should exercise it to change the rules for the future. After all, this is the US Amateur Teams -- the tournament "for the rest of us" -- not the "Tournament of the Stars" or "Tournament Organized by GM Bootlickers So that Some GMs can Make a Buck." If they changed the name to reflect what happened this year, I think you'd have a smaller turnout and maybe some smart person would organize a real Amateur Team event.

No team at USATE 2008 had a realistic chance of beating the GGGgs. It was possible for a team with a GM on top board to draw them, and thus kill their chances at ultimate victory, but even there the single-GM team was laboring under draw odds on top board. And, frankly, I think only GMs Nakamura, Christiansen, and Benjamin had even a remote chance of beating Izoria.

So, basically, you are saying you're fine with the fact that out of over 290 teams only 3 had a realistic chance of drawing the GGGgs? Meanwhile it's fine for the GGGgs to ruin the chances of more balanced teams to compete for the title -- a title which the GGGgs can never really hold anyway, since they are disqualified from going on to represent the East?

You can talk about some theoretical teams that might be able to beat the Three-Gs, but as a practical matter you do not normally see such teams at the event and, as far as I could tell, there really were none present to offer that chance. I certainly have no trouble with most two-GM teams (even the Karpov team, by the way, should be allowed in my view), since most balanced teams would still have a shot at beating them on boards 3 and 4 and maybe even drawing on 1 or 2 (that is, unless Karpov or Kasparov show up, in which case only on board 2). Two GMs is not a completely stacked team. Three GMs is the problem, and at the very least the organizers should acknowledge this and prevent it in future years.

The organizers made a miscalculation in allowing the GGGg team to compete. I hope that in offering a forum for discussing this matter, I eventually help to prevent stacking next year so that the Teams remains a fun and successful tournament long into the future.

Michael Goeller said...

What's the argument for allowing stacked teams at USATE? Is there anyone (besides some GMs who might be paid to participate) who would benefit from that? How so? I really don't understand your perspective in favor of allowing -- or perhaps encouraging -- stacked teams in the future. Perhaps you can enlighten me.

Anonymous said...

The organizers do have a right to determine who belongs in the event. The means by which they do this is the establishment of rules governing the formation of teams. They define an "Amateur team" as a team with a rating under 2200. You might disagree with that definition. Fine, start your own tournament with your own definition.

But to label as unethical a team which acted in a manner fully consistent with the rules established by the tournament organizers is unfair. The fact that they have a 2500 on board 3 instead of a 2200 like the other teams I described doesn't make them a bit less ethical.

Also, you don't think Yudasin, for example, has a remote chance of beating Izoria? A 2500 is going to score about 1 in 4 against a 2700. How many teams does that give a 25% chance of beating/drawing GGGg? Quite a few have a 2500+ on 1st.

Your next argument is that relatively few teams have a chance to win the tournament. Isn't that true in just about every chess tournament you'll ever attend? Come to any major national tournament with me and I'll be able to tell you who the top 5-10 most likely victors are with relative ease, and it would be a shocker if anyone outside that group won it.

Normally out of the 300 teams playing I don't think more than 10-15 would have any real shot at winning. You're saying in this tournament it was more like 5. Even if we grant your premise, I fail to see the problem here.

As a 3rd board, I would have loved to have the chance to get to play a match with a grandmaster. Most people who love chess would rather get to play a tournament game with a former world championship candidate than have a shot at winning a cheap clock and a plaque.

You talked about the "spirit of Amateurism", but your writings aren't consistent with that spirit. "Amateurism" comes from "amatus", which is derived from "amare", meaing "to love". And that's what this tournament is supposed to be about: love of the game, enjoying good chess with your friends in a casual environment. What's better for a lover of chess than getting to play such awesome opponents for so cheap a price?

It's like getting to shoot hoops for a few hours with the Chicago Bulls and being indignant at the opportunity because it interferes with your chances to win a trophy. There's something profoundly sad about that to me.

Michael Goeller said...

Actually, there was only one team -- GGGg -- with a realistic chance of winning the event this year. I forgot Yudasin -- so make that four teams with a GM theoretically capable of beating Izoria and therefore capable of ruining the GGGg's chances to win. But you misrepresent my argument: while these one-GM teams had a chance of drawing GGGg, none of them had a realistic chance of winning the USATE if they had to play GGGg -- they just have a chance to draw them and ruin both their chances. Basically, GGGg spoils the chances of everyone they play.

And yes, the organizers have a right to make rules on what counts as an amateur team. When those rules run counter to the rules at other USAT events and against the rules that most amateur players consider to be fair for the tournament, they should be changed. My reading of the letter written by Carol Jarecki to GM Perelshteyn (e.g.: "others disagree [with Doyle], asserting the rule of not more than a 1000-point difference between boards 3 and 4 has been in place for so long that it no longer needs annual publication") is that it was fully within the rights of the organizers to bar the Three-Gs this year, based on the rules, but Doyle (and perhaps others) felt that having GMs at the event was always good -- no matter what. I hope that the obvious "controversy" will get him to reconsider his position in the future -- and I think most players at USATE (including most GMs) would agree that stacked teams are bad for the event.

As for "unethical": I think it is unethical to be paid to play on an unbeatable stacked team at an amateur event. You are welcome to disagree with that, but I do not think anyone would do that for the "love of chess."

Michael Goeller said...

And as for the "chance to play a GM" argument -- that one has been widely refuted in various places. One need only point out that having three GMs on three different teams would lead to just as many "chances to play a GM" (albeit on only the top boards).

Anonymous said...

"Actually, there was only one team -- GGGg -- with a realistic chance of winning the event this year. I forgot Yudasin -- so make that four teams with a GM theoretically capable of beating Izoria and therefore capable of ruining the GGGg's chances to win. But you misrepresent my argument: while these one-GM teams had a chance of drawing GGGg, none of them had a realistic chance of winning the USATE if they had to play GGGg -- they just have a chance to draw them and ruin both their chances. Basically, GGGg spoils the chances of everyone they play."

You're not giving nearly enough credit to other teams. Our first board, Tom Bartell, can beat pretty much anyone on a good day. I believe that Stoyko's capable of this as well, as are many 1st and 2nd boards. A team with 3 strong masters need only score 1 point in 3 to draw, and that's plenty possible. There are more than enough strong players that this can happen more than once. And yes, they can even lose.

"When those rules run counter to the rules at other USAT events and against the rules that most amateur players consider to be fair for the tournament, they should be changed. "

Most amateur players don't have a problem with this. You and a few other commentators, as well as a few other disgruntled opponents, seem to be the only ones complaining. You acknowledge on your own blog that there isn't any kind of consensus opposition.

"My reading of the letter written by Carol Jarecki to GM Perelshteyn (e.g.: "others disagree [with Doyle], asserting the rule of not more than a 1000-point difference between boards 3 and 4 has been in place for so long that it no longer needs annual publication") is that it was fully within the rights of the organizers to bar the Three-Gs this year, based on the rules, but Doyle (and perhaps others) felt that having GMs at the event was always good -- no matter what. I hope that the obvious "controversy" will get him to reconsider his position in the future -- and I think most players at USATE (including most GMs) would agree that stacked teams are bad for the event."

It's been published for so long that it doesn't need to be published anymore? Are you serious? The vast majority of players never would have known such a rule to exist were it not published. They still bother publishing that the average has to be under 2200, and that's much more obvious and long-standing than a rule that hasn't been in effect for the better part of a decade, so I don't know how you expect people to buy that.

And no, the directors don't have the right to turn a team away because of some non-published arbitrary "rule". That's the whole reason for having rules ahead of time and publishing them: so that people know far ahead of time what formations are permissible.

"As for "unethical": I think it is unethical to be paid to play on an unbeatable stacked team at an amateur event. You are welcome to disagree with that, but I do not think anyone would do that for the "love of chess.""

I've already explained how this team is beatable by formations which have arisen quite frequently in previous years. I'm also not so sure why you think it's an impossibility for a GM and a pair of strong masters to score 1.5/3 against them. But you seem determined to ignore this point, so I'll move along. I do think they're three friends who wanted to play together for the love of chess. They've said they wanted to sit next to each other and play on a team, and I take them at their word. It's not like there's big bucks on the line here.

"And as for the "chance to play a GM" argument -- that one has been widely refuted in various places. One need only point out that having three GMs on three different teams would lead to just as many "chances to play a GM" (albeit on only the top boards)."

One might consider this a refutation if he hasn't the time, ability, or inclination to give it a moment's thought. But upon so doing it'll be quickly evident that putting GMs on 1-3 gives the opportunity to play a GM to 3x more people than just having them on 1st. Everyone on 1st board already had a chance. With this team in the running, 2nd and 3rd boards (some of whom, like me, have never had the opportunity to play a GM in a tournament game) had a shot.

Anonymous said...

Think about the math for a second. Introducing the same GMs into a pool of players without a single GM is going to result in many more opportunities than introducing them to two player pools without a GM in them.

Anonymous said...

Bah, it doesn't let me edit. "Introducing the same GMs into a pool of players without a single GM is going to result in many more opportunities than introducing them to a pool of players which already is populated by GMs" should have been how that read. Mea culpa.

Michael Goeller said...

If you want a chance to play GMs, form the strongest team you can at next year's USATE with you on Board 1 (our Board 1 on the Kenilworth B team has gotten to play lots of GMs over the years). Or play Dzindzi in a simul. If you saw how he played practically every game on Board 3, you'd recognize that the experience of playing Dzindzi on Board 3 at USATE would be basically very similar to playing him in a simul. I think he generally spent a total of 10 minutes actually seated at the board in any game.

Other than offering six amateurs a chance to play a GM on Board 3, I do not see the benefit of having stacked teams -- except for the few GMs who might form them specifically in order to get paid to play by a wealthy patron.

I still don't understand why you want to allow stacked teams, especially when it would improve the tournament experience of many players to ban them in the future.

Anonymous said...

"If you want a chance to play GMs, form the strongest team you can at next year's USATE with you on Board 1 (our Board 1 on the Kenilworth B team has gotten to play lots of GMs over the years). Or play Dzindzi in a simul. "

If you want to play a tournament where pretty much everyone can beat pretty much everyone else, play something with class sections. Virtually every major tournament out there has them. This one doesn't, and part of the reason for that is that anyone might get to play anyone. You might not be comfortable in a situation where you could get paired with a grandmaster or a D-class player, but some of us do enjoy that, and this tournament is one of the few that caters to us.

"Other than offering six amateurs a chance to play a GM on Board 3, I do not see the benefit of having stacked teams -- except for the few GMs who might form them specifically in order to get paid to play by a wealthy patron. "

Well, we have two distinct issues here. One is whether it was wrong/immoral/unethical/illegal for GGGg to participate. All of your arguments as to why this is the case have been, in my view, convincingly refuted. Unless you have anything to add there, I'm content to let that part of our discussion stand and your readers can judge for themselves.

The new question you're posing is whether it's better to allow such teams or not, and here I might be a bit more sympathetic to your position. Personally, I think the best solution would be to dispense of calling it an "amateur" tournament and take the rating cap off. You could have both an open section and class sections, and the winners of the U22 section could be the "Amateur Champions" while the winner of the Open would be the "National Champions".

Within each of the class sections you could prohibit there from being a 500-1k difference between boards and I'd not object. I think that sort of tournament sturcture would be much more interesting and vibrant.

If we're going to keep the 2200 cap on the entire tournament, then I can see an argument for the 1k point difference rule. I don't think there's anything inherently unfair or unethical about it, but it does allow a wider range of team formations to be competitive. If the majority of players wants that to be the case, I'd be happy to go along with that.

What I'll not do is conflate the two issues and pretend that simply because more people would rather something which is permitted by the rules not be, we should act as if some sort of travesty has taken place in GGGg having adopted a perfectly legal team formation. If your purpose be to change future policy, I think you'd also be well advised to separate the issues and argue the former case on its own merits.

While that might not appeal to the sour-grapes resentment of those who lost to GGGg, it does allow you to present a case free of illogical and bitter-sounding assumptions, and has the ancillary benefit of not labeling as unethical three of the most prominent players in the country.

-Brian