Cactus Game Design Message Boards

I might be able to see your side of things as well if players A,B,C were the only top players in the tournament. Their were at least 8 players that could have won the tournament. If your only loss comes from a Maly, Alstad, Earley, Brinkman or Wester you did loss to a top player. Player A played more than one game against the top players because of the way Swiss style works, especially when you play one more than the required rounds.

I also see your point about there being top players who don't end up making the rankings at the end.  But unless we had some sort of ranking system OUTSIDE of individual tournaments, there's no way to quantify that.  Sure I agree that all the guys you mentioned are top players whether they tank a specific tournament or not, but when devising a system for ranking an individual tournament we have to ignore that kind of prejudice (even if it's accurate).

Imagine a tournament where you have 5 players tied for the lead at the end:
A - went 1-0 against the other 4
B - went 3-1 against the other 4
C - went 2-2 against the other 4
D - went 1-3 against the other 4
E - went 0-1 against the other 4
Who deserves to win?  Sure A has a higher win % against top opponents, but they only played 1 game against those others.  Player B won 3 games against the other people placing.  I think everyone would agree that Player B had the best tournament.  The only way I see of making this work is the proposal that I've posted on this and a couple other threads.

If they all had equal points then tht means the other players lost to people outside of top 5 players

Yes, but in my scenario, both A and B lost to people outside of the top 5 (perhaps even the same person). So that doesn't help anything.

That would require all your top players to have lost 2 games an that doesn't even happen at Nats the tournament would have to be huge.

Gabe won Nats 2007 after going 0-2 in the first two rounds...and how would that require all players to have lost 2 games? Perhaps A and B both lost to person X, A in round one, B in round two, then they went on to win all of the rest of their games. It's certainly possible.

There is no system that will be perfect, so I think the best solution is the simplest solution:

For two players tied at the top, the first tiebreaker is head-to-head. If they never played (or tied when they did) then the next tiebreaker is LS differential. If LS differential is equal, then the tourney ends in a tie (or an extra round is played to determine the winner).

For three or more players tied at the top, if any player defeated all of the others, then that player is first (by head-to-head vs. all of them). If there is no such player, then it goes to strictly LS differential for placement.

The scenario I was referring to is when A was 1-0 and B was 2-0.

Quote from: Prof Underwood on July 24, 2012, 03:00:24 PM

Imagine a tournament where you have 5 players tied for the lead at the end:
A - went 1-0 against the other 4
B - went 3-1 against the other 4
Who deserves to win?  Sure A has a higher win % against top opponents, but they only played 1 game against those others.  Player B won 3 games against the other people placing.  I think everyone would agree that Player B had the best tournament.

But what if B was 1-1?

I think it is still clear that B had the better tournament.  He beat 1 top opponent and lost to another.  That is better than beating 1 top opponent and losing to someone who didn't even play well enough to rank at the top (like Player A did).

Or what if A was 1-0 and B was 2-0, but A never got a chance to play the second person B beat because it was randomly chosen in the final round that A would play F (someone who didn't get in the top 5) and B would play E?

Randomness is simply part of a tournament.  Whether it is the randomness of how many LSs your opponent draws, or whether who your random opponents are.  I still think that a player who beats 2 top players had a better tournament than a player who only beats one.  Sure the other guy will complain that it wasn't fair that he wasn't randomly picked to play another top player.  But people complain that it isn't fair that their opponent didn't draw LSs.  My proposal only evaluates who had the best tournament, not whether everyone had a totally fair outcome of all random events.

And I think it's actually doing a pretty good job of that (from a completely biased perspective of course)

So when are we going to implement a top cut in Redemption?

What is "top cut?" You keep using that phrase. I do not think it means what you think it means.

Quote from: EmJayBee83 on July 24, 2012, 07:16:45 PM

So when are we going to implement a top cut in Redemption?

What is "top cut?" You keep using that phrase. I do not think it means what you think it means.

Top cut is a style of tournament where after X rounds (depending on the number of participants) only the top portion of players continue playing. So in a 10 round nationals, everyone might play for 6 rounds and then only the top so many players continue on for the last few rounds.

Someone can add to this or say what is misspoke as I have no played other card games that implement this. Above is my understanding of it loosely.

Kirk

Top cut tournaments are the absolute closest (and most practical) you will come to a perfect system. There is no reason this should have not been implemented in at least Nationals by now.

I don't buy that for a second. Top cut does not impose on the foster and fellowship of Redemption anymore than standard Swiss does. I fail to see how the tournament format has any bearing whatsoever on the fellowship of the game. It's mixing apples with oranges.

From my understanding (from this thread), the majority of the players would only play for 3/4 of the tournament, rather than everyone being able to play every game. That's how the tournament format has a bearing on it. Using the idea of the others still playing and just not included in the possibility for placing could continue to foster fun and fellowship, but I'm guessing that one bad game would be more likely to exclude a person from the top-cut than it would be to exclude them from placing in the current way of playing. If I'm wrong, feel free to explain. For example, the tournament that was already brought up where Gabe won after losing his first two games - would he have been in the top-cut?

I think what needs to be done is some system for determining the winner in the case of a tie should be implemented (whatever it is), and it should be the agreed standard moving forward.

(I wrote the above before Matt's post)

Matt, that's a really good idea! Since having a tie is allowed and happens sometimes anyway, perhaps we should just leave it at that since it would stop all the arguments for the different systems.

From my understanding (from this thread), the majority of the players would only play for 3/4 of the tournament, rather than everyone being able to play every game.

I do not think this is necessarily correct. Let's use a Nationals three-round top cut as an example.  You would do a standard Swiss for the first seven rounds and use the result of those rounds to determine and rank the top eight players. In round #8 you would have 1v8, 2v7, 3v6, and 4v5 play and players 9-the rest would be paired up using standard Swiss. In round #9 the winner of 1v8 would play the winner of 4v5 and the winner of 2v7 would play the winner of 3v6 while the losers of the four named matches would join everyone else in playing a Swiss paired games. The final round would have the two semi-final winners paired for first and second and the other two players meeting in a consolation match for third. Everyone else would play Swiss paired games.

In the end, everyone in the tournament would get to play ten games and we would have an clear 1st through 4th place rankings that were determined by head-to-head competition rather than by a formula or a tie breaker or...

Using the idea of the others still playing and just not included in the possibility for placing could continue to foster fun and fellowship, but I'm guessing that one bad game would be more likely to exclude a person from the top-cut than it would be to exclude them from placing in the current way of playing. If I'm wrong, feel free to explain.

We could probably figure it out if we wanted, but there is a different way to look at this. So far at Nationals we have never had 128 or more players play T2 (and with the exception of the last two Nationals in MN, we have always been closer to 64 players than to 128)--which means 7 rounds would suffice. So a player missing the cut would be one who would have been six or more spots out of placing in any normal tournament. This does not strike me as something that would happen due to one bad game.

For example, the tournament that was already brought up where Gabe won after losing his first two games - would he have been in the top-cut?

According to the spreadsheet Gabe would have been tied for sixth after round #7. So even two bad games out of seven would not necessarily eliminate you.

Matt, that's a really good idea! Since having a tie is allowed and happens sometimes anyway, perhaps we should just leave it at that since it would stop all the arguments for the different systems.

My order of preference would be  top cut > ties > some tie-breaking formula.

It's a game. It's meant to have a winner and a loser. Ties are like kissing your sister...

In the word of the great Herman Edwards, "You play to win the game!"

I agree redemption is less competitive then most games and is intended for fun, but why bother keeping score if you do not care about winning. Or really do the Christ like thing and let the person who does care beat you.

I mean is wanting to win really that bad?  Is being competitive sinful?