New “Auto” number of matches for Swiss Tournaments

[BananaHarambe]

https://boardgamearena.com/tournament?id=405733 This 2p tournament with 61 players just ended with 7 rounds instead of 6 rounds.

It is frustrating to include that extra round for no reason because all that did was make 4 players tie with the same score instead of making clear 1 winner. For 2p Swiss, 1 extra round is a bad configuration IMO.

If Auto Swiss creates such an odd number of rounds that kind of defeats the purpose of it. I am not sure I can recommend it for 2p.

[yodamus]

Hopefully whoever worked on the auto Swiss program will see this thread and correct it. Seems like a possible easy fix. Is there a way to directly contact that person?

[Mathew5000]

It is frustrating to include that extra round for no reason because all that did was make 4 players tie with the same score instead of making clear 1 winner. For 2p Swiss, 1 extra round is a bad configuration IMO.

For any number of players, not just 2p. For example, the link I gave earlier in this thread of a 3p tournament where the "extra round" led to a tie in the standings which necessitated one further round: https://boardgamearena.com/tournament?id=417177

[Meeplelowda]

which necessitated one further round

This makes it sound like it did the further round because of the tie, but we know that's not the case because the tournament BananaHarambe mentioned above ended in a tie (as in the number of points before tie breakers) even though at the end of the previous round there was a single player with the most points, and it didn't go on to another round to break the tie it created. It seems like this extra round in the 3p tournament was just going to happen regardless.

[sillynamenace]

I had this Swiss-Auto tournament with 9 entrants (3p matches): https://boardgamearena.com/tournament?id=413295

It finished in 3 rounds. This didn't seem fair to me.

But for me the real issue was with the points attribution. I used the +6, +5, +3, +1, 0, ... system, thinking that the players would get +3 / +1 / 0 points at the end of each game. But no, it was rather like +6 / +5 / +3 (no 0 pt or +1 pt attribution). Is this normal? Is this how this attribution system normally works? I doubt if it is so, I suspect it had something to do with the Auto system. Any opinion?

[BananaHarambe]

I had this Swiss-Auto tournament with 9 entrants (3p matches): https://boardgamearena.com/tournament?id=413295

It finished in 3 rounds. This didn't seem fair to me.

But for me the real issue was with the points attribution. I used the +6, +5, +3, +1, 0, ... system, thinking that the players would get +3 / +1 / 0 points at the end of each game. But no, it was rather like +6 / +5 / +3 (no 0 pt or +1 pt attribution). Is this normal? Is this how this attribution system normally works? I doubt if it is so, I suspect it had something to do with the Auto system. Any opinion?

Ending in 3 rounds is perfectly in agreement with everyone else. It calculates 2 rounds to sort all players, then added 1 extra round like it always does, none of us are sure why.

I am not sure about that points system since I always use the other one with 4-2-0 scoring for 3p. Someone else might be able to explain better, but I think that is just how it works.

I don’t think it really matters when only 2 of the games completed properly. I see 7/9 games abandoned with 2/1/0 scores so I would be more concerned with getting more games to complete. The tournament winner was likely whoever was clicking their turns the fastest or whoever went to sleep the latest.

[Meeplelowda]

It finished in 3 rounds. This didn't seem fair to me.

You don't say why it doesn't seem fair to you. After just the second round there was already an undefeated player. Swiss is not meant to be a slow version of a round robin.

[sillynamenace]

It finished in 3 rounds. This didn't seem fair to me.

You don't say why it doesn't seem fair to you. After just the second round there was already an undefeated player. Swiss is not meant to be a slow version of a round robin.

It was just a feeling I got, especially knowing that the games didn't finish in time. But the BGA algorithm surely can't take that into account. I'm not gonna be assertive about what I said there, I get your point.

The real issue for me is what I ask afterwards, though.

[Mathew5000]

I used the +6, +5, +3, +1, 0, ... system, thinking that the players would get +3 / +1 / 0 points at the end of each game. But no, it was rather like +6 / +5 / +3 (no 0 pt or +1 pt attribution). Is this normal? Is this how this attribution system normally works? I doubt if it is so, I suspect it had something to do with the Auto system. Any opinion?

I think it has to do with the positioning of the ellipsis. Notice that the two options for "Points attribution based on results", when you set up a tournament, are:

(a) ..., +6, +4, +2, 0

(b) +6, +5, +3, +1, 0, 0, 0...

I take that to mean that in option (a), last place gets 0; second-last gets +2, etc. For example, in this 5p tournament using option (a), the winner of each game gets +8: https://boardgamearena.com/tournament?id=417628 The ellipsis comes before the +6, meaning that if there are more players, each player receives 2 more points than the one finishing immediately below.

In option (b), the ellipsis is placed at the end. I take that to mean that regardless of the number of players, the winner gets +6, the player finishing second gets +5, and so forth.

I see no reason to think that the points attribution is affected by setting the number of matches to "Auto".

[sillynamenace]

I think it has to do with the positioning of the ellipsis. Notice that the two options for "Points attribution based on results", when you set up a tournament, are:

(a) ..., +6, +4, +2, 0

(b) +6, +5, +3, +1, 0, 0, 0...

I take that to mean that in option (a), last place gets 0; second-last gets +2, etc. For example, in this 5p tournament using option (a), the winner of each game gets +8: https://boardgamearena.com/tournament?id=417628 The ellipsis comes before the +6, meaning that if there are more players, each player receives 2 more points than the one finishing immediately below.

In option (b), the ellipsis is placed at the end. I take that to mean that regardless of the number of players, the winner gets +6, the player finishing second gets +5, and so forth.

I see no reason to think that the points attribution is affected by setting the number of matches to "Auto".

Oh this is very informative, thank you a lot.