They all left the room, but not because the winner convinced her of anything. Ast calculated his own guesses thoroughly, but he had no clear idea on what others’ would risk putting forth for evaluation, and this made him reluctant to state his predictions. Sapp didn’t like letting others’ benefit from his work, so while he were more willing to offer guesses, he preferred to accumulate some amount of data, so that others wouldn’t easily learn what exactly he used as evidence. Lastly, Oriss did stick his head out every time, but he did not want to lose too much points at once (and be totally transparent), so he did not make more than two predictions at once.
OK, so guessing anything at all is optional. And I’m not sure how to score things like ‘the child is not AaBB’, which is a guess, but not very specific. And your guess on the parents—do you score that again after every child? If you get it wrong, are you allowed to guess something else and get it right later on?
Aside from that, given all that you’ve said, it seems like it’s going to be really hard to get any useful information from watching anyone else’s score—you need to model their decision process for which things to guess, in order for their score to tell you anything at all, and those are underspecified.
Also, even Oriss can hide his predictions even better if he makes more than two at once. If, say, he holds off on one for a turn and bundles it with another (I’m assuming that delaying doesn’t help—do you get to ‘guess’ the same thing more than once so there’s a reason to make your parental guesses explicit early?).
I’ll try to rewrite it in a sufficiently specified way (right now I don’t have an idea how). They can’t make totally unclear predictions, or Burr is not going to collect enough points. Oriss is going to lose points after the third round. Burr will try to ‘screw around’, guessing that one parent is AaBb, and will not get or lose any points for it, although he will stay in the game.
Fixed the image (I hope).
They all left the room, but not because the winner convinced her of anything. Ast calculated his own guesses thoroughly, but he had no clear idea on what others’ would risk putting forth for evaluation, and this made him reluctant to state his predictions. Sapp didn’t like letting others’ benefit from his work, so while he were more willing to offer guesses, he preferred to accumulate some amount of data, so that others wouldn’t easily learn what exactly he used as evidence. Lastly, Oriss did stick his head out every time, but he did not want to lose too much points at once (and be totally transparent), so he did not make more than two predictions at once.
OK, so guessing anything at all is optional. And I’m not sure how to score things like ‘the child is not AaBB’, which is a guess, but not very specific. And your guess on the parents—do you score that again after every child? If you get it wrong, are you allowed to guess something else and get it right later on?
Aside from that, given all that you’ve said, it seems like it’s going to be really hard to get any useful information from watching anyone else’s score—you need to model their decision process for which things to guess, in order for their score to tell you anything at all, and those are underspecified.
Also, even Oriss can hide his predictions even better if he makes more than two at once. If, say, he holds off on one for a turn and bundles it with another (I’m assuming that delaying doesn’t help—do you get to ‘guess’ the same thing more than once so there’s a reason to make your parental guesses explicit early?).
I’ll try to rewrite it in a sufficiently specified way (right now I don’t have an idea how). They can’t make totally unclear predictions, or Burr is not going to collect enough points. Oriss is going to lose points after the third round. Burr will try to ‘screw around’, guessing that one parent is AaBb, and will not get or lose any points for it, although he will stay in the game.