search

WillFlame14 / hanabi-bot

A bot that plays on the hanab.live interface.
GNU General Public License v3.0
14 stars 9 forks source link

Bluff #240

Open jhgravelle opened 1 month ago

jhgravelle commented 1 month ago

Version (PM the bot with /version): 1.4.5 Convention settings: 11 Steps to reproduce or replay link: https://hanab.live/shared-replay/1173919#5 Additional information: Is bot1s clue a bad bluff or bot2s error not playing into it?

flackr commented 1 month ago

Currently, will-bot2 rules out the bluff because of the visible connection through the b2.

From the level 11 guide I think the last 2 points are most relevant here:

  1. Assume the "move" that results in the least number of blind-plays.

    • In general, this means assuming a one-blind-play Bluff over a two-blind-play Finesse. But keep in mind that only certain cards are legal Bluff targets - you can't just do a Bluff with any card.

  2. Assume the "move" that results in the least number of Prompts.

    • This means that if something could be either a one-blind-play Finesse or a one-blind-play Finesse with a Prompt involved, then the pure Finesse is simpler and you should assume that.

However I'm not sure they give a clear answer the question of whether it is a valid bluff. In particular, the 1's do not need prompting so the decision here is between a delayed blind-play finesse (b2, y2) and a blind-play finesse / bluff (g2 or a bluffed 1).

I tend to feel like this should be interpreted as the delayed blind-play finesse myself and so will-bot1 shouldn't have given the clue.

WillFlame14 commented 4 days ago

This is not a valid bluff because of Schrodinger's Cat Principle: bot2 needs to wait for both b1 and y1 to play before they can play slot 1, which implies that it can't be a bluff due to the delay. bot1 won't give this clue in the next update.