Your #1 is not allowable as per rule 32B3. Either the U1700 rated players take the U1700 prize themselves, and leave all the place prizes to be split by the over 1700 rated players, or the U1700 prize goes into the pot to be split by all the players in the tie. The key word in the first sentence of 32B3 (cited previously) is “only”.
Interesting prize fund 10th-13th are $600, $590, $580, U1800=$900 U1700=$875, $825 with two 1800+, 2 1700-1799, and 3 U1700s would have a total of $4370 awarded to the seven players
2x$600, 5x$634
2x$595, 5x$636
3U1700x$823.33 (1/3 each of U1700 money for 2/3 of a prize, 1/5 each of U1800 money for 1/5 of a prize, 2/45 each of open money for 2/15 of a prize totaling 1 prize) - leaving $360 of U1800 money and $1540 of open money, 2 1700-1799x$565 (1/5 each of U1800 money for 1/5 of a prize, 1/4 each of remaining open money totaling 19/20 of a prize), with 2 1800+ each getting $385
The three-way split of U1700 is $566.67 each. The five-way split of U1800 is $180 each. The seven-way split of open $252.86 each. The one prize per player limitation increases the top four’s money by limiting the share that went to the U1700s.
You are correct per the rule published in the rulebook in 2014.
You may be overlooking the the change to the rule that occurred after that publication and which is referenced on the website. Such changes override the rules as printed and are supposed to be enforced.
uschess.org/docs/gov/reports … hanges.pdf
“Rule 32B3. Ties for more than one prize.
If winners of different prizes tie with each other, all the cash prizes involved shall be summed
and divided equally among the tied winners unless any of the winners would receive more
money by winning or dividing only a particular prize for which others in the tie are ineligible. No player may receive an amount greater from the division of those prizes than the largest
prize for which he would be eligible if there were no tie. No more than one cash prize shall
go into the pool for each winner. For examples see 32B5, Offering a choice of prizes:”
OK. Now you have me confused. What in rule 32B3 that you cite contradicts what I am saying?
This was a Bob Messenger-amended suggestion of mine from a similar discussion in 2013. This gets rid of that unfortunate one-at-a-time rather than group-at-a-time language from the current rule:
The “No player may receive an amount greater from the division of those prizes than the largest prize for which he would be eligible if there were no tie” isn’t a operative description of how prizes should be divided (it’s a limit, not an algorithm). It’s simply a check against gross mistakes—if you do the distribution correctly, you wouldn’t have that situation. And, that one-at-a-time language creates more confusion than it resolves (see much of this thread).
This seems like a good suggestion for alternate wording, and would help to clarify the prize distribution here. I wonder why it didn’t make it into the current rules? I hope someone will author an ADM next year to get this language into the rulebook.
I think that, with an absolutely literal reading of rule 32B3, Mr. Parker is correct in limiting Alan, Barbara, and Charles to $600 each. That is to say, literally considering each of those players in isolation, it is true that if there were no other players with the same score, each would be eligible for 10th place ($600). I think that I wrote this earlier in this thread. However, I believe this distribution violates rule 32B1 (which I will discuss below).
Mr. Relyea is correct that the change to rule 32B3 did not accomplish what the delegates intended, or at least what this delegate intended. (Mr. Immitt and I submitted two separate ADMs in 2013 (ADMs 13-39 and 13-40) proposing the same change to rule 32B3. I cannot speak for Mr. Immitt’s intention.) To me, it seems unreasonable that adding players into a tie should increase the winnings for any subset of the players in the tie compared to what that subset collectively would have won without the other tied players being in the score group. Let me be perfectly clear: I am not saying that, taken literally, rule 32B3 prevents awarding each of Alan, Barbara, and Charles $600. It just doesn’t make sense to me that those three players collectively win $1800 when they would collectively win $1770 if all of Donna, Edward, and Francine had scored even a half point less.
In fact, I continue to argue that awarding each of Alan, Barbara, and Charles $600 violates rule 32B1 insofar as there is no possible way to pay $1800 using just the sum of 10th, 11th, and 12th place prize money (which totals $1770). Those three players may not be paid any of the Under 1700 prize money, so paying them requires dipping into (presumably) the 13th place prize money. Paying out more than the sum of three overall place prizes to three players appears to be a case of paying out more than one cash prize per player. (As I pointed out earlier, rule 32B1 is itself flawed insofar as we regularly (and correctly) pay out more than one prize to tied players when some of the prizes involved are “under” prizes. That was the point of my 1st $100, top U2000 $50 example with an expert and an A player tied for top score.)
Sigh … I suppose I’ll be submitting YAA (Yet Another ADM) to fix the fix to 32B3.
Is that not $900 for the U1700 and $600 for the 1700+ players?
- Enrique
Out of curiosity I plugged this into both WinTD and SwissSys to see how they would interpret it.
SwissSys uses the $600 limit version and distributes like this:
3 x $600 (1700+) and 3 x $666.67 (U1700)
WinTD splits the U1700 and 1700+ players into two groups and distributes as such:
3 x $590 (1700+) and 3 x $676.67 (U1700)
Since both seem to be easily defensible by the current rules, we clearly need better language in the rules. But, if new wording is added, then which one is correct?
Needless to say, the WinTD calculation matches the description in viewtopic.php?p=311884#p311884 (which is what I’ve always understood the correct process to be). The $600 each for the 1700+'s is IMO simply wrong—no player or players should make more money as a result of other players being added to the tie.
While I understand Tom’s sentiment, I’m not 100% certain that I agree with it. Here is why. (This requires stepping back to a slightly larger discussion. Sometimes when a can of worms is opened, its worth stepping back to see if the best way to get the worms back into the can is to consider using a larger can.)
As I recall - looking back into the dark haze of my past chess career - tournaments did not always have class prizes. The idea of adding a class prize was to entice more people to enter even though they knew that they were not likely to win an overall prize - because they could take pride in and recover some of their costs by winning a class prize. Class prizes were, at an earlier time often viewed as supplements to overall prizes. They were beneath but ancillary to overall prizes
Then the situation would occur where ties would spread among the top players - for example experts and masters - and the top A player.
So we might have this:
1st $100 2nd $75 3rd $50 Top A $40
Suppose that 1st was taken by a 5-0 Master, 2nd by a 4.5 Master, and 3rd-4th by 2 experts and an “A” player at 4.0. The prize distribution I would see was $50+$40/3 = $30 each.
And I do recall seeing cases where perhaps the A player scored 3.5, won the A prize outright, and people thought it was “wrong” that the A player received $40, while the two experts each received $25. The situation could be more pronounced if the top “class or under” (for now I’ll use “class” to mean both) prize exceeds the lowest overall prize.
Note that if the 2 experts did better - scoring 4.5 - the A player also did better prize-wise - and the Master joined at 4.5 did worse prize-wise! This is also a problem here - why should a player do worse when their score is unchanged, and another player do better when their score is unchanged? Note that the relative position of the “A” player doesn’t improve, but his/her prize increases. The relative position of the Master doesn’t worsen, but his/her prize decreases.
This would not happen if the “Top A” prize was “4th”. In the originally prize scoring the A player would tie for 3rd-5th and receive $30. In the revised scoring where the two experts score an extra half, 5th place is out in the cold. The 3 players receive $165/3=$55 - a distribution that makes more sense based on the relative positions.
So, irrespective of what the current rules are - I DO THINK there is a reasonable perspective and argument for treating class prizes as “subservient to” overall prizes - that perspective is that merit should always come before playing level - and those assumptions come with their own set of issues. There are implicit issues to class prizes.
Another potential way to view class prizes is to view them as “separate from” the top overall prizes - in the same way one might view an upset prize or perhaps a “top junior” or team prize. That is, to treat them solely unto their group, separate from the group of overall individual prizes.
In the case at hand:
One might argue under such a methodology that Donna, Edward and Francine would first receive the Under 1700 prize/3 = $300 each, and that then ALL six players would also share in 10th -15th: $3,450/6= $575 each.
So the players above 1700 would receive $575 and the players below $875.
The “1 prize per player” rule has not been broken because each PRIZE GROUP (overall and class) is considered completely separately, and the rule is applied within each group, not across groups.
I realize that both of the above perspectives might currently be a bit less traditional, but its not clear to me that the perspectives are less valid - its just that the underlying assumptions are different.
So now lets consider the situation where Overall and Class Prizes may be pooled, the original problem:
I view the two answers (as outlined by SwissSys and WinTD):
SwissSys uses the $600 limit version and distributes like this:
3 x $600 (1700+) and 3 x $666.67 (U1700)
WinTD splits the U1700 and 1700+ players into two groups and distributes as such:
3 x $590 (1700+) and 3 x $676.67 (U1700)
Where NEITHER is problematic. The issue is what is the additional underlying assumption about hierarchy? In the first case, class prizes are considered somewhat hierarchically less than overall prizes - in the second case they are not.
To me - both answers are justifiable. It’s just that the underlying assumptions are different.
Simple question, deserving of a simple answer. If A-C scored 1/4 point more than D-F, how much would they win? If it’s not splitting $600+$590+$580, then you’re proposing a completely different method of allocating prizes.
Simple answer: In one method I wrote about above, it is $600+$590+$580, and in a second method I wrote about it is different. As I said up front in proposing additional methods, I was stepping back and looking at the larger problem, and fully admit that SOME of what I said was “…a completely different method of allocating prizes.” I was looking not only at the question of what the rules currently say, but also “what did they use to say?” “what do we want them to say?” “are there alternatives that make sense that might simplify this?”
The answers when stepping back might not be as useful as we would like, but sometimes it produces pleasant results.
Of course Mr. Bachler is correct here. This is why many have said to make all place prizes larger than any class prize. I’d venture to suggest that in Mr. Bachler’s “dark haze” tournaments didn’t pay 50, or even 15, places, either. This is a big part of where the problem lies.
Alex Relyea
OK. Then I saved time by not reading what you wrote.
Not sure how.
A friend of mine (a Life Master) has argued ever since I have known him that all class prizes should be eliminated, that there should be only place prizes. I have grave doubts that model would fly in the US, but it would have some advantages. You would pretty much eliminate sandbagging, and the kind of prize distribution problems like this one that we agonize over would be a thing of the past. If I ever get rich enough to be able to afford to take the financial beating that I think would come my way I may try it out.
To give anything more than a share of $600+$590+$580 to A-C (if they scored higher than D-F) would require fishing into a lower score group and glomming onto money for which they are ineligible. We have a hard enough time writing a description of what should be a relatively straightforward top-to-bottom calculation. “Look-ahead” calculations would require overriding the judgment of the tournament organizer regarding an appropriate prize fund.
They are “ineligible” only by definition. Definitions can be changed. Its a rule, nothing more.
I’m not suggesting a “look ahead” calculation or over-riding the judgement of the tournament organizer. I’m suggesting that a better understanding of the PRINCIPLES at work would likely simplify the currently difficult description. I’m suggesting that part of the reason for the difficulties are that people have different views and perspectives of how this SHOULD work, so when it gets defined, they interpret the definition in different ways.
If instead we look at how its being interpreted and recognize that there are some differences in interpretation, it should become a matter of creating a few rules for which principle is being followed, and then creating a few settings to match - even if we choose not to allow multiple methods of prize distribution, it will eliminate questions because we will have eliminated the grayness of some definitions.
It’s possible that the entire discussion I raised was unnecessary - it appears that some posters were “far enough along” in the thought process that it may not have been. But based on the responses, it doesn’t appear that everyone was on that same page.
I hope you’re missing a smilie face.