Right, that should not be named 'self color bluff', but 'self color known bluff'. I will propose that on discord/github.Blacktango wrote: ↑27 September 2022, 17:30 Oh, fair enough. I didn’t think about it like this.
I guess the "bluff" part of this technique’s name doesn't suit me, and didn’t help me to understand it.
Maybe I would call it "self canceling reverse" or "anti reverse" instead, or something like that, because no one is bluffed here.
Layered vs. reverse bluff
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: Layered vs. reverse bluff
-
Blacktango
- Posts: 433
- Joined: 18 April 2015, 12:15
Re: Layered vs. reverse bluff
Hello,
What do you think about the following situation ?
- ravingbeauty’s 1 is known as trash.
- kaptekop just gave the 2 clue to ravingbeauty.
What should happen if I clue 1 to kaptekop ?
What do you think about the following situation ?
- ravingbeauty’s 1 is known as trash.
- kaptekop just gave the 2 clue to ravingbeauty.
What should happen if I clue 1 to kaptekop ?
-
Travis Hall
- Posts: 180
- Joined: 12 April 2020, 14:13
Re: Layered vs. reverse bluff
Raving Beauty plays 2k, Gunthario may play 1k (but is also likely to clue to get the multis moving; if you’re using Flamboyants, it would be a good idea to try to get 3m and 4m played quickly for your best chance of retrieving 5m).
I see no reason to believe that anything fancy is going on. Why? Is there a reason you think there is something complex about this?
Frankly, I’m not convinced that 1k is a great clue in the situation. White for 2w gives Gunthario options (since he already has 5r to play for a flamboyant any time he wishes) and even if a blocker 1 is drawn, you can still clue 1k as a save. Or you can clue multis to let Gunthario play 5r.
Re: Layered vs. reverse bluff
So firstly, that 2 clue is for k2, so raving got a play.
Then r1 clue promesse b1/k1 in another hand after k2 play.
I guess we are in the 'out of position bluff' I talked on that subject. Ravingbeauty got negative b1/k1 in her whole hand, so should imo assume the interpretation with goes well with the lowest number of blind-play/play on her hand. Which is... w2.
So raving plays k2.
Then Gunthario play w2 as b1/k1, and r1 is known trash.
Nothing else happen.
And like Travis Hall said, just clue w2 instead.
[[[Hgroup's convention have no out of position bluff for simplicity, but if they got some, that would be a nice example of Unnecessary move (level 23, pretty high: https://hanabi.github.io/docs/level-23/ ): since you got a clean clue on w2 and choose instead something complicated, it promesse one extra blind play. Here by default it should 'push' the card to the left of r1 which is r3 and make w2+r3 playable ( https://hanabi.github.io/docs/level-23/ ... trash-push ).
But since r3 isn't playable, it start to be complicated even for me. I think raving should play her third finesse position to tell the card is trash (first would be like g3+w2+g4, second would be like b5 (following logic of 5 color ejection, off in bga too: https://hanabi.github.io/docs/level-15# ... ection-5ce ), and third would be trash). But it's not possible here. r5+w2+trash would be a non conventionnal option, but r5 is globally known.
From Gunthario's perspective, he doesn't know if his slot 1 is clueable directly, he could hold w2+w2+y4+r5, and so that clue make sense, so he can't do any weird stuff.
So yeah, even if you play with hgroup's unnecessary moves, I don't see Gunthario ever play w2+k1 even with non conventionnal stuff and full trust.]]]
Then r1 clue promesse b1/k1 in another hand after k2 play.
I guess we are in the 'out of position bluff' I talked on that subject. Ravingbeauty got negative b1/k1 in her whole hand, so should imo assume the interpretation with goes well with the lowest number of blind-play/play on her hand. Which is... w2.
So raving plays k2.
Then Gunthario play w2 as b1/k1, and r1 is known trash.
Nothing else happen.
And like Travis Hall said, just clue w2 instead.
[[[Hgroup's convention have no out of position bluff for simplicity, but if they got some, that would be a nice example of Unnecessary move (level 23, pretty high: https://hanabi.github.io/docs/level-23/ ): since you got a clean clue on w2 and choose instead something complicated, it promesse one extra blind play. Here by default it should 'push' the card to the left of r1 which is r3 and make w2+r3 playable ( https://hanabi.github.io/docs/level-23/ ... trash-push ).
But since r3 isn't playable, it start to be complicated even for me. I think raving should play her third finesse position to tell the card is trash (first would be like g3+w2+g4, second would be like b5 (following logic of 5 color ejection, off in bga too: https://hanabi.github.io/docs/level-15# ... ection-5ce ), and third would be trash). But it's not possible here. r5+w2+trash would be a non conventionnal option, but r5 is globally known.
From Gunthario's perspective, he doesn't know if his slot 1 is clueable directly, he could hold w2+w2+y4+r5, and so that clue make sense, so he can't do any weird stuff.
So yeah, even if you play with hgroup's unnecessary moves, I don't see Gunthario ever play w2+k1 even with non conventionnal stuff and full trust.]]]
-
Travis Hall
- Posts: 180
- Joined: 12 April 2020, 14:13
Re: Layered vs. reverse bluff
Oh, wait, 1 clue to kapkekop, not Gunthario? That can be read as a trash bluff, which means raving beauty could respond with a blind play. Her only blind play is 5y, which misfires.
I don’t see enough value in that to make it worth the risks. It would be so easy to misread. Raving could misfire, Gunthario could think that playing once was enough, Gunthario could just think it isn’t for him because he’s not in bluff seat. Technically it could get two plays, but strategically I don’t think it’s worth it.
I don’t see enough value in that to make it worth the risks. It would be so easy to misread. Raving could misfire, Gunthario could think that playing once was enough, Gunthario could just think it isn’t for him because he’s not in bluff seat. Technically it could get two plays, but strategically I don’t think it’s worth it.
Re: Layered vs. reverse bluff
It wouldn't make sense to play 5y. Trash bluff (any bluff) means you should play the card that makes the clue valid (i.e. not bomb). A 1 clue to kapkekop would be calling for 1b, and 5y cannot be 1b, so it shouldn't be "called" by a 1 trash-bluff clue. It should cause either 1m to bomb (if it's not as clear that it's trash or not blue), or Gunthario to play 2w blindly (and stop there, not assume layered).
-
Blacktango
- Posts: 433
- Joined: 18 April 2015, 12:15
Re: Layered vs. reverse bluff
Ok, then it would be just a distant trash bluff (not a layered one).
I agree, if the clue doesn’t trigger both W2 and Bk1 to be played, there is no gain to such complexity.
I guess I thought at the same time :
- raving knows they can’t answer and won’t, then Gunthario will
- raving didn’t answer, so Gunthario will know it’s a layered and will play two cards.
Which can’t be both right.
I agree, if the clue doesn’t trigger both W2 and Bk1 to be played, there is no gain to such complexity.
I guess I thought at the same time :
- raving knows they can’t answer and won’t, then Gunthario will
- raving didn’t answer, so Gunthario will know it’s a layered and will play two cards.
Which can’t be both right.
Re: Layered vs. reverse bluff
In this situation, for me 1 clue to Kap should just trigger a 5y misfire. This is just because 1r is clearly a bluff clue, and the usual convention is that the very first person that can reveal a bluff has to reveal it (I just strongly disagree with what Jellby has written).
A more interesting situation would happen if Gunthario was holding 1b leftmost...in fact, if u have the "trash finesse" in your convention package, then a bluff is no more necessarily revealed by the very first person that can reveal it, but it can be revealed also by some other person next to it...if this holds a card "connected" to the marked one. So in this case, adopting such convention, clueing 1r would trigger no bomb by ravingbeauty and would get Gunthario's 1b played.
Personally, I find this convention nice and potentially useful...but I have never seen it used on bga.
A more interesting situation would happen if Gunthario was holding 1b leftmost...in fact, if u have the "trash finesse" in your convention package, then a bluff is no more necessarily revealed by the very first person that can reveal it, but it can be revealed also by some other person next to it...if this holds a card "connected" to the marked one. So in this case, adopting such convention, clueing 1r would trigger no bomb by ravingbeauty and would get Gunthario's 1b played.
Personally, I find this convention nice and potentially useful...but I have never seen it used on bga.
Re: Layered vs. reverse bluff
Yeah it's called a 'trash finesse' in hgroup: https://hanabi.github.io/docs/level-13/ ... sh-finesse and I saw it talked more and more on bga.
What you are talking about was debated on the whole subject. You know you don't have the connecting card. So you can either play anyway as a 'known bluff', or choose the interpretation the most likely which make things work (usually that mean that your next player have a playable in finesse position and it can be gotten instead).
The problem with that second possibility, is that if you play with layered finesses, it can be really hard for new and average player to differenciate the cases where it's a layered finesse and when it's an 'out of position bluff'.
And I think that's the reason why h-group would strangly assume a 'known bluff'.
Remember that with that layered finesse/out of position bluff you need for all players to be agree on what is going on, else it cause one bomb.
ie: Start of game. First player clue red on r2 to forth player. Third player have b1 in finesse position and r1 in second finesse position.
If Bob plays it's no.
If Bob clue something unrelated it's yes.
What you are talking about was debated on the whole subject. You know you don't have the connecting card. So you can either play anyway as a 'known bluff', or choose the interpretation the most likely which make things work (usually that mean that your next player have a playable in finesse position and it can be gotten instead).
The problem with that second possibility, is that if you play with layered finesses, it can be really hard for new and average player to differenciate the cases where it's a layered finesse and when it's an 'out of position bluff'.
And I think that's the reason why h-group would strangly assume a 'known bluff'.
Remember that with that layered finesse/out of position bluff you need for all players to be agree on what is going on, else it cause one bomb.
Jelby, are you playing with layered finesse on your games?
ie: Start of game. First player clue red on r2 to forth player. Third player have b1 in finesse position and r1 in second finesse position.
If Bob plays it's no.
If Bob clue something unrelated it's yes.
Re: Layered vs. reverse bluff
The reason why (I believe) I wouldn't bomb 5y is because I know I don't have the card, and I know the person who gave the clue knows that I know it, and I see it could be a bluff for the next player, and I have a play and no need to save.
I assume there could be layered. But sometimes it's a mess because a clued play gets delayed when the player thinks it could be a layered finesse for someone else...Jelby, are you playing with layered finesse on your games?