Tom,
Your distribution is correct:
The 2100 and 1799 each receive equal shares of 1st place and 1st Class B (B is a higher $$ amount than 2nd place). $65 each
The 1702 and 1599 each receive equal shares of 2nd place and 1st Class C (1st C is a higher $$ amount than 2nd B). $35 each
In addition,
2nd B goes to the next highest scoring B player(s) and,
2nd C goes to the next highest scoring C player(s).
This is the reason why Bill Smythe gave the following very good advice:
Quite a statement there. Certainly it’s a defensible policy and less of a headache perhaps if you’re the type to “stick to the rules”. But, nonetheless, it’s what I would call - wrong.
The whole point of having a rulebook is to increase equitability, not to stifle it. The rules are guidelines; they give us direction and provide a proper spirit for interpretation.
They are adaptable, too. The normal method, of course, is through the Delegates and the Rules Committee. But, I doubt a majority would object to them being “adjusted” ex-tempore if it can be of help.
This same type of situation (re: the prize distribution) occurred in a tournament that I directed with my friend, NTD Steve Dillard. I disagreed with his interpretation of the split, seeing the point-of-view of the forsaken party. I asked Steve if I could call an NTD from the ratings list page, and with his (somewhat exasperated) permission, the NTD (I forget whom) agreed with my interpretation. Namely, that 50% of each prize should move up since both class players were equally eligible to have their class prize move up for the first place prize. Now, the random NTD could have been wrong, of course, but it was thought fair by all concerned.
All things considered, the answer to our problem boils down to how you frame the question. If you say that the tied players get the two highest of the three prizes for which they are eligible which is general rules interpretation, then the B-prize moves up and gets split with 1st place. If, however, you say that 1st prize gets shared between two people from different classes, and both are equally entitled to being considered first, with their corresponding prize moving up, then you are at an impasse.
TW told us to check rule 32B4 for clarification. Let’s see how that applies. First sentence: “A player who is eligible for a place prize and class prize of an identical amount shall receive the place prize.” $32 does not equal $20 and we are comparing two class prizes, so this does not seem to apply, at least in a strict sense. Second sentence: “A player who is eligible for more than one class prize of an identical amount shall receive the prize for the highest class involved.” Again, strict class prizes were used, not ‘under’ prizes so this sentence does not apply, at least in a strict sense. The third and last sentence is: A player who is eligible for prizes of identical amounts, with one being a rating based class prize and the other being a prize for juniors, seniors, etc. shall receive the rating based class prize." Nope, way off target. So rule 32B4 does not apply in any way.
Indeed, I think the rule that most specifically covers this case is Rule 1A: "Scope. …However, the rules of chess cannot and should not regulate all possible situations. In situations not explicitly covered, the tournament director can usually reach a fair decision by considering similar cases and applying principles analogously. The USCF assumes that its tournament directors have the competence, sound judgment, and absolute objectivity needed to arrive at fair and logical solutions to problems not specifically treated by these rules.
Most of you would be hard-pressed, I imagine, to come up with a situation less covered by the rulebook owing largely to the thoroughness of the editors.
The solution I gave which is not to arbitrarily bring either the B-prize nor the C-prize, but to recognize the equal claim of their representatives being considered first. This is done by taking 1/2 of each class prize and adding it to the first place prize. Besides being extremely equitable, it does not hurt that many more players receive a small sum.
Finally, for those that wish to adhere to the rule 32, I think B3 is the extrapolation of choice: “Ties for more than one prize. [The case here.] If winners of different prizes tie with each other, all the cash prizes involved [Here: 1st, B, and C] shall be summed and divided equally among the tied players unless…[followed by an inapplicable proviso]. No more than one cash prize shall go into the pool for each winner.” So only 2 prizes can advance. My proposal 1 + 1/2 + 1/2 = 2 seems to loosely abide within this framework, especially given the fact that none of the situations mentioned in 32B4 applies. Again, I still vote for invoking 1A for doing so, but there is something to be said here as well.
Regards to all,
Ben Bentrup
Well, that’s an opinion not shared by me (that it’s “wrong”). Perhaps the tone came off as harsh, and for that I apologize. I am interested in hearing other’s thoughts, but it seemed to me that the rules were clear here, and player’s opinions or interpretations, while perhaps interesting, are nonetheless not valid.
Isn’t this, though, awarding parts of three class prizes to two tied players? 32B1. “One cash prize per player” is the rule.
I suspect that Steve was correct (he’s seldom wrong). I am almost positive that the other NTD was incorrect!. Perhaps he didn’t understand the situation clearly enough, but if he did, I see no way he could have justified throwing in parts of three prizes for two tied players.
Here, you are using 1A to “create” a rule that violates 32B1. This is not an impasse. The rulebook is clear.
32B4 “… A player who is eligible for more than one class prize of an identical amount shall receive the prize for the highest class involved”. I don’t know how that could be any clearer. Here, the rules say that you bring up the highest class prize involved. In MY case, this would have been B class. This rule is in no way restricted to Under prizes (and perhaps not even implied).
This is, of course, for suituations not clearly defined by the rules.
I think the players might disagree with you there, particularly if the amounts were quite larger.
This is a total liberal dilution to what is given in 32B1, and the last part of what you quoted, namely that “No more than one cash prize shall go into the pool for each winner.” Your “1 + 1/2 + 1/2 = 2” though mathematically correct, is very liberal indeed, and is not what the Rules Committee had in mind, I’m sure.
Okay, maybe 32B1 isn’t the exact rule to quote as “the clear rule” here. I did say, in my other post from which you quoted, that “it was a bit clearer”. It doesn’t speak to tied players, but it should be able to be used analogously according to your 1A rule 
Hey Ben, I think you got it in for me for some reason!!!
Yeah, I sounded a little harsh too - I wasn’t aiming at you, Terry, but rulemongering gets my back up a little; and I like to philosophize. Dangerous, I know.
Again the rule you use for justification of your solution is: “… A player who is eligible for more than one class prize of an identical amount shall receive the prize for the highest class involved”.
Terry, this doesn’t apply. The B-player was eligible for his class prize and a place prize. The C-player was the same way. So no one is eligible for “more than one class prize.” You didn’t use the “Under” category, so C-player is not eligible for the B-prize, and vice-versa.
So then, you too are left with making up a rule. You are basically inserting the words that I put in caps here: “… A player who is eligible for more than one class prize OR COMBINED PLACE/CLASS PRIZE of an identical amount shall receive the prize for the highest class involved”.
So, now the question definitely goes to the discretion of 1A. Your interpretation does have the merit of being quite analogous, but I will still hold that, and it seems that others here have backed me up on this opinion, that my solution was more “fair and logical” [i.e., equitable]. With regards to my solution that it “breaks” 32B1’s limit on cash prizes per player, I would think about first saying “so what? we’re already on uncovered territory” but I think I would also try to stand for my idea that a half of a class prize is not the same as equaling the full class prize, thus barely, squeaking through that rules dilemma.
I anticipate your subsequent reply. 
Ben Bentrup
OK, the rule I quoted b[/b] was not correct.
Let me quote from the 4th edition, and unless the rules have been modified, I assume the same applies. I don’t know why some of the language has changed between the two editions, but I don’t think it’s a major change that overturned the previous rule. Perhaps Tim Just could jump in and help us here.
32B3. (4th edition) Class winners tie with place winners. If winners of class prizes tie with winners of place prizes, all the cash prizes involved shall be summed and divided equally among the tied winners unless the class prize winner(s) would receive more money by winning or dividing only the class prize(s). No more than one cash prize shall go into the pool for each winner.
The only thing this rule is missing, in order to make it clearer, is which class prize to bring up to go with 1st place when all class prizes are of identical amounts. Bill Smythe’s suggestion #2 is the best way around this, as it would then be clear that B would come up since it would be a higher amount.
This is still 32B3 in the 5th editon. The only change is that the language “Class winners tie with place winners” has been taken out of the 5th.
Ben, I’ll have to respectfully disagree with your idea of using 1+ 1/2+ 1/2 idea, while conceding that I quoted the wrong rule. Of course, under your idea, more players would receive money, and we all like to give the most to the most. I concede that it’s more equitable, but, we have to stay within the rules.
I don’t believe that 1A would apply here. Consider the amount of trouble the C players would give you if the prize amounts were larger. (they would probably give you trouble with this piddly amount, too). Your only argument would be that we are “on uncovered territory”. But we aren’t. It’s the way we’ve been doing it for at least my 16 years of being a TD. And, it’s covered under 32B3 (at least in the 4th edition).
I believe the above rule is the most correct. It’s the way I was taught by my TD Kindergarten teachers, Dillard and Anders, back in '87. 
And I think John’s quote above agrees.
What do you think?
Debates like this, about how to combine and split place and class prizes, have been going on since Noah was a boy.
What many don’t realize is that these arguments occur because the rule itself is bad – it contains inconsistencies, or perhaps I should say discontinuities.
Consider the following example:
1st: $150
2nd: $101
Under-1800: $99
Crosstable results:
2100 5.0
1900 4.0
1700 4.5
1650 3.5
Since the 1700 is eligible for both 2nd and Under-1800, he is awarded the higher, leaving the $99 to the next under-1800 player, the 1650. The 1900 gets nothing.
Now let’s change the prize structure by a mere $2:
1st: $150
2nd: $99
Under-1800: $101
Now the 1700 still gets $101, but the $99 now goes to to the 1900, with the 1650 getting nothing.
A change of just $2 in the prize structure has resulted in a $99 difference to each of two players!
Of course, the discontinuity occurs precisely at the crossover point:
1st: $150
2nd: $100
Under-1800: $100
– which is exactly where the arguments begin, including those that started this conversation.
What is REALLY needed here is a rule change. Something like the following:
“If a player is eligible for two prizes, he receives the higher of the two, with the remaining prize being split among the players next in line for each of the two prizes, in proportion to the two original prize amounts.”
For example, in the crossover case:
1st: $150
2nd: $100
Under-1800: $100
– the 1700 wins $100, with the remaining $100 split evenly (because the two prizes are identical) between the 1900 and the 1650.
In the first example:
1st: $150
2nd: $101
Under-1800: $99
– the 1700 wins $101, with the remaining $99 split not-quite-evenly (because the two prizes are not quite identical) between the 1900 and the 1650). The 1900 would get 101/200 of $99, while the 1650 would get 99/200 of $99. This would amount to about $50 and $49, respectively.
Likewise, in the second example:
1st: $150
2nd: $99
Under-1800: $101
– the 1700 would win $101, with the remaining $99 being split at about $49 for the 1900 and $50 for the 1650.
No discontinuities here. Small changes in the prize structure result in small changes in the final awards, and everything seems more equitable.
Bill Smythe
I actually typed a nice reply for Terry last night but my daughter needed putting to bed before I finished and I fell asleep. The biggest idea was that I was impressed that Terry could actually step out of the Bible for once to philosophize on a rule. Granted he took an earlier version for scriptures to back up his ruling, but I thought that he had been hard and fast on the idea that whatever he did had to come from the current rulebook. His logic seemed sound, and although I no longer have retained a 4th ed. for my reference, I accept his point.
I still hold that the philosphy of the rules is to search for the most equitable ruling, in fact this is quoted in the preface to the 5th, and he has conceded that my take was more equitable, but I have to accept he has the more analogous rule to what is current and past.
But, Bill, you genious! You hit the nail on the head.
You showed us we were arguing ‘toMAYto’ vs. ‘toMAHto’. I have to admit that while I barely read a single detail of your post, I grasped the main idea in a few seconds. You may make us fetch our calculators, but its sound advice if we can ever get that worded correctly into the rulebook.
The WinTD/SwissSys and other programmers will hopefully see an opportunity in this thread to write some code to automate the prize-giving process even more. Once the prize information is entered, the program tells you exactly who wins what amount.
Ben Bentrup
Thank you for your kind words.
I had hoped that my post would generate a lot of discussion – maybe even heated arguments and lots of examples where my scheme allegedly wouldn’t work – but the result has been, instead, dead silence. Maybe everybody is still eating turkey, or maybe the idea is SO good that it simply ends discussion (I very much doubt it, however).
I guess what the rule-makers have overlooked, for at least the 36 years I have been a USCF member, is that just because a player is entitled to $101, he may not be entitled to THE $101. His prize could as well be forged from parts of two others, as long as the total is $101.
The suggestion in my post was:
But now I have an idea for an improved version:
“If a player is eligible for two or more prizes, he receives an amount equal to the higher, divided proportionally among the prizes for which he is eligible.”
This wording (I think) is mathematically equivalent to my original idea, but may be easier to understand, especially if followed by an example such as the following:
Let’s say a player is eligible for both of the following prizes:
2nd $150
Under-1800 $100
He is eligible for $150, which is 3/5 of the total $250. So he would receive 3/5 of the $150, plus 3/5 of the $100, i.e. $90 of the 2nd place prize plus $60 of the under-1800 prize. This would leave $60 in the 2nd place pot and $40 in the under-1800 pot, which would go to the players next in line for each of these prizes.
NOW let’s have your criticisms, counterexamples, complaints, logic, emotions, etc. By rights this should be a lively discussion.
And remember, we haven’t even TOUCHED another related question, namely what to do if TWO players are tied for 2nd and under-1800, and only one of them is rated under 1800.
Bill Smythe
I am making one more desperate attempt to revive this conversation.
Please re-read my post of November 22, and the few replies that followed. I had hoped that this post would become a conversation-starter, not a conversation-ender!
Thank you, all.
Bill Smythe
Senior TD Bill Smythe:
It has been a nice debate with the members of this forum, with the after problem with class and under prize. It has to be a issue with the director when building the prize award before the tournament. The under prize award or the class prize award works best if and only if the tournament has an aceptable amount of entries; it does not work best if the entries are small in the first place, for example: if the tournament only has less then 20 entries.
Theory, only one person in you’re tournaments’ has a Class E rating: nobody has ever gone to you’re tournaments with this level of rating except this one person. Forming a prize award for a class prize, as a Class E prize, with the past empirical evidence only to state of only one person would be looked on as a bias from the ‘organizer/director’ for building such award. The player with the past history will be thought on as the possible winner of such a award even before the start of the event.
My particular problem with following this thread (which I suspect is everyone else’s) is that I’m too lazy and math-challenged to articulate arguments or even to understand what it is that you have proposed. Shameful, since I was once also a Northwestern engineer.
When I apply what I think is your method to the original problem, I get results that are in line with Ben Bentrup’s method, that is moving half of each class prize up so that the second place class B gets a little money. Ben’s method appealed to me because it there seemed to be a certain arbitrariness to saying that a $20 Class B prize is worth more than a $20 Class C prize, therefore the second place class B gets nothing even though he tied for the second highest overall score in the tournament. Perhaps “arbitrary” and “discontinuity” are analogous terms in legal and mathematical injustice?
It seems simple enough in your example, but this number seems to be the one which is hard enough to find under the current method. You almost seem to be advocating some sort of recursive method, analogous to the current rating system method for unrateds.
Under what I believe is your method, the U1800 is “eligible for” half of 2nd or $75 and half of U1800 or $50 which totals $125, which turns out to be half of the involved prizes. Therefore he gets the same as what he should have gotten under the ($150 + $100)/2 method. However, the tied player ABOVE 1800 will only get the $75 remaining from 2nd. Some third player U1800 will get the $50 remaining Under-1800 prize.
I don’t really object to what you propose except that it gets a little complicated with fractional prizes being left over and summed up with other fractional prizes. If I have totally missed your point and misapplied your system, then perhaps that furthers my point that it may be too complicated for us simple chess folk to understand.
I just realized that I have insulted everyone’s intelligence. If that doesn’t get 'em riled up Bill, I don’t know what will. 
Ernest:
It is very clear you have not insulted my intelligence; its’ so clear for everyone in the universe, as there is ‘no’ proof or evidence of any intelligence on Earth. With the American election of 2004, its’ clear to other life form as proof and self evidence of no intelligence. Even if there is intelligence in the universe, and they do come to Earth and study us: like those UFO’s stories. It just gives proof that they have way too much free time to study humanity; with the members of these crews that study this planet, as clear proof they do not have a life.
Yes, my method would come up with the same result as Ben’s, in that particular case. In fact, my proposal was intended to extend Ben’s method to cases where the two prizes are not identical, in a fair, logical, and continuous manner.
It’s not recursive, it’s just proportional – unless certain complicated cases could become recursive (let’s hope not!).
If a player is eligible for both a $150 and a $100 prize, then he gets $150, taken from both prizes proportionally:
Total of the two prizes: $250.
Player is eligible for: $150.
This is 150 / 250 of the total, or 3/5. (Granted, I picked an example where the fractions can be reduced to much simpler terms.) Therefore, the player gets 3/5 of each prize, and the remaining 2/5 of each prize goes to the players next in line for each of these two prizes:
3/5 of $150 is $90. 3/5 of $100 is $60. $90 plus $60 equals $150 (of course). This $150 goes to the player in question.
2/5 of $150 is $60. This $60 goes to the player next in line for the $150 prize.
2/5 of $100 is $40. This $40 goes to the player next in line for the $100 prize.
I am hoping that more complicated cases won’t really be more complicated to calculate, but I’m waiting for possible counterexamples (and hoping for additional helpful remarks) from the thoughtful among you (Terry, Ernest, Ben, rfeditor, thunderchicken, WildTommy, etc).
As for the other related question (two players tied for two prizes, one of them ineligible for one of the prizes) – well, let’s put that one on hold for a while.
Ah, yes. The age-old argument of “let’s do it a simple way, even though it’s wrong, just so that everyone can easily understand it.” That argument is used frequently against the recent improvements in the rating system, too.
Well, thanks for the effort. We’ll see what happens!
Bill Smythe
just want to make a quick reply. I haven’t investigated all that you propose, and will have to go back and read your post some more. My initial thoughts were that your examples weren’t relevant to my original post because you’re using “under” prizes and my case involved “class” prizes. I don’t think the two are analogous, but I may be wrong. The second issue I had with your proposal was that I was searching for a clarification to the existing rules regarding “class” prizes. I realize the rules may be changed to be more fair, and I wouldn’t object to that. I was just interested in clarifying existing rules.
Bill, are you proposing that the rules are ambigious as currently written? And if so, are TDs responsible for deciding their own policies based on their own objective opinions?
Your post isn’t a discussion ender, I just need more time to delve into it.
Please reread my original case and tell me if I’m wrong in my thinking (particularly the back and forths between Ben and me)
Doesn’t matter. In the example I used, both of the under-1800 players were also class B players – neither was under 1600. The logic works equally well either way.
Our approaches are different. When a rule is inherently illogical, unfair, and discontinuous, I tend to become less interested in how it should be applied in a specific case, and more interested in discussing how it should be changed.
I suppose that, when a rule is illogical, unfair, and discontinous, it could also be regarded as ambiguous. If applying the rule literally can lead to absurd results, and to feelings of “surely they didn’t mean for me to do THAT”, then there exists, at the very least, an ambiguity between the letter of the rule and common sense. Having to choose between the two can put a TD in an extremely uncomfortable position, to say the least.
And that’s why a discussion of a desirable rule change is FAR more interesting to me than a discussion of how a flawed rule should be applied in a specific case.
Oh, all right. Eventually. Meanwhile, you can do me the same favor in reverse, by examining my proposal in the light of how things can be improved, rather than just how things are at present.
Bill Smythe
Okay, here’s my “eventually” (it didn’t take as long as I feared it might):
As a literal interpretation of the flawed rule in the case you cited, your method (and that suggested by Wild Tommy) seems absolutely correct, and for the exact reasons you stated in your November 16 post. A literal interpretation is the safest way for a TD to go, as the TD can blame the rulebook for any inequity, and cover himself better in the event of an appeal.
As far as common sense is concerned, I like Ben’s idea. But it doesn’t seem supported by any current rule. My suggested change incorporates Ben’s idea, and extends it to cases where the prizes are non-identical, thus removing the discontinuity in the current rule.
Okay, NOW you can comment on my November 22 post, and those following. Bon appetit!
Bill Smythe
In an open tournament, seven players tie for first prize which is $700. One of the seven is additionally eligible for the U2000 prize which is $99. We’ll call this person the dual prize winner. So then this dual prize winner is “eligible for” both a $100 share of first prize and a $99 prize, then he gets $100 taken from both prizes proportionally. 100/199 of the $100 first prize is $50.25 and 100/199 of the $99 U2000 prize is $49.75. $50.25+49.75 equals $100. The 1st prize has $649.75 left and the U2000 prize has $49.25 left. Do the six remaining first place players get $649.75/6 = $108.29? The dual winner cries foul. How come he gets $8.29 less than the other six when his points are the same? Does the “eligible for” $100 assumption need to be changed? The old way of prize distribution would have $799 split 7 ways with each of the seven tied players getting $114.14. Would you then start with the dual prize winner being “eligible for” $114.14? Then 114.14/799 * 700 = $100 and 114.14/799 * 99 = $14.14. So the dual prize winner gets his $114.14. The remaining six first place finishers get $100. The second place U2000 gets $84.86.
Hang on here a minute! You gotta tell me how the 1900 is eligible for any money at all. The 1700 gets 2nd place (he outscored the 1900, and 2nd place should be regarded -IMO- as more prestigious than an Under/Class prize). The only remaining prize is the u1800, and why would a 1900 get part of that?
Same comments as above! How is the 1900 eligible for any money?
Here, since the prize fund is screwed up by making an u1800 prize more valuable than a place prize, the 1900 is eligible for 2nd place because he outscored the 1650. In this case, how is it that 1650 is eligible for an award (the u1800 is gone, and the 1900 clearly wins 2nd place).
I can’t see where I’m wrong here, but I await your guidance.
Sorry for the long delay, and thanks for re-reading my earlier posts.
The whole point of the “limit one prize per player” rule is to spread the prizes over more players, so that some players not otherwise eligible for a prize might win some money after all.
In this case, one player is eligible for two prizes, so part(s) of one or both prizes are awarded instead to the next player(s) in line.
One way awards 2nd place to a player who was not really 2nd. Another way awards 1st under-1800 to a player who was not really 1st under-1800. For some reason you are objecting to the former, but not to the latter. Still a third way (my proposal) would do both, but to a lesser extent in each case.
Personally, I don’t see how it is any worse to award the 1900 a prize he didn’t “really” win, than it is to do the same for the 1650.
Again, you seem to want to go by the letter of a flawed rule, rather than to stand back and look at the big picture. Your post does not address the problem of the discontinuity at all.
Bill Smythe