Why punish Hanabi players if we all agree to abandon the game?

[Travis Hall]

Now there is no grace/mercy. Because who wants to take -10 for your unfortunate situation? You ask to concede and there is no incentive to agree. You will be directed to simply quit.

I too regret the loss of our capacity to apply human judgement to these situations, but if people have been abusing abandons and a hard line has to be taken, the loss of rating and reputation for the player who has to drop out is appropriate.

If leaving games is not a habit for a person, it should not be a problem to regain the reputation points pretty quickly.

[Earthboundia]

Hello Hanabi players,

I realized recently that now if we choose to abandon the game in Hanabi, all players will lose 10 ELO points.

Who implemented this new rule and why was it implemented?

In other games, you do not lose ELO points if we chose to abandon the game together.

How do i ask the Hanabi developers to change back to the old rule whereby no ELO points are lost when we choose to abandon the game together?

Do i have to create some sort of petition or what? What do i have to do?

Please help!!

Thank you
patton

In most games people collectively abandon when they don't have time to complete the game for whatever reason. In Hanabi people abandon to keep their record and ELO intact when someone makes even the slightest miscalculation. A 10 ELO loss would encourage players to complete their games.

[ChiefPointThief]

A 10 ELO loss would encourage players to complete their games.

No it doesn't. The only difference now is that before people would cave into the bomber before the final blow. Now the bomber takes everyone with them. These players already benefitted from the old system so the concede change doesn't do much in that regard. Even if someone only purposely bombs 1/4 of the time they are still gaining elo overall. A lot of these players have already reached into the 900s and 1000s. It would take lots of games before they would fall under what is considered the highest metric (master level).

[Travis Hall]

Even if someone only purposely bombs 1/4 of the time they are still gaining elo overall. A lot of these players have already reached into the 900s and 1000s. It would take lots of games before they would fall under what is considered the highest metric (master level).

No, they aren’t still increasing their ELO rating at that point. You overlook the fact that players with very high ratings also gain much less when they complete a game with a good score.

I don’t believe anybody with a master ranking is deliberately scoring 0 in a quarter of their games, and still increasing their rating. And frankly, if they can do that, their legitimate rating would be higher than their actual rating is (because that’s how ELO systems work).

I advise that when you encounter a player who deliberately bombs out a game, you simply give them a red thumb and move on. Once you give a red thumb, you have achieved the purpose of the rating system - you have excluded a bad player from your games. Leave any jealousy you may have concerning their rating aside, and just play with good players. Their behaviour will inevitably result in a correction to their own ratings, even if it takes longer than you might hope.

[Stroom]

Even if someone only purposely bombs 1/4 of the time they are still gaining elo overall. A lot of these players have already reached into the 900s and 1000s. It would take lots of games before they would fall under what is considered the highest metric (master level).

No, they aren’t still increasing their ELO rating at that point. You overlook the fact that players with very high ratings also gain much less when they complete a game with a good score.

I don’t believe anybody with a master ranking is deliberately scoring 0 in a quarter of their games, and still increasing their rating. And frankly, if they can do that, their legitimate rating would be higher than their actual rating is (because that’s how ELO systems work).

I advise that when you encounter a player who deliberately bombs out a game, you simply give them a red thumb and move on. Once you give a red thumb, you have achieved the purpose of the rating system - you have excluded a bad player from your games. Leave any jealousy you may have concerning their rating aside, and just play with good players. Their behaviour will inevitably result in a correction to their own ratings, even if it takes longer than you might hope.

Exactly this. Bad players are brought out more often so they get red thumbed and ignored in the future. What's sad about this is that even if they change their ways, I don't have any way of knowing and I can't remove the red thumb.

This change was a very good addition overall. I have had much more pleasant games. I would argue that as players are forced to play through tough decks and they can't just quit, they are going to learn more advanced strategies which increases their skill level.

[ChiefPointThief]

Even if someone only purposely bombs 1/4 of the time they are still gaining elo overall. A lot of these players have already reached into the 900s and 1000s. It would take lots of games before they would fall under what is considered the highest metric (master level).

No, they aren’t still increasing their ELO rating at that point. You overlook the fact that players with very high ratings also gain much less when they complete a game with a good score.

I don’t believe anybody with a master ranking is deliberately scoring 0 in a quarter of their games, and still increasing their rating. And frankly, if they can do that, their legitimate rating would be higher than their actual rating is (because that’s how ELO systems work).

I advise that when you encounter a player who deliberately bombs out a game, you simply give them a red thumb and move on. Once you give a red thumb, you have achieved the purpose of the rating system - you have excluded a bad player from your games. Leave any jealousy you may have concerning their rating aside, and just play with good players. Their behaviour will inevitably result in a correction to their own ratings, even if it takes longer than you might hope.

No. Not overlooking anything. 1 bomb = -10. Which means to maintain their current elo they would need an average of +3.33 in games in which they aren't bombing if they are bombing at a 25% rate. This was a hypothetical math equation. Even if said person does this in 20% of games which is still a very high rate they will maintain or slightly rise as well. I've encountered such players and just took a look at a few of their elos to see if my statement stands. 2 masters and one in the upper 500s (I'm sure I have encountered more but 3 is enough for research purposes). They do this habitually looking at their history.

I don't know how you got jealousy from my post. I stopped responding to you on the other topic for commenting me ****** things.

Exactly this. Bad players are brought out more often so they get red thumbed and ignored in the future. What's sad about this is that even if they change their ways, I don't have any way of knowing and I can't remove the red thumb.

This change was a very good addition overall. I have had much more pleasant games. I would argue that as players are forced to play through tough decks and they can't just quit, they are going to learn more advanced strategies which increases their skill level.

The community as a whole may not be better. You've probably just excluded most of the bad apples at this point. I've re thumbed several people as well. So I think my rate of bad games is going down. Some of them weren't bad people but our playing styles didn't mesh. I supported the suggestions to only red thumb players in particular games for this reason but that is probably difficult to accomplish and won't happen. I also switched to playing only 2p and flams though which I think is a factor as well. The less people at the table lowers the chances of being paired with a ticking time bomb and flams fix mistakes.

[Stroom]

No. Not overlooking anything. 1 bomb = -10. Which means to maintain their current elo they would need an average of +3.33 in games in which they aren't bombing if they are bombing at a 25% rate. This was a hypothetical math equation. Even if said person does this in 20% of games which is still a very high rate they will maintain or slightly rise as well. I've encountered such players and just took a look at a few of their elos to see if my statement stands. 2 masters and one in the upper 500s (I'm sure I have encountered more but 3 is enough for research purposes). They do this habitually looking at their history.

I don't know how you got jealousy from my post. I stopped responding to you on the other topic for commenting me ****** things.

Bombing happens rarely. Maybe once every 100 games. You have to play really badly for that. Most often you end up with 1-2 points less than maximum.

Also, from your history it seems like you only play with 5 colors. That is a super easy mode and it justifiably provides you less elo. So your skill level lands between 700 and 800 with those settings. If you would try to add multicolors and adjust your tactics to that, you will see more elo gains from getting 30 points. At the moment, what you are experiencing (elo going down a lot and slowly rising back to the old value and then dropping again) is your elo reaching the representation of your skill level. This is the expected outcome and works as it should. Elo is not supposed to increase indefinitely and it is supposed to have drops from time to time.

The community as a whole may not be better. You've probably just excluded most of the bad apples at this point. I've re thumbed several people as well. So I think my rate of bad games is going down. Some of them weren't bad people but our playing styles didn't mesh. I supported the suggestions to only red thumb players in particular games for this reason but that is probably difficult to accomplish and won't happen. I also switched to playing only 2p and flams though which I think is a factor as well. The less people at the table lowers the chances of being paired with a ticking time bomb and flams fix mistakes.

from my experience, the games are better now. I red thumb people less. Players are learning to become better players. Less people are bombing the game too.

[Travis Hall]

No. Not overlooking anything. 1 bomb = -10. Which means to maintain their current elo they would need an average of +3.33 in games in which they aren't bombing if they are bombing at a 25% rate.

You can’t get that when your team has an average rating of 900. When I play 4-player games with players of a similar rating, we won’t even get 1 point for scoring 30. With 5, I think we get about 3 points, which still won’t get us the 3.33 average.

The community as a whole may not be better. You've probably just excluded most of the bad apples at this point.

My experience is certainly improved by having red thumbed a lot of the most toxic players. However, my play experience has notably improved (further) since the abandonment adjusted ELO ratings as a 0 score.

I used to regularly have to deny abandon requests repeatedly during games - something like 1/4 of all games I played would have persistent requests to abandon. Often they were accompanied by players complaining about being forced to play (which we all know was really a complaint about how they were going to suffer a rating decrease). Since the change, I’ve seen two abandon requests in precisely one game, out of dozens, possibly hundreds. A few other games have had one abandon request, but after a single denial those players settled down and just played.

The change certainly has not fixed all the problems in this community, but it has helped.

[Wreckage]

These players already benefitted from the old system so the concede change doesn't do much in that regard. Even if someone only purposely bombs 1/4 of the time they are still gaining elo overall. A lot of these players have already reached into the 900s and 1000s. It would take lots of games before they would fall under what is considered the highest metric (master level).

I can see why you would think this is unfair. What you don't realize is it doesn't work like you are describing. If your team average is 900s and 1000s, a perfect score probably gains 0.00 elo. You lose elo for scoring 29 or less. You can gain elo if there are some 800- players in the game, well they might have to be 700-.

The system is much better now. Now some players will maintain 900+ elo legitimately, and the players who had 900+ elo by abandoning all their losing games, will now start dropping to where they belong.

[ChiefPointThief]

Bombing happens rarely. Maybe once every 100 games. You have to play really badly for that. Most often you end up with 1-2 points less than maximum.

Also, from your history it seems like you only play with 5 colors. That is a super easy mode and it justifiably provides you less elo. So your skill level lands between 700 and 800 with those settings. If you would try to add multicolors and adjust your tactics to that, you will see more elo gains from getting 30 points. At the moment, what you are experiencing (elo going down a lot and slowly rising back to the old value and then dropping again) is your elo reaching the representation of your skill level. This is the expected outcome and works as it should. Elo is not supposed to increase indefinitely and it is supposed to have drops from time to time.

from my experience, the games are better now. I red thumb people less. Players are learning to become better players. Less people are bombing the game too.

I am happy for you that bombing occurs in 1% of your games. I know if there was some bga stats as a whole the # is higher. Personally I have more than 5 games in which bombing occurred in my 500 game history. You can lump losing on purpose in general among those stats. Just in the last 3 weeks I've seen several topics from players concerned about people losing Hanabi games on purpose. It is very possible though that bombing occurs less now and people typically probably remember those games more than others. But it is still a problem.

Also I don't play w/ 5 colors normally. Recently I have been playing a lot w/ a friend who is still learning. That is the only person I play 5 colors w/. But my stance has nothing to do w/ my elo or me being envious of others elo. Unsure how my original comment was perceived that way.