Being able to abandon games

[Stuff by Bez]

The idea of counting all co-op games as a loss when abandoned seems perfectly reasonable to me.

[Wreckage]

...ELO has very little meaning.

So you think there is no difference between say a 500 ELO and a 900 ELO?

Haha, let me rephrase. Change "ELO has very little meaning" to "ELO is not currently a reliable indicator of skill". I would agree, generally speaking players with 900 ELO tend to be better than players with 500 ELO. I do like that the rest of your post goes on to support my statement though.

...because they mostly play 2p or playing Basic or Tricky and don't learn the nuances of MC and BP.

How did it get this way? It's some combination of...
- Players who cheat by abandoning every game where they won't get a 30.
- Players who fluff their ELO via playing tons of 2p games
- Players who fluff their ELO via playing easier versions of Hanabi (basic, tricky, and/or no black powder.)

Two things here... Players who have substantial artificially high ELO are rarely getting it from playing basic or tricky settings. I have never seen it. If they are "fluffing" their ELO, they are doing it on avalanche.

The second thing is this arrogance against players playing 2-player games. I get it. I understand it. They are not doing finesses, and for you, that's the whole game. It is a different game. But this doesn't give them any advantage in ELO. It's not an easier game to reach 30 points. And it's not an ELO boost.

Let's consider 2p vs 5p games. I have to think all the advantages go to the 5p game. First, the ability to get multiple plays from a single clue is huge. Second the number of cards visible is another huge advantage. Even 3p has the advantage of another set of eyes making sure there's not a double discard.

I do think players who have exploited the ELO system the most, have done it on 2-player Avalanche games. The reason is it's easier to find 1 strong teammate than 4 strong teammates with really low ELO.

[Travis Hall]

The second thing is this arrogance against players playing 2-player games. I get it. I understand it. They are not doing finesses, and for you, that's the whole game. It is a different game.

At least many of us recognise that our skills in 3+ player Hanabi do not directly translate into 2-player. (Honestly, some of our skills do translate, but for best results you need some 2-player-specific modifications to your play.) The real problem with many 2-player specialists is that they don’t understand that their skills won’t work for larger games.

When played by skilled players, you’d be less likely to get a perfect score in 2-player. Unfortunately that means that there are 2-player specialists who think they are hot stuff, because they are playing the “hardest” form of the game. That’s not how it works. And it results in players joining larger games and playing extremely poorly because they don’t know what they are doing.

A lot of 2-player specialists, if they want to transition, really need to join some lower level games for a while and build up their understanding before joining games at their supposed rank. I’d have to do the same in the other direction, but less so… mostly because I’m already aware of the differences.

But this doesn't give them any advantage in ELO. It's not an easier game to reach 30 points. And it's not an ELO boost.

The ELO calculations are a little different for different player counts, and I’m pretty sure this does make it a bit easier to build up ELO with certain player counts. I don’t remember whether it is most favourable to 2-player, but my opinion is that it’s 3-player that the settings are hardest on.

I do think players who have exploited the ELO system the most, have done it on 2-player Avalanche games. The reason is it's easier to find 1 strong teammate than 4 strong teammates with really low ELO.

I think the bigger advantage of 2-player for ELO building is even simpler than that. It’s mostly just that the games are faster. Even if total play time per game is similar, you’ll usually wait less time in the lobby for a single partner, compared to three partners. And these players are trying to churn through games as quickly as possible. When you’re abandoning the games that will cause an ELO loss anyway, it doesn’t matter much if the rush causes a mistake. Nor does it matter much if your partner us as weak as their rating says, because if the game goes well, that’s just more ELO points, and if it doesn’t you’ll abandon quickly.

[Jellby]

Also, 2-player games are easier to abandon, you only need 1 other player to agree.

[Wreckage]

When played by skilled players, you’d be less likely to get a perfect score in 2-player. Unfortunately that means that there are 2-player specialists who think they are hot stuff, because they are playing the “hardest” form of the game. That’s not how it works.

A 2-player game is much easier to learn. I can't disagree with that. Multi-player games, typically may have several finesses in action simultaneously. Not only that, in a 2-player game, when my partner gives a clue, I know it's for me, there's no question. So in terms of complexity, there's more going on in multi-player, so there's more to learn to perfect it.

In terms of "hardest", Difficult/2-player/Black Powder/No Flams is the hardest setting on BGA (to reach 30 points), but it's a lesser learning curve to master than the same settings with 5 players.

ELO calculations are a little different for different player counts

We've had this discussion before. I believe 2p and 5p gain the same ELO for the same scores (could be wrong). 4p gains less. And 3p definitely gains the least. So, in a way players who play 5p games gain less than 2p gamers, because players who play 5p also tend to play 4p games too.

The other points about getting more games played in a short amount of time also make sense. In my own experience, there have been a few times I have found a partner with 200-300 ELO who is really a much better player than expected. If I play 10 games in a row with them I might gain 50-60 ELO in a single sitting.

I think there are a number of reasons why 2-player gamers will generally have higher ELO compared to multi-player gamers, probably by about 150 points or so. If you had 5 Masters with 800 ELO at one table, and 2 players with 800 ELO at another table, and they played the same number of games, the advantage would go to the 5 players. But in practical terms, that's not what's happening.

And one last point that some of you are trying to make, it's possible for a player to be the best player on BGA in 2p, and still be a baby at 5p. These are two different games.

[Travis Hall]

In terms of "hardest", Difficult/2-player/Black Powder/No Flams is the hardest setting on BGA (to reach 30 points), but it's a lesser learning curve to master than the same settings with 5 players.

Yes, exactly. Mastery gets you better chances in 5-player, but you need that mastery.

One one player playing without mastery can drag the whole team down, even if everyone else has mastery. Some of us know how to adapt to a semi-competent player to some extent, but an incompetent player can easily be bad enough that nothing can save a team.

ELO calculations are a little different for different player counts

We've had this discussion before. I believe 2p and 5p gain the same ELO for the same scores (could be wrong). 4p gains less. And 3p definitely gains the least. So, in a way players who play 5p games gain less than 2p gamers, because players who play 5p also tend to play 4p games too.

Yeah, the relevant constants are somewhere in this forum, but I can’t find them. I just don’t remember exactly how 2p, 4p, and 5p compare.

But 3p is the one that it’s harsh on. It’s not just the constant in the calculation, although I think that’s part of it; 3p has a tough token economy, and a locked hand can put you in a cycle you just can’t break, all while needing to understand the multiplayer signals. (But not layered. At least you don’t have to worry about layered in 3p.) If you want to sharpen your hand and token management skills, play 3p.

I think there are a number of reasons why 2-player gamers will generally have higher ELO compared to multi-player gamers, probably by about 150 points or so. If you had 5 Masters with 800 ELO at one table, and 2 players with 800 ELO at another table, and they played the same number of games, the advantage would go to the 5 players. But in practical terms, that's not what's happening.

Again, simpler than that. 800 ELO is not perfect; everyone still makes mistakes. In 5-player, you’ve got 5 players whose attention can wander, or who just don’t figure out what a clue means. In 2-player, only 2.

If everyone plays very well, you are much more likely to get 30 in 5-player. *If*. But you have 5 players who can fail to play well.

[Jellby]

Yeah, the relevant constants are somewhere in this forum, but I can’t find them. I just don’t remember exactly how 2p, 4p, and 5p compare.

viewtopic.php?t=4814

Last time I checked, the values for multicolor 5-player applied also with black powder + flams (at least for scores 28-30). I guess all other values are unchanged too, but i didn't check them.

[Travis Hall]

Yes, that’s the topic.

So, according to those numbers, Avalanche 2-player is regarded as slightly easier than 5-player. Very marginally so, though.

And 3-player is easier than 4-player, which is easier than 5-player, in much more significant jumps. And that’s just wrong. 3-player is definitely harder than 4-player. 3-player vs 5-player is very tricky to evaluate - each is difficult in its own way - but 4-player is the easy form, I think. (I host a lot of 4-player games basically to allow experts to train up. And 5-player masters-only because it’s just plain fun.)

It’s interesting to note that 2-player is rated easier than other player counts in non-Avalanche games. I doubt that is accurate.

[Wreckage]

I never saw this before, it's interesting. I have often heard people say that 3p is harder than 2p. I don't think this makes sense on any variant. The only possible reason it could be harder that I can think of, is because you can't save non-unique cards on chop. But this is just a self inflicted rule of playing by BGA conventions.

In the past we had some competitions which I called "Challenges". I would add a challenge rule, and all the teams involved would play the identical preset decks and compare scores. Our biggest week we had 13 2-player teams playing 4 decks each, and the rule was "you cannot clue 1, 2 or 5". Out of the 52 decks played, 25 games scored 30 points. All 13 teams were playing different conventions that they made up themselves.

Other weeks the rule was "Every clue must touch at least 2 cards" "A player can only have one number marked in their hand at any time", "Multi cards in the discard must be discarded in order" and many others. Some are very hard. We’ve played without giving a single color clue during the game, and got 30s. We had 3p games too, and a couple times 4p games. But it's hard to get very many teams with more than 2 players.

So, the skill of playing 2p was pushed during these events, and the actual skill was deciding what your conventions were during the week before the games started. You have to find new ways to clue things. There are many ways to clue that the vast majority of players have never considered. For instance, your partner is holding a 5 and several other cards on chop, you clue 5 to them again. It can't play, it can't be finesse of 4-5 cards, what can it mean? It can mean anything you want it to, but a clue like this is generally not even thought of by thousands of regular players.

Here's one that can help everyone in games of any player count with no black powder. If someone gives you a 1 clue at the start, and it hits three 1s, which do you play? Most of you would immediately say the right 1. But what if you played the middle one, or the left. What can you be telling them by playing out or order? We’ve used it to mean if I play middle, save your chop card. If I play left, save two cards. So it’s possible to save a 5 and a 2 without spending those tokens, and move the game forward at the same time.

Playing with weird restrictions can open your eyes for the first time to all the possible ways to clue you’ve never considered before. I have given a single clue to mean save different cards in slots 2, 3, and 5. Or save chop and play leftmost. Or play slot 2, then slot 1.

Now, I know you can’t do anything new when you are playing with different people every day. But, if you have the same team play together a couple times a week, there are things to discover, and that’s fun. Right now the best masters on BGA play the same conventions again and again, and it’s mostly about paying attention, not figuring out something new. I do this too. It’s pleasant to relax in your comfort zone. So the repeated complaint is playing with people who don’t have the conventions memorized as well as you.

[Travis Hall]

Just going to call attention to an edit I made in my last post. I meant to say what I have edited to, but I also considered “don’t think” instead of “doubt”, and ending up posting “don’t doubt”, which reverses my intended meaning.