Hugh wrote:  In Swiss, the system above should hand the tiebreak in a fair way to the person who played the better opponents in the tournament.

Isn't that pretty much what is happening?  Win later against higher ranking players (players who have won games) is better than winning later against lower ranking players (players who have lost games).

Raptor wrote: So here is how you game the system: Set your account so that you play more than one tournament game at a time. Look at the games that you get set up in. Play the noobs as fast as you can and take 1 day and 23 hour turns against top 5 players.

This is a concern of mine, or a possible strategy, but again, I think it seems to be the same as in Ranked games.  Also if you play the noobs fast then play the top players you probably have a higher chance of losing against the top players which would result in losing more points than hurrying up and losing to them and then beating the noobs.

*edit* 4 posts in a row!  I win!

Yertle wrote: *edit* 4 posts in a row!  I win!

I can beat you there!

But first things first. It seems like my idea got lost along the too many lines I wrote before, so I'll reapeat it here:

if two players end up tied at the end of a Round Robin tournament, then the player who won the "direct match", that is the game against the other tied player, win the tie breaker. If the tournament is set up to have games with more than two players, than the game where the tied players are is considered and if no one of them won that game, then the current system or another one (see next) is used.

Also I could propose to use the score used in chess tournaments: the Buchholz system. And for more information this page could help as well (though I guess you already visited it)

(=

Tesctassa II wrote:

But first things first. It seems like my idea got lost along the too many lines I wrote before, so I'll reapeat it here:

Yertle wrote:

That said, I do think the idea of Ties goes to the winner between heads-up play (and score if no heads-up play) has merit.

Yertle wrote:

Hugh wrote:  In Swiss, the system above should hand the tiebreak in a fair way to the person who played the better opponents in the tournament.

Isn't that pretty much what is happening?  Win later against higher ranking players (players who have won games) is better than winning later against lower ranking players (players who have lost games).

Mostly, I think.   It heads down a straightforward path, because new rounds don't start till an old round finishes.  I haven't analyzed fully the situation where the championship game is reached with some of the players having the maximum number of points, and others with one less point.  All is fine when the max pts people win, but if not, you get a massive tie.  I believe the players who won earlier will have the higher score on account of losing to the higher rated players in the championship (who would have lost earlier to a lower rated player).  Also, those players entering at a max number of pts could end in a tie for score.  Might be worth working out some examples to see how that goes down.

Yertle wrote:

Tesctassa II wrote:

But first things first. It seems like my idea got lost along the too many lines I wrote before, so I'll reapeat it here:

Yertle wrote:

That said, I do think the idea of Ties goes to the winner between heads-up play (and score if no heads-up play) has merit.

I meant to say that I agreed with Tect/Yert on that point :)  I don't actually believe the rationale behind it is sound,  but all kinds of sports use that and they are happy with it.  Also, it is a time-independent tiebreak system.  However, imagine a round robin 3-way tie where the 3 leaders beat each of the others once - from a performance standard, I can't make an argument for why any of the 3 should win.  It is truly tied (though our system currently will pick a winner).   What then?  B/C it is round robin, even the mentioned Bucholz system fails to pick a winner.

Yertle wrote:

Hugh wrote: (With normal ranked games, this bias is negligible b/c people keep playing, but tournaments are finite and small differences determine tournament winners.)

I'm not so sure it is really all that different between Tournaments and Ranked games.  I would much rather lose early in CP hunting, when I'm at 1000 score, than when I'm at 1400.  I'd also much rather win against someone with a 1400 score when I'm at 1400 than beat someone with a 900 score. 

For the most part, I would say that Ranked games are finite, we all know that there is a "ceiling" that we strive for, the 20 CPs at 1500 which, for the most part, indicates a finite system (the being passed and losing CPs does create more of an un-finite system, but a lot of maps the top player will be the top player for a while).

The bias I refer to is time among equal performers.  Two 1000 ranked players go 1-1 against each other.  It is better to lose and then win than it is to win and then lose.  The difference is very small (1001 vs 999 in fact), but in a tournament, this small difference matters.

Yeah, ranked games are finite in a sense, but in principle your 1500 score isn't safe.  This already plays out on popular boards.  I'm convinced that under ideal circumstances (huge user base, all boards getting lots of play by skilled players), such biases get smoothed out.  My main point about the finiteness is that in a tournament, if two players go 6-1, whatever biases the calculations have impacts who wins the tournament. Contrast that with ranked play: the player who has the lower score can just go and start a bunch of new games.

Yertle wrote:

Tesctassa II wrote: But first things first. It seems like my idea got lost along the too many lines I wrote before, so I'll reapeat it here:

Yertle wrote: That said, I do think the idea of Ties goes to the winner between heads-up play (and score if no heads-up play) has merit.

Sry! But that's what happen when you write four post in a row! Never do that again! XD (ok I'm wrong, just trying to save some face)

Seriously, here's something that should convince Hugh about the actual system:

"the rationale (for system that weigth more early wins), is that a player who scored well early in the tournament has most likely faced tougher opponents in later rounds and should therefore be favored over a player who scored poorly in the start before subsequently scoring points against weaker opponents" (Just & Burg 2003:200-201)

This statement holds for swiss system tournaments of course. But then again there are some solution:

• one could chose to evaluate "direct-match" if any, since in a swiss tournament it's not obvious that top players play agains each others

 or, using a more general criteria,

• use a score like the Buchholz, which evaluates the strength of each player's opponents (contextually to the tournament) summing opponents' scores,
• use a score which evaluate the performance of a player based on the starting rank on that board. That is, at the end of the tournament evaluate that player A played like a player with a X ranking score

Another good idea should be to count the booted players one has encountered in his/her games, as it is obvious that playing a 4-players game were two get booted is far more easy than a game where all players partecipate actively. Not to mention 1-vs-1 games; for such games the win could (should) even be halved.

I guess this last suggestion should be taken in account seriously, since here swiss tournament last 3 rounds, rarely 4, which is a small number of rounds to farily establish a winner, in my opinion.

(=

Hugh wrote:

I meant to say that I agreed with Tect/Yert on that point :)  I don't actually believe the rationale behind it is sound,  but all kinds of sports use that and they are happy with it.  Also, it is a time-independent tiebreak system.  However, imagine a round robin 3-way tie where the 3 leaders beat each of the others once - from a performance standard, I can't make an argument for why any of the 3 should win.  It is truly tied (though our system currently will pick a winner).   What then?  B/C it is round robin, even the mentioned Bucholz system fails to pick a winner.

I think it is valid, as it is supposed that if two top players, or teams, end up tied it means they are the strongest in the tournament, and hence, the one that beat the other is stronger. After all it's unprobable that, of the two tied players, one lost against strong opponents but won against weak opponents, while the other one obtained the opposite results.

But if you're concerned with this possible situation, then we should start to consider more the ranking score one has on a board, and use that as a reference point instead of the tournament score alone.

What do you think?

(=

I'll throw in my 2 cents I guess. While CiscoKid knows who won he's right in that under the current tournament configuration he won. It sucks for me but in the end it has brought about this discussion which will hopefully bring about a fix to what I think is a broken scoring system.

For regular non-tournament games the scoring system obviously works great. For tournament games, however, it as has been stated above is broken. By setting the Multiple Tournament Rounds in your settings to play all the rounds at once and then you could pretty much pick and choose who you play early and who you take your time with. As Hugh said you could play the people you know/think are going to lose a lot early and play the better players later.

The problem I think lies in that each round the score follows you and it really shouldn't. I think each round should start with the base score and then each round tallied seperately is your overall score.  This of course wouldn't fix the problem of having the same win/loss record, but it would even out the fairness of playing early or late games.

The second part of this is that potentially if you don't play all your games at once your place in the points is randomly determined by what games and in what order you happen to play them in. Had I had played one round in a different order in the Arm Wrestle tournament things may have come out quite differently. My problem with this is the order you play the games in should not be the determining factor for a Round Robin match.

I also think that in a situation such as in this tournament that the winner of the heads up match should be the winner or if the functionality is there it should kickoff a Tiebreak match automatically.

You guys probably have more on this than I could come up with..

I stick with my two ideas:

for RR tournaments buchholz and "direct-match" should be the criteria, since everyone plays everyone once, there shouldn't be the problems that there are now.

for Swiss tournaments actual systems works fine, since all games in a round start simultaneously, and players are paired considering their win/loss score, and hence the tournament score follows them accordingly. Maybe one could use buchholz in Swiss tournaments as well, or maybe using the board ranking.

And again I think booting situation should be reconsidered in computing the score

(=

Tesctassa II wrote:

Seriously, here's something that should convince Hugh about the actual system:

"the rationale (for system that weigth more early wins), is that a player who scored well early in the tournament has most likely faced tougher opponents in later rounds and should therefore be favored over a player who scored poorly in the start before subsequently scoring points against weaker opponents" (Just & Burg 2003:200-201)

This statement holds for swiss system tournaments of course.

This is very convincing in a Swiss system tournament because it is very nearly a bracketed tournament:  players who won must play each other, then players who won two games must play each other, etc.  If you lose early, you got to play the people who lost round 1.  This quote was used to justify the "cumulative sum" tiebreaker, which I think our system might be mimicking automatically.  I haven't worked it out whether that is true or not.  I like this tiebreaker for Swiss, but round robin tournaments are far from being bracket-like.  This logic fails dramatically in the round robin case since every player will have played all the strong and all the weak players (in fact, all the players!)

Tesctassa II wrote: I stick with my two ideas:

for RR tournaments buchholz and "direct-match" should be the criteria, since everyone plays everyone once, there shouldn't be the problems that there are now.

Bucholz fails to produce a winner in a round robin tournament:  The sum of the opponents' scores will be equal for the leaders.   "Direct match" (aka "head-to-head") will produce a winner in a two-way tie, but not always in a 3-way (or more-way) tie in round robin tournaments.

Ok, I didn't read everything here but I agree that I don't like the current tie breaking system. Here is my suggestion:
Add 1 more game with the top X players based on the number of players for the game. If it is a 4 player tourny than the top 4 play 1 more game and the winner is the winner of the tournament. Yes, I understand that if there are no ties than this system is stoopid but I don't know if ties can be coded to do this so I think if it is just what it is all should be good. You win by playing and winning not on a formula and I think that is the best solution anyways.
Yes, this is the anti-BCS and pro playoffs argument here.

Hugh wrote:

Tesctassa II wrote: I stick with my two ideas:

for RR tournaments buchholz and "direct-match" should be the criteria, since everyone plays everyone once, there shouldn't be the problems that there are now.

Bucholz fails to produce a winner in a round robin tournament:  The sum of the opponents' scores will be equal for the leaders.   "Direct match" (aka "head-to-head") will produce a winner in a two-way tie, but not always in a 3-way (or more-way) tie in round robin tournaments.

You're right! I didn't see that. And thanks for telling me the correct way to say "head-to-head" (=

About RR tournaments, I think we could use a modified version of the Buchholz which count only the score of the opponents the player won against, or maybe those who lost to.

Or maybe board ranking score could be used instead of the win/loss score. That score is harly the same for three players.

Also it could be an option to consider (I know I'm stressing this point a lot, but it looks important to me) the booting situation: the player who encountered less "booted players" in his game win.

I can't come up with more ideas for now, but always remember one thing: finding a criteria to determine the winner at any cost, isn't always the best way. If two players played equally well even after comparing two or three different scores, then they both deserve to win or, if it is required that only one win, than drawing lots becomes the solution. In our case I think an ex-equo placement is more correct. I don't think it'll be common anyway.

(=

I'm not a fan of the boots playing a role in a winner. It's not the player's fault that someone else in a game was booted. Not to mention they could have been booted when they only had 1 troop left and were pretty much all but eliminated.

RiskyBack wrote: Ok, I didn't read everything here but I agree that I don't like the current tie breaking system. Here is my suggestion:
Add 1 more game with the top X players based on the number of players for the game. If it is a 4 player tourny than the top 4 play 1 more game and the winner is the winner of the tournament. Yes, I understand that if there are no ties than this system is stoopid but I don't know if ties can be coded to do this so I think if it is just what it is all should be good. You win by playing and winning not on a formula and I think that is the best solution anyways.
Yes, this is the anti-BCS and pro playoffs argument here.

About the new tournament type, here's a thread I opened some time ago. Apparently it's been planned to add some new type of tournament.

;)

Yertle wrote: I'm not a fan of the boots playing a role in a winner. It's not the player's fault that someone else in a game was booted. Not to mention they could have been booted when they only had 1 troop left and were pretty much all but eliminated.

I agree with you about the last case you mentioned. But about the first one I don't. I know it's not a player's fault, but nevertheless whoever is gonna fight less opponents has an easier task.

But since I can't come up with an easy solution for the first case, I guess you're right. It's not possible to apply this method.

(=

From what I read to break a tie for Round-Robin:

First - Use the head-to-head comparison. If still tied then ...
Second - Declare a tie.

I think if we insist in "no-ties" then either some external factor has to be used (e.g. most eliminated- which would not work for 2 player games btw), and any of these could still result in a tie, or have a rematch or rematches until a clear winner was found. This paragraph sounds like no fun to me.

I thought of having an automatic' heads-up elimination between the tied players... the problem as ever with tourney math is making sure it's always conclusive - e.g. what if there is a 3-way tie, how do you have a 3 way elimination battle on a 2 player board?

Whatever solution is come up with it has to guarantee a tie-breaker result in all possible situations.

I don't think tied tournaments are the answer.

So how comes does my presence always cause a disturbance in the worlds of RISK?