Question re Dividing Prizes (Rule 32B3)

Running Chess Tournaments
1

An interesting division of prizes arose in a tournament recently and the way the new rule was applied is different to what I would have done so I’d like to pick a few brains in here.

Prizes: 1st $12000, 2nd $6000, 3rd $3000, 4th $1500, 5th $1000, 6th $800, 1st U1700 $2000, 2nd U1700 $1000.

Player A, 1753, 8.5 pts
Player B, 1702, 7.5 pts
Player C, 1700, 7.5 pts
Player D, 1700, 7.5 pts
Player E, 1750, 7.0 pts
Player F, 1676, 7.0 pts
Player G, 1675, 7.0 pts
Player H, 1650, 7.0 pts

Clearly Player A takes 1st ($12000) and Players B-D take 2nd-4th ($3500), the question is the remaining prizes.

My opinion: Since just awarding the Under 1700 prizes to the 3 Under 1700 players would only yield $1000, all the prizes are added into the pool and divided by 4 ($1200 each). However, I would have said that the largest prize Player E is eligible for is the 5th place prize of $1000 and so I would have added that extra $200 to the remaining Players F, G and H, giving them $1266.66 each.

Actual Decision: The way it was distributed was that the largest prize Player E was eligible for was a split of 5th/6th since he shared that prize (as there was only 2 U1700 prizes) and therefore that player is awarded $900 and the other players were awarded a share of 5th/6th and the 1st and 2nd U1700 prizes giving them $1300 each.

Any thoughts on the which way is the correct way to interpret this rule?

2

I think the reasoning expounded in the “Actual Decision” paragraph is faulty. I believe the correct reasoning was that advocated by Mr. Bird. In the absence of any other players with 7.0 points, the largest prize to which player E would have been entitled was the overall fifth place prize of $1000. I believe it was incorrect to consider the other players with 7.0 points when determining the largest prize to which player E was entitled. The penultimate sentence of rule 32B3 includes the phrase “if there were no tie,” so I believe evenly dividing the fifth and sixth place prizes is a misapplication of that sentence. I thus believe the correct distribution would have been $1000 to player E and $1266.67 to each of players F, G, and H.

This question is remarkably similar to the question posted four years ago by Mr. Immitt that led to the penultimate sentence of rule 32B3 being added in 2014.

3

Yep.

4

Yep.

5

Yes, but the wording of the rule could certainly be improved.

“winning or dividing only a particular prize”. In this case, it’s actually prizes, not prize.

“for which others in the tie are ineligible”. I recall that this specific language caused at least some problems before with those who want to be confused. The one player is literally “eligible” for both the 5th and 6th prizes. Of course if he takes 5th solo he can’t take 6th, which is what is intended. While the language added does resolve some difficulties, the way the rule actually should work is that you check to see if the group of U1700 players (or any other subgroup in the tie) does better by letting the other player(s) get their money first. That avoids loaded words like “eligible”. It’s what the other group would win if they took their share first.

6

The critical phrase is “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.”

If a player is eligible for both 5th and 6th (and is not eligible for any of the U1700 prizes) then the LARGEST prize the player is eligible for is 5th, and that is the maximum amount the player could win if various prizes are pooled together in a tie. The wording has prize instead of prizes, so only the single largest eligible prize is checked to determine the maximum.

Interesting distribution
5th=$1000, 6th=$900, 7th=$800, 1stU1700=$2000, 2ndU1700=$1000. Five way tie with two 1700+ and three U1700s.
3-way split of the two U1700 prizes is $1000 each
5-way split of all five prizes is $1140 each, so the three U1700 players do better by pooling all prizes.
The $1000 cap means the two 1700+ players are limited to $1000 each, leaving $1233.33 for each of the three U1700 players.

Note that this is different from giving the two 1700+ players $950 each for splitting 5th and 6th (the top two prizes they are eligible for). The rule says each is limited to the largest prize they are eligible for, and they are both eligible for the $1000 5th prize.

The way Tom said the rule should work makes sense, but does not follow the exact wording. It is better than the original wording ($1140 each for all five is better for the U1700s than $1000 each for the three U1700s, so originally $1140 would have technically been acceptable). That wording was better than the even older wording where a well-scoring expert that tied with multiple GMs for the final place prize ended up seeing the top expert and bottom place prize lumped together and split (he ended up getting less than he would have if he’d scored a half point less and been able to take the top expert prize all by himself).

One difficulty with getting more and more exact on prize distribution rules is that eventually a prize fund can be devised that doesn’t do what the rules-changers would like to see it do.

7

I don’t see that as the “critical phrase”. It’s more of a check to avoid making a major error. As you show in your example, it is unlikely to help (and, in fact, might be a bit confusing) when each subgroup within the tie has at least two players. A better phrasing would be that “No subgroup within the tie may receive an amount greater from the division of those prizes than the prize they would receive if there were no other players in the tie” (or something like that). Thus the over 1700’s would be limited to splitting $1000+$900, leaving the rest for the U1700’s.

8

Mr Doan seems to be changing the meaning of the rule. It was intended to, in this case, limit the 1700s to $1000, the amount they would get if there was no tie. It is very clear. If they wanted to do as he suggests, they would have phrased it differently. “No ties” is vert specific. It means none, not none outside of your class.

9

Absolutely, positively not. The 1700’s are limited to a max of $1000 each, but that’s irrelevant, because they actually at most can split $1900. The one-player-at-a-time wording in the rule is pointless (at best) and misleading (at worst) when there are multiple players in the group with the more limited prize options. As Jeff points out, one could (I believe incorrectly) interpret that stray sentence to mean that the 1700’s get $1000 each, which gives them more than they (collectively) could get if the U1700’s weren’t in the tie.

10

As currently worded, the 1700+ players can split the sum of the five prizes that the five players split but are limited to $1000. The two 1700+ players are not sharing more than two prizes. They are sharing one fifth each of five prizes with a limit on what they can get. As the 32B3 rule was previously worded they could take part in the five way split without a limit (getting $1140 each).

The author of the current change would likely have agreed with you. Players that are not rules lawyers would probably accept your interpretation as quite reasonable.

11

I’m confused. Are you saying that the 1700’s winning $1000 each is correct? That sounds like a rules lawyer interpretation. If the 1700’s had 0.1 points more than the U1700’s, they would unquestionably split $1900. How could they make less money by scoring more points?

12

As I said, I’m guessing the author of the current change would agree with what you want to do. The current change came about because under the previous rule the 1700s would have won $1140 each, which is more than either could have individually won if there was no tie. The change capped the winnings at the maximum possible ($1000) if they were the sole person contending for the prizes. The wording quite definitely allowed for winning more money in a tie than could be won if the other players had not scored high enough to tie (the case that sparked the current change was a tournament where the higher rated player agreed to a winnings cap even though it was not part of the rules). The change did not address capping players based on a combination of prizes, but only based on the single, maximum, prize that could have been won if there was no tie.

Also, I did say that players that are not “rules lawyers” would probably accept your interpretation. For that matter, many players would accept your interpretation even if they realize that the letter of the rules could give them a larger share of the prize fund in the event they are in one of those unusual ties.

Somebody who is bound and determined to maximum his winnings could point to the current wording as technically saying both 1700s each have an individual $1000 cap, not a combined $1900 cap.

13

Is there anybody who believes that the “correct” distribution gives the 1700’s each $1000? It seems like the current language is an improvement, but still doesn’t really explain things properly.

14

One of the problems I had with getting the language to explain things properly is the “at most one prize per player” rule. Of course, proper prize distribution violates this rule all the time. For example, suppose the prizes for a small four round tournament are 1st place $100, top under 2200 $50. Suppose a master and an expert tie with 3.5/4. The correct prize distribution is for each of the two players to receive $75. However, there is no way around the interpretation that the expert is walking away with the full $50 under 2200 prize plus one-quarter of the $100 1st place prize. (However, it is true that at most one prize per player went into the distribution.)

In the present example, it is clear there is no way to pay both of the 1700s $1000 each without paying them some of the under 1700 prize money (to which they are not entitled).

15

The total U1700 money is $3000 in the revised example ($2000 for 1st U1700 and $1000 for 2nd U1700) and the three U1700 players receaved $1233.33 each ($3700 total). The total open prize fund (5th-7th) was $2700 and the two 1700+ would have gotten $1000 each.

16

How about this wording for the second sentence: “No player or players may receive an amount greater from the division of those prizes than the largest prize(s) for which they would be eligible if they were the only players winning or sharing the prize(s).”

17

Maybe it’s necessary to include (as a TD tip or directly in the rule), that if N players equally divide N prizes that one can not and should not attempt to come up with a breakdown of individual shares of the prizes. Here, the master apparently shares in the U2200 money so the expert can properly share in the higher overall prize. That’s how we (US Chess) have decided to divide up the prizes in such situations: they’re jointly entitled to the $150, and the U2200 can’t do better by letting the master take first, therefore you divide evenly. It might be possible to come up with a different systematic way to distribute this which would give more to the U2200, but that wasn’t the choice that we made.