FAQFAQ   SearchSearch   MemberlistMemberlist   UsergroupsUsergroups 
 ProfileProfile   PreferencesPreferences   Log in to check your private messagesLog in to check your private messages   Log inLog in 

Reply to topic Go to page Previous  1, 2
View previous topic :: View next topic  
Author Message
FredactOffline
Vip
Vip
Avatar

Joined: 20 Feb 2014
Total posts: 18
Gender: Unknown
Usa
PostPosted: Tue Mar 11, 2014 5:15    Post subject: Reply with quote

I'm new here, so maybe this has been explained before, if so, I'm sorry.

But why do tournament finals not simply play an odd number of games. Wouldn't that solve all these tiebreaker issues? And if it's a game that's prone to ties within the game, play until someone actually wins 2 (or 3 or whatever) games?
Back to top
View user's profile 
bramOffline
Superadmin
Superadmin
Avatar

Joined: 14 Jul 2005
Total posts: 1065
Gender: Unknown
Netherlands
PostPosted: Tue Mar 11, 2014 9:23    Post subject: Reply with quote

Fredact wrote:
I'm new here, so maybe this has been explained before, if so, I'm sorry.

But why do tournament finals not simply play an odd number of games. Wouldn't that solve all these tiebreaker issues? And if it's a game that's prone to ties within the game, play until someone actually wins 2 (or 3 or whatever) games?


1. Odd number is not fair. Most games have an advantage for the first or second player. In our opninion a odd number of games is less fair then our current system!
2. Starting new games when there is a tie would prolong tournaments in theory indefinetly.
Back to top
View user's profile 
Pianomatic
PostPosted: Wed Apr 16, 2014 15:00    Post subject: Reply with quote

bram wrote:
Fredact wrote:
I'm new here, so maybe this has been explained before, if so, I'm sorry.

But why do tournament finals not simply play an odd number of games. Wouldn't that solve all these tiebreaker issues? And if it's a game that's prone to ties within the game, play until someone actually wins 2 (or 3 or whatever) games?


1. Odd number is not fair. Most games have an advantage for the first or second player. In our opninion a odd number of games is less fair then our current system!
2. Starting new games when there is a tie would prolong tournaments in theory indefinetly.


In response to your two responses:
1. It's true that in some games, going first or second can give an advantage to one player, but the way that TPR works is confusing to many (as you can see from this discussion). My hunch here is that many players would feel as though they are treated more fairly if there were a way to break ties that was more transparent
2. Every tournament will eventually come to an end, in reality. If you're playing a dice game that ends when a player rolls a 5, eventually it will happen, and the game will be over. It's impossible for the game to go on forever. With infinite dice rolls, every number will be hit somewhere along the way. Similarly, a tie breaker may prolong a tournament, but somewhere down the road, it will definitely end.

Some additional food for thought:
There's a TPR calculator, but players have a lot of work to do in order to find out whether a tie will be in their or their opponent's favour. In addition, since the equation used is not available for users to examine, it doesn't make sense to many players. (Again, see the discussion in this thread.)

Based on the frustration expressed in this thread over the last several months, I believe a reexamination of the use of TPR is probably a very good idea. If you insist on continuing to use it, you really ought to be more open about how it works. There's no reason why it needs to be a secret.
Back to top
FredactOffline
Vip
Vip
Avatar

Joined: 20 Feb 2014
Total posts: 18
Gender: Unknown
Usa
PostPosted: Fri Sep 26, 2014 5:11    Post subject: Reply with quote

bram wrote:
Fredact wrote:
I'm new here, so maybe this has been explained before, if so, I'm sorry.

But why do tournament finals not simply play an odd number of games. Wouldn't that solve all these tiebreaker issues? And if it's a game that's prone to ties within the game, play until someone actually wins 2 (or 3 or whatever) games?


1. Odd number is not fair. Most games have an advantage for the first or second player. In our opninion a odd number of games is less fair then our current system!
.


Ok thanks, I'd not thought about the games with player order advantage. With that in mind, you could still go to 5 games though, and let higher TPR determine which player gets the advantaged position in three of the games.
Back to top
View user's profile 
karaklis
PostPosted: Sun Sep 28, 2014 9:47    Post subject: Reply with quote

bram wrote:

1. Odd number is not fair. Most games have an advantage for the first or second player. In our opninion a odd number of games is less fair then our current system!

So actually the task is to minimize the unfairness. There are two options:

a) TPR as tie breaker
b) odd number of games

Is there a method to measure the unfairness? Probably not, but my gut feeling tells me that an odd number of games is less unfair. Let me elaborate.

a) When you have advanced to the knockout phase, it most likely means that you have beaten all or all but one of the players in your group. That applies to your opponent as well. The TPR will be determined by your rating and the rating of your previous opponents. This is microscopically better than drawing by lots, and drawing by lots is maximally unfair. If drawing by lots as tie breaker is 100% unfair, TPR as tie breaker is 99.9% unfair, since you have no influence on the fluctuations of your opponents' rating. They might be advantageous or disadvantageous for you.

b) If you have an odd number of games, let's say five for a tournament final, or nine for a championship final, you might have an advantage (e.g. chess) or a disadvantage (e.g. atoll) or neither measurable advantage nor disadvantage (e.g. go, due to komi). Let's assume that on average the first player has an advantage of 52.5% vs. 47.5% and that this advantage is granted to the person with the better TPR, the winning ratio will be roughly 50.5% (for five games) or 50.28% (for nine games), so the unfairness is just 1% or 0.56% respectively.

Even if the assumption about the TPR as tie breaker is just a gut feeling, it would not change anything about the fact that an odd number of game would be *much* fairer, since the TPR would also determine who will get the minimal advantage of starting the fifth or ninth game. So if the unfairness of TPR as tie breaker is actually only 50% instead of 99.9% as explained in (a), it would also reduce the unfairness to 0.5% or 0.28% respectively in (b).

Conclusion: Comparing (a) and (b), an odd number of games would be *much* fairer than TPR just as tie breaker.
Back to top
bramOffline
Superadmin
Superadmin
Avatar

Joined: 14 Jul 2005
Total posts: 1065
Gender: Unknown
Netherlands
PostPosted: Sun Sep 28, 2014 12:02    Post subject: Reply with quote

You are allowed your opinion Smile But keep in mind you did not prove anything but stated your opinion as I did mine. What do you think will happen now?? Smile Smile
Back to top
View user's profile 
karaklis
PostPosted: Sun Oct 12, 2014 17:26    Post subject: Reply with quote

Hey Bram Very Happy

of course everybody may keep their opinion. I also know that everyone has their own feeling of what is fair and what is unfair (though as far as I understand the scientific approaches that try to clarify the term "justice" suggest that most humans have a common basis in this regard).

I just wanted to elaborate why I think that an odd number of games would be objectively fairer than an even number of games. The given numbers are just examples to show this, but even if the real numbers differed, it would change nothing about the statement itself.

What I expect to happen now? Well, to be honest, nothing, because before any action on this will be taken, two requirements must be fulfilled:

1) The insight of the responsible persons that and why an odd number of games would objectively be fairer

2) The capacity of doing the necessary changes

I have no knowledge about the code base, but my experience as a programmer tells me that such changes would likely have a deep impact on the coding. You probably have your own priorization of the changes that you would like to implement, and I wouldn't wonder if high-effort changes had a low priority. Very Happy
Back to top
Display posts from previous:   
Reply to topic All times are GMT + 1 Hour
Go to page Previous  1, 2
Page 2 of 2

 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum


Powered by phpBB © 2001, 2005 phpBB Group