An interesting philosophical question in its own right. Other than organizer/TD reporting error, when should a result be changed?
Isn’t any file directly out of TMS easiest?
No, we don’t. The organizer is free to create his own tiebreak system, although this needs to be announced and/or posted. The Massachusetts Chess Association, for example, uses a non-standard tiebreak system for its scholastic team tournaments. I’m strongly opposed to forcing organizers to use a standardized tiebreak system.
There are at least two important differences between offering crosstables in score/rating order and allowing rooks to move diagonally. For one thing, the first is reasonable and the second is ridiculous. For another, in the first we would be depriving Jane of something she has always had, while in the second Jane has never had it. To say the least, Kevin, your logic is peccable – and all on the minor sub-issue of “ramming”, at that.
Our main disagreement is whether MSA should compute the sort order and tiebreaks (apparently your preference) or should simply accept the sort order and tiebreaks furnished by the pairing software (my preference).
The pitfalls of the former have been well documented in this thread and elsewhere. But let’s look at a likely scenario – a scenario which, I’m sure, will become anything but hypothetical on many occasions if MSA is allowed to compute sort order and tiebreaks:
Jane (now an organizer) is running a scholastic tournament where there are zillions of trophies. She carefully announces in all pre-event publicity that the sort order and tiebreaks will be those recommended by USCF. She sets her tournament software to produce the USCF-recommended sort order and tiebreaks. At the end of the tournament, she awards trophies in the manner determined by her tournament software.
After the ceremony, she uploads her crosstable to MSA, but MSA has been brilliantly designed not to include the option to accept sort order and tiebreaks from the tournament software, so instead she chooses the option to have MSA compute the sort order and tiebreaks. But wait – there is a problem! Jane has already given the 4th-place under-1000 6th-grade trophy to Johnny, following the dictates of her tournament software, but MSA says the trophy should have gone to Fred. Somehow there is a discrepancy in the two sort orders, caused perhaps by a subtle difference in the tiebreak computation. Maybe cross-section pairings were handled differently, or house players, or whatever. Fred’s parents, now at home, look at MSA and become livid, calling Jane, also now at home, to demand that Johnny, also now at home, give up his ill-gotten trophy to Fred.
As should be obvious, we would not be doing organizers or players any favor by having MSA compute sort order and tiebreaks.
Bill Smythe
I don’t see why this would be the case.
If there is one error in the report, even though it might affect hundreds of final ranks and tiebreak points, all the organizer would have to do is correct the one error in his tournament management software. The TMS would then recalculate the sort order and all the tiebreaks, etc. This revised report from the TMS would then become the correction required by the office.
The corrected report from the organizer’s TMS would be handled by the “approved designate”, rather than through MSA as it would be if it were an original report.
Bill Smythe
As stated, an opinion, not an argument, so not convincing. Additionally, I could replace the example with one you find more palatable-writing moves first for example.
There is nothing wrong with the logic. You’re making the error of focusing on attributes of an illustration that you dot like. It’s an example, not intended to be all encompassing, simply illustrating the point. Just pick another example-the points till holds.
If that is what you think, then you missed Several points of my post. I suggest you re-read.
The rest is not currently relevant.
No, Kevin, I have absolutely no desire to discuss further the minor concept of ramming stuff down people’s throats, nor to suffer through any similar examples of your “logic” in other areas such as move-before-writing.
Let’s continue instead with the topic of where the tie-breaks (and sort order) should come from, MSA or the organizer. Upon re-reading your posts in this thread (as you suggested), I’ll admit I find very little that would give me a clue one way or the other as to which you prefer. About all I found was:
– which suggests to me that maybe you want everybody to use the same tiebreaks so that MSA can do the job itself.
So, let’s get our cards on the table. Which do you prefer?
- MSA should compute the tiebreaks, using the same (standardized) list of tiebreak systems for all events, OR
- MSA should allow the organizer to supply the tiebreaks, via the rating report (oops, tournament report) submitted to MSA by the organizer.
After I know where you stand, we can continue from there.
Bill Smythe
Bill, I’m sorry that you didn’t get it-but sarcasm such as “logic” is inappropriate. Your argument was faulty, I demonstrated that it was faulty, sarcasm is unnecessary.
I don’t have a big preference because the preference is largely meaningless. It’s all based on the same data. So results should be the same.
I’m ignoring non-TMS run tournaments in that statement. Including “by hand” calculations tilt toward using organizer reports because of the increased possibility of organizer error.
There is an argument that if tiebreaks are used in prize distributions, that TMS must be used.
Since all tiebreaks should follow the same methodology irrespective of where the calculation is performed, why does your question matter?
There have been a number of proposals where information would have to be entered in TD/A on a player-by-player basis. My previous statement quoted above includes potential ranges for manual changes that may be brought about by a modified result entered into MSA by the national office. Specifically, if prizes or tiebreaks have been manually entered, either by the director in TD/A or by the national office in the event of a mailed rating report, those changes would have to be manually entered as well.
Again, this does not address the cases where prize/tiebreak information would first be entered manually, and then changed later. If the initial information was manually entered by the TD, then the changes would likely have to be manually entered as well. This leads back to the question of who will do the manual entry. Again, I strongly suspect that the national office - which is already thin on resources - will not want any part of a system where that work will fall upon them. If this, or any other, proposal relies on their willingness to do this kind of manual data entry, then said proposal is likely DOA.
Whether one thinks the manual-entry use cases are significant is irrelevant (though I certainly believe they are). Their mere existence requires the time of the administrators and developers, and must be addressed in any subsequent development effort. They are a big part - but only one such part - of the reason why the cost-benefit analysis for all these proposals seems, to me at least, to be upside-down.
All tournaments do not (and should not) use the same tiebreaks.
Your demonstration that my argument was faulty is faulty.
Should be, in theory. Also in theory, practice and theory are the same, but in practice, they’re not (Yogi Berra?). It’s just a matter of time before a discrepancy shows up between MSA and TMS. The rulebook explanations of the various tiebreak systems are complicated (perhaps not even entirely clear in certain borderline situations), and the person who writes the MSA algorithm probably won’t be the same as the person(s) who write the TMS algorithms. Murphy’s law is bound to rear its ugly head eventually.
I would to hate to prohibit a TD without on-site TMS from running a trophy tournament, if that’s what you’re saying.
As I said: Theory. Practice.
Bill Smythe
Has anyone calculated the cost for making all of the changes to the MSA to show tiebreak order in one particular way over another? What will be the costs to reprogram WinTD, SwissSys, and other pairing software concerning tiebreaks? If all of the changes were made, from a hypothetical point of view, who would they serve?
As an organizer, I do not use the standard tiebreaks for trophy tournaments. The present standard ordering is fine for a 7 to 12 round tournament. In fact, the tiebreaks were first created for such tournaments. In smaller events with only 4 rounds, I prefer to ignore median and use Solkoff, head to head, cumulative and coin toss. The people on site are the only ones interested in the trophy order. I will show them the numbers and calculations, if necessary, but nobody asks. Afterwards, when they look at the MSA, the only thing they care about is the rating change. The way ratings are presently shown post tournament are easy enough for players and parents to find. I have yet to have anyone ask why the rating report on MSA does not show trophy order. Three months or three years down the line, who is really going to care who got the 2nd U600 trophy in a particular tournament? Should the programming of the MSA be turned inside out to show this result?
The office should never be required, or expected, to enter tiebreak or prize data manually. Initially, extra fields could be provided on the MSA screen which organizers use when they submit a rating report without using TMS (tournament management software, like SwisSys or WinTD). I assume that, when the organizer does use TMS, this same MSA screen eventually becomes available so that the organizer can enter corrections manually. (If this last statement is not true, then whoops, maybe we have another monkey wrench.) It is at this point that the organizer could enter the tiebreaks and prizes manually, player by player.
That could be Stage One of the software development process. No changes would be necessary, yet, in the TMS. An organizer who submits a report by snail-mail would simply not have the option to enter tiebreaks or prizes at all, nor to expect that his tournament report will ever be available to viewers in any format other than the “standard” score-then-rating sort order.
At Stage Two, the TMS and MSA software will have both been enhanced so that tiebreak and prize data from the TMS can be shot straight through to MSA. It would still be true (and would always be true) that organizers submitting snail-mail reports would never have any options regarding tiebreaks and prizes.
Yes, by the organizer – the same person who entered the report to begin with, and who is now asking for corrections.
Bill Smythe
Wouldn’t it just be much easier to allow for a link to an organizer’s site where the organizer can format up a report however he wants (and change it if need be)? For instance:
estima.com/chess/cochess/000325.htm
where I included the team codes and grades (which would be of interest to many of those viewing) and manually added the prize field.
EDIT: An excellent point raised via PM: Even this relatively labor-cheap suggestion still adds serious complexity to both the implementation and post-implementation planning. The national office would need to review and approve all submitted links prior to actually linking them.
That wasn’t asserted.
Not necessarily, no, although it may be a great interim step.
Yes, we do. There are standardized tie-break systems.
The fact that SOME OF IT is not standard doesn’t change that. The question remains whether there should be greater forced standardization. If there are good additional systems, there isn’t a reason NOT to standardize those.
I swear some people just try to be argumentative.
In what universe would that not be easier? It’s one URL which wouldn’t even necessarily have to be part of the upload spec. Most organizers of significant tournaments already create a formatted tournament report and would probably be quite happy to have a link to their site out of the MSA.
The alternative is the vaporware upload spec that you and Bill Smythe are having a food fight about.
Bill, that has already been demonstrated to be incorrect. It’s observable.
And yet TMS has already implemented standardized methods.
Organizers and TDs should use established theory and practice.
In a universe where we have comprehensive end-to-end TMS and member data management.
As I said, your point is well taken and is a great interim and current step.
That’s not a reason to avoid striving for a standardized, comprehensive, cost-saving, time saving, centralized system in the longer term.
MSA has a UNIQUE advantage. It is a centralized, permanent, record. Organizer websites cannot match that that feature in the foreseeable future- and hence MSA should be the base for development- not organizer sites.