Unexpected elimination patterns in multiplayer elim tournaments

[Bartissimo]

I set up the following tournament as a 3-player elimination some time ago: https://boardgamearena.com/tournament?id=435606 . It's a tournament with only 3-player games, with 2 players qualifying for the the next round. I had expected that regular elimination pattern to run right up till the final three-player table. Instead, when 5 players remained in the tournament, a 3-player table was created, 2 players were skipped and auto-progressed as normal– and surprisingly, two players were eliminated from the 3-player table, leaving only one player from that table to progress. Those three players are now playing a game that is the final match, with places 1/2/3 in the tourney determined as you would expect, simply by the placement in this game. At least, that is what I hope - otherwise the system would throw another major surprise

Is it clear to anyone when and why the system should break the standard pattern of eliminating only one player in a set-up where 2 players should progress, as it did in the penultimate game? I ask this also because there are a few more multiplayer elimination tourneys running. The previous game in the tournament I linked to looks like an anomaly and I wonder how players can know upfront what the elimination approach is in the games leading up to the final.

[FrankJones]

My suggestion would be to require the tournament to begin with a player pool that will avoid having the number of players reduce to a non-multiple of 3.

Which might be easier if it were just one player advancing. With two players advancing, everyone will be playing to avoid last rather than playing to win the game

[Azoken]

My suggestion would be to require the tournament to begin with a player pool that will avoid having the number of players reduce to a non-multiple of 3.

Which might be easier if it were just one player advancing. With two players advancing, everyone will be playing to avoid last rather than playing to win the game

I tried, and it's not possible. First game can be a multiple of 3 but if you keep 2/3 you'll eventually grant byes.

there are not so many interactions in TFM that not trying to loose is a viable option, but that's only my opinion.

i think the case mentionned by OP is kind of an exception as if you keep 2 players and mix with the other 2, you still have to do two more rounds with almost the same people, but i wouldn't have expected that behavior either.

[FrankJones]

there are not so many interactions in TFM that not trying to loose is a viable option, but that's only my opinion.

Removal of plants, microbes, or animals. Reduction of [production of "any player". Map blocking. Draft decisions. Making plays that ensure the player in 1at place gets TR bonuses rather then the player in 3rd place. decisions with regard to awards, milestones.

I would say there are quite a few ways in which a player in 2nd place could make plays that elevate 1st place while further burying the person in 3rd place.

I haven't played in a TFM tournament, but I would think tournaments drag on long enough as is, and having only 1 person from each 3-player game advance would serve several useful purposes.

[Azoken]

And the very same could be said for king making. At least if you're last you can't make the same excuses as if you're 2nd.

In a tournament, if the game does not end, the winner is kinda random. If you keep two, only the guy who took the longest time is eliminated, which is very important.

[FrankJones]

Oh, true, I see what you mean. If one player rarely moves, and the other two move almost immediately, one of those two players would be eliminated if only one player advanced, even if the thinking time for the three players was something like 2 hours, 2 hours one minute, and 7 days.

[Azoken]

Oh, true, I see what you mean. If one player rarely moves, and the other two move almost immediately, one of those two players would be eliminated if only one player advanced, even if the thinking time for the three players was something like 2 hours, 2 hours one minute, and 7 days.

here is a recent real life exemple. Due to favorable timezone, one player had 45h left in bank 42h before the end of the round. He had the game lost on board (his last move would be placing a forest tile). So he decided not to play. For 42h all his other game would advance but this one. And he won the round thanks to the clock. Here, at least, only the player who took the most time is punished by this behavior. The player who would have won the game sill advance in the tournament.

[FrankJones]

Oh, true, I see what you mean. If one player rarely moves, and the other two move almost immediately, one of those two players would be eliminated if only one player advanced, even if the thinking time for the three players was something like 2 hours, 2 hours one minute, and 7 days.

here is a recent real life exemple. Due to favorable timezone, one player had 45h left in bank 42h before the end of the round. He had the game lost on board (his last move would be placing a forest tile). So he decided not to play. For 42h all his other game would advance but this one. And he won the round thanks to the clock. Here, at least, only the player who took the most time is punished by this behavior. The player who would have won the game sill advance in the tournament.

That's a good example. I don't think 2/3 players advancing is ideal, but as your example shows, under the current tournament time rules, 1/3 advancing would be far worse.

In the end, it all comes back to the same problem we have in non-tournament games: There is often no good way to determine a single winner in a 3+ player game when one person forfeits or quits.

[xspuipuke]

Sounds to me like the problem is the elimination style tournament. A simple swiss system one with a fixed number of games might be the better option for 3p.