Yourturnmyturn.com

Posted: Mon Sep 09, 2013 13:12 Post subject: Penalty for wrong guesses

Am I the only one who wishes the penalty for wrong guesses were more severe?

At the moment, it's statistically wise to guess if you have at least a 1-in-3 chance (since at worst you would lose 2x2 points for wrong guesses and then gain 5 points for the third guess, making a net gain of 1 point). This encourages a much less cautious, more hit-and-miss style of play than in the solo computer game on which this was modelled.

I really like more careful, strategic play though.

Any chance of a Minesweeper variant, with perhaps a -5 penalty for wrong guesses?

Posted: Wed Sep 25, 2013 4:41 Post subject: Analysis of guessing in Minesweeper

This may not fit very well with the topic you just discussed, but it may help us understand why the penalty is only two points.

Assuming both players guess at a 1 in 3:

The first player has a 1/3 chance to guess correctly and gain 5 points. (33,33%)
The second player has a 1/3 chance only if the first player misses. (2/3*1/3, 22,22%)

The first player then would have a 1/2 chance if both guesses were wrong. (4/9*1/2, 22,22%)
The second player then would have 1/2 chance of guessing correctly, if his guess is incorrect, then the first player would get the points for the mine. (11,11% each)

This gives the first player a 66,67% chance to gain 5 points under this system and the second player a 33,33% chance to gain 7 points (5 for the mine plus 2 for the extra miss by the first player). If the score is tied and this was the last mine, then the first player would have a 2/3 chance to win.

I believe that the YTMT version of this game was designed to reward courage unlike the BrainKing version (Frog Legs) which reveals the guesses you make to both players, and therefore encourages players not to guess, wasting time. This is why the YTMT version automatically reveals squares that do not contribute any information. If the penalty is say 5 points per miss, then players would not be inclined to guess as the expected values in that situation would be the same for both players. Of course if the score is tied and there is only one mine left, this no longer becomes a problem.

If there seems to be a flaw or error in my logic, then post a reply. I will be happy to respond.

Posted: Thu Apr 11, 2019 19:10 Post subject:

The penalty should be -10, also you don't get to see where they guessed wrong. I am currently in game vs an opponent that is making wildly poor guesses. And I'm being shut out...makes no sense

Posted: Mon Apr 15, 2019 0:37 Post subject:

guessing usually has a break even if 1/5 chance of mine

non guessing moves if uncovering a blank should be parity wise
if uncovering a one, should give lowest chance of mine