doubles in Castles of burgundy

[tbhp]

Wait, OP is the one conceding all those games?

I was under the impression that it was their opponents who were conceding the games. OP says they care about fairness, so I assumed they were the one being conceded against, not the other way around, silly me.

This makes the situation potentially very ironic. It seems that OP has an obsession with double rolls working against them. So, if every time they get a lot of double rolls in a game and they aren't doing well, they decide to concede because of it, then later they decide to go collect the data to prove they are right, and they count the conceded games as if they had 25 rolls when in fact they had fewer, giving those the same weight as the other games, then obviously the results are gonna make it look like there were more double rolls than there actually were.

So, with their paranoia and their conceding tendencies, it seems that maybe OP created themselves the conditions for their data collecting to be flawed. Like a self fulfilling prophecy.

[FrankJones]

It looks as though most games are played to completion (meaning without either player conceding). I assume that means it's usually a close game.

[tbhp]

OP has 6 conceded games in their past 100 games, 5 of which were conceded by them.

This is the kind of thing that should impact karma and should be listed alongside clock violations. Who cares how many times OP ran out of time when next to that, you can see that they have a 5% chance of conceding when playing against you? That's clearly the bigger of the two issues.

[FrankJones]

This makes the situation potentially very ironic. It seems that OP has an obsession with double rolls working against them. So, if every time they get a lot of double rolls in a game and they aren't doing well, they decide to concede because of it, then later they decide to go collect the data to prove they are right, and they count the conceded games as if they had 25 rolls when in fact they had fewer, giving those the same weight as the other games, then obviously the results are gonna make it look like there were more double rolls than there actually were.

True, that could explain the unusually high numbers.

[FrankJones]

OP has 6 conceded games in their past 100 games, 5 of which were conceded by them.

This is the kind of thing that should impact karma and should be listed alongside clock violations. Who cares how many times OP ran out of time when next to that, you can see that they have a 5% chance of conceding when playing against you? That's clearly the bigger of the two issues.

As you know from my past posts and our discussions, I, similar to you, hate conceded games. But, I think you and I ( and a few others) are in the minority on this one.

That said, if people have no objection to conceding, I see no reason why we can't see "concede percentage", because I'd like having a filter to screen out anyone who concedes a lot. I like my games to be completed.

[tbhp]

As you know from my past posts and our discussions, I, similar to you, hate conceded games. But, I think you and I ( and a few others) are in the minority on this one.

It all seems very arbitrary. On what basis do we even decide that a behavior counts as a "red flag"? Shouldn't we list all the things that people may have an issue with?

People like me have relatively no issue with opponents going out of time (unless it leads to things like games not being able to be completed in time), yet we are forced to see a metric we don't care very much about being marked on the user's profile, while there are other metrics that would be much more relevant to know.

How many times a user has kicked out is also something that should be marked. There should also be a distinction with the timeouts that were associated with an early game termination, versus those that didn't have an impact on the conclusion of the game. Of course, the former would exclude the 2-player games where the termination was due to the opponent expelling, since in those cases the game termination is solely the fault of the expelling player, not the out of time player.

All in all, those factors could be put together in a "player reliability" metric, as an alternative to the karma system. This reliability score would give us an idea of how likely this player's games are to be played to completion (not counting the cases where another player was at fault for the termination).

That would be much more useful to me than knowing how often someone goes out of time.

[nik592]

OP has 6 conceded games in their past 100 games, 5 of which were conceded by them.

This is the kind of thing that should impact karma and should be listed alongside clock violations. Who cares how many times OP ran out of time when next to that, you can see that they have a 5% chance of conceding when playing against you? That's clearly the bigger of the two issues.

As you know from my past posts and our discussions, I, similar to you, hate conceded games. But, I think you and I ( and a few others) are in the minority on this one.

That said, if people have no objection to conceding, I see no reason why we can't see "concede percentage", because I'd like having a filter to screen out anyone who concedes a lot. I like my games to be completed.

I don't like conceding either, but I do accept it's a valid way for a game to end (as is someone being kicked for being over time, which I don't often do, and I will sometimes judge people (and red thumb them) for being too quick on the kick button). And I don't agree that there should be penalties for either (kicking other players or conceding), and I'm not sure how I feel about being able to filter on players who concede. That said, I am ok with red thumbing people on that basis, and that's more or less filtering, so I don't know if my feeling about it really holds weight.

I do realise it wasn't you who was advocating for penalties for conceding here, so I don't know if you were agreeing with that idea, or just proposing filtering.

On a practical side though, the more options you give players on which to filter their opponents, the more difficult it becomes to make matches (and the more work servers are doing to find matches), so there is that side too. One more thing may not cause much overhead, but if you agree to that, does it open the door to more things to filter on (eg. kick percentage, time violations, etc.).

[FrankJones]

True. I think. I think I agree with nik592.


I hate concedes

I also had two recent opponents who did this:

Queued for fast turn based.

Made No love during the 12 hour window.

I cannot force abandon or expel.

So, I eventually forced abandoned game.

I could have taken two cheap wins. But I did not.

That said, both of these players annoyed the hell out of me. MAKE A GODDAMN MOVE.

[Meeplelowda]

Can confirm end-game stats are displayed for abandoned games:

https://boardgamearena.com/table?table=925105170

Okay, so, the only remaining issue is, how much are the stats skewed by not taking a weighted average to account for conceded or abandoned games?

There would need to be a lot of conceded or abandoned games to explain the differential between 16.67% and 17.25%.

Might there be a correlation between conceded and abandoned games and a deviation from expected number of doubles? Would this correlation compound the error of giving them equal weight?

[tbhp]

I don't like conceding either, but I do accept it's a valid way for a game to end (as is someone being kicked for being over time, which I don't often do, and I will sometimes judge people (and red thumb them) for being too quick on the kick button). And I don't agree that there should be penalties for either (kicking other players or conceding), and I'm not sure how I feel about being able to filter on players who concede. That said, I am ok with red thumbing people on that basis, and that's more or less filtering, so I don't know if my feeling about it really holds weight.

So you don't like it and you even judge people for it, yet you seem to just accept that it's valid for people to do it. Why does it have to be that way? Why is it us who have to bow down to the people who like conceding and expelling, when there are already other behaviors that are considered not acceptable by the website? It all seems very arbitrary.

I mean, the people who don't like people who go out of time could also say "I don't like it, but I accept that it's valid for people who do it". Except, they don't have to do that, the site is already taking care of their issue by making the behavior penalizable and visible on their profile. Why can't we have that too with conceding and expelling?

On a practical side though, the more options you give players on which to filter their opponents, the more difficult it becomes to make matches (and the more work servers are doing to find matches), so there is that side too. One more thing may not cause much overhead, but if you agree to that, does it open the door to more things to filter on (eg. kick percentage, time violations, etc.).

But if the system encouraged the idea that conceding and expelling are not good things to do, then people would try to be good users and a lot less people would do those things. So it would become less a case of "I need to filter out the people who do those things" and more that people would avoid doing them in the first place, just like how people avoid going out of time.