Semi-bluff ?

[AlexT2067]

Let's say there is a yellow1 already played.

Player A is the clue giver.
Player B have white1 in finesse
Player C have yellow2 in finesse
Player D have white3 in his hand

Can player A clue white3 ?
Player B pov : missing white 1/2, should be both in his own finesse positions, blind play the white1
Player C pov : missing white 2, not in player B hand, blind play yellow2
Player D pov : the white clue trigger two blind plays, should be a white3
Back to player B, the white clue trigger a blind play from player C hand, no white2 to play, end of bluff

[_mg6maciej_]

TL;DR: Yes.



If the player who is to play 2w is immediately after the player who played 1w, it's a bluff.

So like in your example:
Alice: X X X X
Bob: 1w X X X (thinks 1w 2w X X)
Cathy: 2y X X X (thinks 2w X X X)
Donald: X X 3w X <- Alice clues 3 or white (thinks 2w X 3w X if clued 3, or X X 2w X if clued white)

This also works if 1w is marked. For example Bob was clued 1 to a card on chop and thinks it might be 1k, so won't play it without a prompt or another kind of information from others. Note: only works when white clue is used, 3 would result in a bombing 3w as 3y, because from Donald's pov 1w might have played for another reason than clueing Donald.

1w play does not have to be on the very next player for this kind of semi-bluff (also called finesse bluff), for example:
Alice: X X X X
Bob: 3w X X X <- Alice clues white using the last clue (Bob thinks 2w, and has to discard)
Cathy: 1w X X X (thinks 1w 2w X X)
Donald: 2y X X X (thinks 2w X X X)
Similarly to your original example 2y play reveals a bluff and Bob's 3w is known as 3w.

However if there is someone between 1w and 2y play who did not immediately react, creating so called gap, the clue becomes layered finesse:
Alice: X X X X
Bob: 1w X X X
Cathy: X X X X <- did not blind play immediately after 1w, creating a gap
Donald: 2y 2w X X (thinks 2w X X X, is surprised by 2y, but must continue to play next rounds until 2w is found)
Emily: X X 3w X <- Alice clues 3 or white

[Romain672]

I want to add something important, because of this convention, player B must answer the clue and can't clue y2.

In this example:
Played: y1
A: x-y3-y4-x
B: ?-?-?-?
C: y2-x-x-g5
D: w3-x-x-x
A clues white to D.

B here can't clue yellow to A, because by doing so they remove channel information and after that B doesn't know anymore if they hold w2 or not.
That can seem obvious, but it's a huge cost.
When optimising and A give the clue the initial clue, if B hold w1+w2, A must check what would happen with C's hand, and if it's problematic they can't clue w1+w2 anymore (D should have likely helped A one turn earlier).

There is exceptions here, but removing channel information or not answering a bluff have a huge cost outside of 3p games (and is still confusing at 3p), so that stay rare.