WER server apparently crashed today (and is still down?). Fired up this app and it worked like a charm. Real lifesaver. Thanks again!

This works pretty good, just used it yesterday.
I should suggest a manual.pairing option, it was after first round when i noticed TO running the tournament manually.
Thanks a lot

Edited Emmanuel Leal (March 31, 2014 02:15:43 AM)

Good to hear!

Manual pairings would be nice, but for now I won't include them, until I absolutely need it and I figure out a nice way to do it (from a usability standpoint).

Johannes,

Thanks for this great app!

Last week I tried to run an FNM in both, WER and your app, just to be able to compare the tournament outcomes. Everything went great, but there was a small discrepancy in tiebreaker percentages counted by the applications. You should probably doublecheck the formula you use for computing those (or does WER deviate from the formulae described in MTR?).

From my experience, it's the later. Both DCIR and WER don't follow the MTR tiebreaker formulas.

Wow! And no one does not seem to care about it?

I think this may be the percentage assigned to people with low points. The MTR says 33% which WER follows but you may be using 1/3 or 33.3333%.

Thanks for the work dude!

I'm definitely going to use this instead of doing the pairings by hand at non-sanctioned events run at LANs :)

So, I investigated the tiebreaker issue a bit. My findings:

1) WER completely ignores the “0,33” rule (!). In my test tournament, one player had the final GW percentage of 28,5714.

2) TiebreakerJS follows the MTR - it uses value “0,33” instead.

Originally posted by Milan Majercik:

1) WER completely ignores the “0,33” rule (!). In my test tournament, one player had the final GW percentage of 28,5714.

33% rule is only for Opponents percentages. Others tiebreakers can be lower

Edited Amand Dosimont (March 31, 2014 08:56:49 AM)

Negative!

The 0.33 shall be used everytime (and WER does not use it when computing ANY percentage).

Originally posted by MTR:

Match-win percentage
A player’s match-win percentage is that player’s accumulated match points divided by the total match points possible in those rounds (generally, 3 times the number of rounds played). If this number is lower than 0.33, use 0.33 instead. The minimum match-win percentage of 0.33 limits the effect low performances have when calculating and comparing opponents’ match-win percentage.

Game-win percentage
Similar to the match-win percentage, a player’s game-win percentage is the total number of game points he or she earned divided by the total game points possible (generally, 3 times the number of games played). Again, use 0.33 if the actual game-win percentage is lower than that.

You are quoting wrong paragraph.

This is important for Game Win %:
Game-win percentage:
Similar to the match-win percentage, a player's game-win percentage is the
total number of game points he or she
earned divided by X times the number of games played.

The 0.33 corrective is applied when calculating OMW% and OGW% (1st and 3rd
tiebreaker) in order to limit the effect of low performances when
calculating and comparing opponents' percentages,but there is really no
reason to apply it on GW%. It would be equivalent to saying that player's
match points cannot be lower than a third of maximum.

Actually WER uses 0.33 corrective exactly as written in MTR, even though it
generally calculates everything to 4 decimal places. That's a reason why
you can have very little differences in OMW% even in tournaments with very
small amount of rounds.


2014-03-31 16:04 GMT+02:00 Milan Majercik <

Thanks David!

Where does that quotation come from? I can't find it in MTR (not even in the old ones, like 2 years old).


I computed manually the percentages for both cases (0.33 used and 0.33 not used). And the result was very exact: WER used GW percentage 28,5714 when computing OGW.

Edited Milan Majerčík (March 31, 2014 09:43:39 AM)

The quote is from 5-years-old MTR. I agree with you, that current MTR says
that 0.33 corrective applies even to GW%, but by using logic and common
interpretation rules I came into realization that it's wrong and most
probably just an error, rather than philosophy problem.

Here's my reasoning (feel free to disagree):
1. GW% is a tie-breaker. It should help to break ties. By saying that
everything between 0 and 0.33 is equal, you effectively diminish its
tie-breaking function. OGW% and OMW% use the corrective because the
affected player has no possible way to influence who his or her opponents
will be. In that case, the corrective should ensure a little bit more fair
results. In case of GW% there is no need to do that, because GW% is 100%
determined by the player's own gameplay.

2. In OGW% section of MTR, there is a sentence “And, as with opponents'
match-win percentage, each opponent has a minimum game-win percentage of
0.33.” If the 0.33 is really intended to be applied to the GW%, there would
be no need for this sentence, because if no GW% can be lower than 0.33,
then OGW% (which is average of them) cannot be lower than 0.33 either.

3. WER still doesn't apply it. If this change was done on purpuse, I would
expect it to be for some important reason which would mean that it would be
implemented in WER (if it's not really stupidly programmed, just a few
lines of code would have to be deleted).

I don't know why it has been changed. I spent last hour searching for the
version of MTR in which it has been changed, so I could look into its
appendix A and consider the reasoning behind that change, but I couldn't
find it. Personally I think, that someone just wanted to make GW% look the
same way as MW% (which used the corrective even 5 years ago) and didn't
realize, that even though MW% is never used as a tie-breaker and therefore
doesn't matter at all, GW% is a tie-breaker by itself.

I would be glad if the person who made the change (or at least understands
it) can explain it. And if it really it a mistake, I think that MTR should
be updated, so that diligent judges like Milan, who try to undestand
everything, did not get it wrong (or at least inconsistently with reality,
which is determined by WER).

TLDR; You are right that MTR says it, but MTR or reality are wrong :)


2014-03-31 16:27 GMT+02:00 Milan Majercik <

I will look into this over the next few days, and provide the sources.