Lets say you have XX 2G 3G XX XX and you're clued green. After you've played 2G, should you wait for another clue to be certain the next card is 3G, or you just play it even no further clue is given?
How should clue residue be handled
Forum rules
Please DO NOT POST BUGS on this forum. Please report (and vote) bugs on : https://boardgamearena.com/#!bugs
Please DO NOT POST BUGS on this forum. Please report (and vote) bugs on : https://boardgamearena.com/#!bugs
Re: How should clue residue be handled
I'd say "don't play yet". If possible, peoples try to clue only cards that will be eventually playable, so you can hope that your other green is a 3G/4G/5G (so not a 1G/2G), and depending on who you play with, it might even be reasonable to assume that it's the case so that it is one of those 3 cards (it's the "good touch principle"), but it's really dependent on the convention you use (and it's been quite a while that I didn't played on BGA so I'm not sure to which degree BGA follows the good touch principle).
However, expecting it to be a 3G exactly is quite strict, and I don't think anyone play like that.
Re: How should clue residue be handled
This is wrong though, many players have that idea.
But from experienced player on the internet, I rarely saw people argue for it.
On your example you have a g2 known. The other card is 1/1/1/2/3/3/4/4/5. 9 possibilities.
1/1/1/2 is unlikely, but with your way, that causes a bomb immediately. If later you can clue 1: r1-x-x-(g)1-x, you get that correction for free.
But so you get 5 main possibilities left. 40% of those is a play. 60% would require a fix.
But on those 60%, you can finesse the card saved. So as soon as a g4 is drawn, you can clue it to make the g3 played for free.
If both g4 & g5 are visible, that g3 can be played later on. It can look unlikely, but if yourself you discard a g4, or if you think you should take a risk later on, it's still some strong options if you see another g4 (even if there is an high chance it could be clued if it's playable).
Same if you receive a 4 card or a 3 clue which is for another card.
That leave to lots of situations where if you don't play it, you can still play it later on.
.
Playing a card make the next card of that suit playable (but then you would see one copy of it already) and make you see one more card on the deck.
Discarding a useless card make you win one clue and make you see one more card on the deck.
Discarding a useful card is pretty annoying and can cost you the game, but still make you win one clue and see one more card on the deck.
Playing a card early is good, but isn't that good.
-
Travis Hall
- Posts: 180
- Joined: 12 April 2020, 14:13
Re: How should clue residue be handled
“Play one” vs “play all”.
There are a few people who like play all, but they don’t reach high ranks on BGA. The only advantage to play all is high efficiency, and there’s enough other techniques to improve efficiency that play all really isn’t worth it. Meanwhile, adhering to valid clues under play all means you have to let many clues go, and you miss out on using splash to save cards that is very useful under play one.
At a certain level of skill, emphasis starts to shift somewhat from efficiency to preventing loss of non-trash cards.
So basically if you hope to reach high levels of play on BGA, you need play one, and virtually everyone uses it.
Re: How should clue residue be handled
This convention is called "continuation" and is not uncommon. I agree with Travis though, it is limiting and when I switched away from it my scores improved.
It is a nice way to play and learn about playing without direct clues, the idea is that you need a second clue to "correct* it from playable to non-playable. The other way around you need a second clue to make it playable. If you play both conventions enough you will probably conclude the former uses more clues on average, which makes it less desirable.
That said, both plays are correct, it just depends on if the table has agreed to that convention
It is a nice way to play and learn about playing without direct clues, the idea is that you need a second clue to "correct* it from playable to non-playable. The other way around you need a second clue to make it playable. If you play both conventions enough you will probably conclude the former uses more clues on average, which makes it less desirable.
That said, both plays are correct, it just depends on if the table has agreed to that convention
Re: How should clue residue be handled
Agree that emphasis should be more on preventing loss of non-trash cards rather than efficiency if you're playing for ELO.
Guess it also depends on who you're playing with, I find it rather frustrating with play one but maybe it's because I play all with friends and play one with strangers. Some strangers compete to become the top clue dispatcher and clog your hand with transhable cards, ignoring good touch principle.
With continuation, I rarely need a fix because you have a chouce to either clue with color or clue with number and you can use the one which doesn't touch an unplayble card. If you have XX 2G 4G XX XX in a 3 player game where 3G is in another's hand, you can clue 3G right after 2G is played so the play is seamless. More beneficial if you have XX 2G 3G 4G XX.
Guess it also depends on who you're playing with, I find it rather frustrating with play one but maybe it's because I play all with friends and play one with strangers. Some strangers compete to become the top clue dispatcher and clog your hand with transhable cards, ignoring good touch principle.
With continuation, I rarely need a fix because you have a chouce to either clue with color or clue with number and you can use the one which doesn't touch an unplayble card. If you have XX 2G 4G XX XX in a 3 player game where 3G is in another's hand, you can clue 3G right after 2G is played so the play is seamless. More beneficial if you have XX 2G 3G 4G XX.
Re: How should clue residue be handled
Suppose r2 & g1 are on the stacks:dgjxqz wrote: ↑07 September 2022, 12:49With continuation, I rarely need a fix because you have a chouce to either clue with color or clue with number and you can use the one which doesn't touch an unplayble card. If you have XX 2G 4G XX XX in a 3 player game where 3G is in another's hand, you can clue 3G right after 2G is played so the play is seamless. More beneficial if you have XX 2G 3G 4G XX.
Hand is g2 x r1 g4 x
You can clue 2 to get g2 played that's true. But then you don't save g4. That mean after one play and one discard we suppose is trash or couldn't be avoided, you are left with x x x r1 g4. And you can't save g4. Then if the other copy of the g4 is at the bottom, you lose the game.
(an hand as g2 r2 x g4 x is unlikely, but is still possible. It require two clues: one to get g2 played, another one to get g4 saved.
if the hand is instead g2 r2 b4 g4 x, it became pretty good because you use one clue to save g2, and another one to save i suppose b4 & g4. Problem is that you can't choose to not save b4&g4. You rarely lack of clues at this point on the game, but that could happen)
(note that g2 x x g5 x use 2 clues instead of 1)
But remember my first post, it's where the main arguments came from.
I'm really biaised since I only play the bga/hgroup way, but it's really hard to come with upsides of your way. Like I said on my other post, playings cards early isn't that good.
Last edited by Romain672 on 07 September 2022, 14:02, edited 1 time in total.
Re: How should clue residue be handled
To make me slightly change my position on it, you would need to come with new ways to save useful non critical cards with the few extra clues you get with your convention.
The obvious way is playing with 2-saves & 3-saves, but then you kinda directly gains clues with your convention you immediately convert into saves in others spots which look fine, I know pianoblook which is a great hanabi player and create lots of stuff really like 3-saves (2-saves are always on on hgroup), so I suppose I underestimate those.
But I don't believe you play with 2-saves.
After a while, when you start to consistently score max score, you start to try to not lose those useful non criticals cards.
The obvious way is playing with 2-saves & 3-saves, but then you kinda directly gains clues with your convention you immediately convert into saves in others spots which look fine, I know pianoblook which is a great hanabi player and create lots of stuff really like 3-saves (2-saves are always on on hgroup), so I suppose I underestimate those.
But I don't believe you play with 2-saves.
After a while, when you start to consistently score max score, you start to try to not lose those useful non criticals cards.
Re: How should clue residue be handled
Agree that it is good to save useful non-critical cards and playing cards early can make it hard to distinguish save clues from play clues.
Perhaps I play with the "wrong" players, but I find that some players over emphasize on save. Sometimes you can defer a discard by clueing a player to play another card. If his play yields him drawing another playable card, this may repeat long enough until his supposed-to-be-saved card becomes playable. So no save clue needed. You can also discourage a player from discarding by not clueing the third player and leave the clueing job to him. These won't work well if your hands are clogged because you're less likely to have another card that is playable and your supposed-to-be-saved cards will reach chop sooner, which is why I try to stay away from clogged hands.
And of course, I am biased. So I may have missed a lot of things here, lol.
Perhaps I play with the "wrong" players, but I find that some players over emphasize on save. Sometimes you can defer a discard by clueing a player to play another card. If his play yields him drawing another playable card, this may repeat long enough until his supposed-to-be-saved card becomes playable. So no save clue needed. You can also discourage a player from discarding by not clueing the third player and leave the clueing job to him. These won't work well if your hands are clogged because you're less likely to have another card that is playable and your supposed-to-be-saved cards will reach chop sooner, which is why I try to stay away from clogged hands.
And of course, I am biased. So I may have missed a lot of things here, lol.
Re: How should clue residue be handled
It's already what bga does (with more experienced players at least), called 'chop management' on my bga guide, and 'tempo clues' in hgroup.
The cases where it's good are a combinaison of: your chop could be useful, that player has no safe action and have a useful non savable card in chop, next card of that suit has been drawn, other's players chop are trash, you have lots of clues, you have lots of saved cards, (it's late on the game & you are fine & only one or two useful non critical cards are reamaining).
On bga, you use one clue to reclue a saved card to make it playable, that's a '0for1', which is bad in effeciency.
With your way, I think number still save the others cards, so it's the same.
You was arguing with the fact i said 'plays aren't that good'. And it's a really hard claim to talk about. It depend of so much, but that's my experience.
The main thing is, converting a card from the state 'playable or useful' to a state 'playable' as a cost in any convention set.
You can have gain related to tempo, ie the fact to make the right player play or discard a card ; but outside of the tempo gain which is situational, the only gain is if the next card of the deck is the next card of that suit (or the next card is playable, and all cards played from it make draw the next card of the suit).
Hope you followed me
But again, there was imo stronger arguments in my first post, main one being that card coule became playable for free later on. So you don't have to convert the 'playable or useful' to a 'playable'. It's done for free.
The cases where it's good are a combinaison of: your chop could be useful, that player has no safe action and have a useful non savable card in chop, next card of that suit has been drawn, other's players chop are trash, you have lots of clues, you have lots of saved cards, (it's late on the game & you are fine & only one or two useful non critical cards are reamaining).
On bga, you use one clue to reclue a saved card to make it playable, that's a '0for1', which is bad in effeciency.
With your way, I think number still save the others cards, so it's the same.
You was arguing with the fact i said 'plays aren't that good'. And it's a really hard claim to talk about. It depend of so much, but that's my experience.
The main thing is, converting a card from the state 'playable or useful' to a state 'playable' as a cost in any convention set.
You can have gain related to tempo, ie the fact to make the right player play or discard a card ; but outside of the tempo gain which is situational, the only gain is if the next card of the deck is the next card of that suit (or the next card is playable, and all cards played from it make draw the next card of the suit).
Hope you followed me
But again, there was imo stronger arguments in my first post, main one being that card coule became playable for free later on. So you don't have to convert the 'playable or useful' to a 'playable'. It's done for free.