That changes the question from “how does a TD handle this existing prize fund?” to “how does an organizer sidestep needing to worry about it?”. You have just provided a valid option to address the second question. Does your answer to the first question remain “the organizer pays an extra $100 to the unrated and uses the full original prize fund for everybody else”?
I would be OK with adding that to the rules. It’s a political solution, but the rulebook has many other political solutions. I suspect it wouldn’t pass at a Delegates meeting, though.
I don’t think that’s a correct interpretation of the ratings formulas. We compute an ESTIMATED pre-event rating for someone who doesn’t have one based upon the player’s age, but only when we have no other basis for making an estimation. (And if we don’t have the player’s age, we make the estimate based on the player’s membership type.)
It used to be that we encouraged people to say that such a player has no published rating and was therefore unrated, but these days many organizers will see if the player has an unpublished rating online and use it. (Whether that’s advisable or not is a separate issue, my views on this are well known.)
Not his post-tournament rating. It is a bit complicated (especially if the player is young), but it would likely be around 400 points below the rating of the lowest rated player he lost to (with a minimum around 100.) All we really know about a player who has lost all his games is that he is probably weaker than his weakest opponent.
Ah, yes. That nixes, or at least revises, my previous response. Player A was probably in the bottom class of some class event. OK, so 5-0 in U1100, wins $100, post-tournament rating in the high 1400s (P5), probably still some room to win a class prize next time around.
I appreciate the discussion of my earlier question. Agreed it’s likely that player A entered a class D or below section of her first event.
I have two ways I have handled unrateds as an organizer at a class event:
-
Have an unrated prize in the lowest section and specify that unrated players may win the unrated prize only. Pros: Rest of field is protected from a potentially strong unrated player. At least one unrated player goes home with something. Cons: You have to add a prize to the prize fund. There may only be one unrated player.
-
Let unrated players in the lowest section without a prize limitation. Pros: Simple. Prize fund structured as normal. Nine times out of ten, the new player is lucky to score 0.5/4 against a murderers’ row of nine year olds, and it doesn’t matter. Con: one time out of ten, an unrated player with years of very good self study gets the perfect score at the player here did. Counterweight to con: the new player gets a rating of top opponent plus 400, and she won’t be in this section again. She probably ought not have been in a class section in an ideal world, but miscategorized players are an inherent risk of class sections and class prizes. Sandbagging? Yawn. One cannot participate in a system that encourages the behavior and complain about it afterwards.
tl;dr: There is a better way to solve the problem of unrated prizes, assuming it’s a real problem. And it may not be a real problem.
Now that I’ve made the philosophical case for either stand-alone unrated prizes or no limit at all, if we are faced with the facts in Col. Hater’s OP, the correct answer is Tom Doan’s in the first reply. (The corollary to the correct answer is Ken Sloan’s reply.)
I agree with Mike’s point.
I also think that going back to Jeff’s options, the validity of the options is impacted by understanding Mike’s point. There is, for example, an argument for option #5 in Jeff’s list; The argument falls in line with both what Mike is saying, and what Tom is saying (with a bit of a twist).
If we consider “the players can’t do better than if player #1 didn’t exist” is the implication to player #1 only, or to player #1 and the prize that would have been associated to a random unlimited player #1? There is a reason for this question - and it comes back around to Mike’s general question. Is the point to say that, had player #1 not existed, that players #2 and #3 would have finished as #1 and #2 (a ridiculous argument since all the conditions would have changed and who knows what would have happened) or is the point that Player #1 should not benefit from questionable (incorrect or insufficient) information about player #1’s strength?
If its the latter, then there is a very strong argument for Jeff’s option #5 (i.e. players 2 and 3 did in fact finish 2nd and 3rd and should receive what reality is, not some phoney baloney guess at what reality would have been had reality been completely different 
) - and the question - based on Mike’s thought process, should be “What happens to left over $$?”
Whenever we have to go through complex analysis like in many of the posts here, it seems fundamentally wrong. In the words of Lasker “long analysis is wrong analysis.”
I would however, reword Mike’s question: Not “who is to benefit from the limitation”, but rather, “what is the point of the limitation?”
If the point is to discourage players from intentional or unintentional sandbagging, should the players who finished 2-3 do better in terms of prizes than their actual finish warrants?
Assuming this was a section of a tournament, then Player #1, knowing they were limited, may have an incentive to play in a more appropriate section. (BTW-Creating higher limits for higher sections , would create at least some motivation for this player to self-select to a more appropriate section.)
If the point of limitations is to remove sandbagging (intentional or not) motivation, then why immediately assume that the other players in the event are harmed? It’s POSSIBLE that the unrated-who-performed-well chose the correct section and played well or just benefited from tournament luck. (We’ve each had those events where things just went our way.) But the underlying assumption of 32C6 is that intentional harm was done to the other players, and therefore the other players should receive the benefit of the limitation. It’s not clear to me that should be so immediately and strongly assumed - because if it is, the implication is that the limitation didn’t work anyway! So why do it?
I’m not in favor of awarding prizes based on hypothetical results.
A whole lot of confusion about prize fund allocation could be eliminated by stating the following (what I think are obvious) principles.
Other things being equal
- A player or group of players cannot win more if their scores were lower.
- A player or group of players cannot win less if their scores were higher.
- A player cannot win less if another player’s or group of players scores are lower.
- A player cannot win more if another player’s or group of players scores are higher.
Most of Jeff’s 5 possibilities fail one of these. (The $400 each to the 4.5’s fails point 3 since if you give the limited player 4.0 rather than 5.0, they would only get $375 each. Several others fail point #3 because 2 and 3 get less than they would if the limited player also were at 4.5, or equivalently fails #2 if 2 and 3 were also at 5.0).
It is very difficult to come up with a plain English description of how to pay out prizes which works for all prize funds and all collections of scores, as long as there are any restrictions on who can win what. (WinTD uses a variant of linear programming, and I haven’t yet figured out how to extend that to do prize distribution with limits since the limited player is not trying to maximize anything). No matter what you try to put into the rule book, someone will figure out some way to parse it to confuse things—and in most cases, that will run afoul of common sense (i.e. #1 to #4).
Ding, ding, ding–we have a winner.
It’s simple the winner “may have bagged” This is purpose of limited prizes. 100 400 400.
And the winner may not have bagged, (which has already been covered in this thread.) And why do only 2nd and 3rd benefit from this state of epistemic ambivalence? Why not other players? Or why not the organizer? And even if the winner bagged, why would these two players be the only beneficiaries of that, or the beneficiaries at all?
It looks like some people will simply and firmly disagree.
The prize fund was 500 - 250 - 150 - 100
Tom Doan said it should be treated as 100 - 500 - 250 - 150 (or he would have if he had known about the fourth place prize when he first answered it). This is exactly what I’ve done in multiple years working the back room at the National Open.
Others say it should be treated as 100 - 650 - 150 - 100. I don’t particularly like this, but 32C6 does not explicitly disallow it. Tom Doan suggested four points that WOULD disallow it but those points are not currently rules.
I don’t like the idea of a seven-round tournament with the only 6-0 player playing the only 5.5-0.5 player and no 5-1 players, and the 5.5-0.5 player having more of an incentive to draw than to actually beat the limited-prize 6-0 player (with a draw the unlimited 6-1 player would essentially gain all of first plus the better part of second behind the limited 6.5-0.5 player, while with a win the unlimited 6.5-0.5 player would “only” gain all of first).
Again, I ask the question, why should the prizes change at all for the non-limited prize players? They entered at least in theory with no knowledge of whether or not there were any limited prize players. There doesn’t appear to be any direct logic for doing this without making significant additional assumptions, the truth of which are not known. It may be a NICE thing for the organizer to do, but it doesn’t seem like it should be a REQUIRED thing to do.
32C6 requires that if a player winning a large prize is limited to less than the full amount then the balance still needs to be paid out.
Either increase the subsequent prizes or add new prizes, but the money is required to go back into a prize fund in some manner.
In this particular case, if you don’t redistribute to the existing prizes you could simply add 5th-8th at $100 each, or add 5th-11th at 90 - 80 - 70 - 60 - 50 - 30 - 20, or something else. If it was a CCA event then the tournament information generally says that the balance of a limited prize is passed down to the next prize(s), and often other tournaments say the same thing if they have limited prizes.
[Blue coloring mine.]
I’d like to agree with you, at least in principle, but I’m having trouble figuring out your antecedents.
If by this and it you mean offering a limited prize to a player who appears to be ineligible for any prize, then I agree completely – it’s a nice thing to do. However, if the nice thing is advertised in pre-event publicity, then the nice thing becomes required.
OTOH, if by this and it you mean changing the prizes for the non-limited players, then it’s certainly NOT nice, and whether it’s required depends on how the rule is written (or interpreted).
When an unrated player is allowed to play in an Under-nnnn section, there is already a problem, and if the player is declared eligible for prizes (whether overall or class) in that section, there is a further problem. It rips off the other players who are legitimately under nnnn. Prize-excluding that player removes the rip-off. Prize-limiting that player only reduces the rip-off.
So I tend to agree that the best course would be to simply give the non-limited players their full prizes, and give the limited player his limit (provided that his score would have made him eligible) as an extra prize.
Of course, that could get expensive for the organizer, but it might be worth it for the improved PR.
Bill Smythe
Bill, if the organizer doesn’t want the risk of paying out additional monies because of prize-limited players, he can always not allow them in that section, thus avoiding the risk, or not place a limit on their winnings.
Does this make organizing a bit harder and increase the risk to the organizer of taking a financial loss? Certainly. But the rules are already pretty much stacked against the organizer, players appear to have little or no sympathy for the organizer’s risk and God forbid the organizer should actually make a profit!
As I used to tell undergrads at the University of Nebraska-Lincoln business school when I was a grad student there, profit is not a four letter word!
And loss is, as I’m sure you also pointed out to your undergrads.
Bill Smythe
I guess I misstated my point.
I understand what the rule requires. What I’m stating is that its not clear to me that this rule makes sense. WHY is this rule in place, is perhaps the question.
I’m noting the latter, Bill - there seems to be no logical reason/mathematically justifiable reason (at least none that I’ve seen so far) for increasing the prizes to the OTHER players if a limited prize is won by a player. In fact, it creates so many questions that as an organizer one would be inclined NOT to limit a player/prize, which doesn’t necessarily seem to be in the overall best interest of players, US Chess, the potentially limited entry, etc.
Prizes are not won as a group - they are won by participants eligible for each prize. If each eligible participant receives the prize for which they were deemed eligible at entry, then why should their prize change based upon the information or lack of information about another player? The idea that it should appears to be rooted in the concept that the “limited” player is highly likely to be sandbagging. But if that assumption is true, then limiting the player is not a real solution anyway; certainly not without an enticement to that player to self-select to the most appropriate section.
When announced in advance, it seems that it should be fine to limit a player and any excess prize not paid to the limited player should accrue to the house, since each remaining prize winner will be receiving the exact and same amount each should have received for winning the prize that they won had the limited player not existed and had they won the same prize. This is simpler, easier, cleaner, and seems to avoid the issues that Tom Doan raised with his 4 points.
Yes. 32B3 is the Stupidest Rule in the Rulebook. And now 32C6 is a close second, for similar reasons. Both have generated interminable discussions where ultimate consensus seems impossible.
Bill Smythe