good job for the +, thrax

storm after using your system for awhile (thanks for leaving it open on the 5k table often), i have come up with the following hypothesis:

risktakers have inflated luck, while conservative players have deflated luck.

since your system squares the luck of each roll, it gives higher weights to outliers. so that if you have a game with 10 average rolls, and one absolutely terrible 5v1 loss, your luck will appear as ~30% instead of ~45%. i think most agree this is a much better system than the current one implemented.

but after a few hundred recorded rolls, i realized my luck was hovering around 40%, even though it felt* like it was closer to 50% (*i understand feelings are perhaps a poor reflection/measurement of reality). then i noticed that some other players' luck, after a few hundred rolls, were hovering around 60%. after a few hundred rolls, this seemed like too large of a discrepancy... how much longer before they would regress to the mean of 50%?

then i realized that the 60% luckers were people who were more aggressive as far as risks were concerned, whereas i am known (in a shameful way) as someone who plays a very conservative style.

as far as outlier rolls go (the ones that will sway your luck calc the most), there are four types:

1) losing a highly probable attack.
2) losing a highly probable defense.
3) winning a highly improbable attack.
4) winning a highly improbable defense.

if all of these happen equally often, then the mean luck should be ~50%. the first two will subtract a ton from your luck, while the last two will add a ton. winning or losing defenses is mostly independent of player personality: everyone gets attacked whether they are conservative or aggressive. losing probable attacks also is mostly independent of player personality.

however, people like me (conservative weenies), hardly ever have situation 3 occur because we hardly ever roll improbable attacks. risktakers, however, are known for negative attacks (2v3s, etc...).

i still love your calc. i think as we all see our !savedluck number of rolls build up, it is very interesting to see what luck everyone has, not because it is an accurate reflection of true luck (i think it partially reflects true luck), but moreso because it is a reflection of how aggressive/conservative a player is.

thoughts?

my thoughts were that you wasted your time writing that entire essay. you suck. i rock. roll 4v5s, they usually win. and then u can jack off to !luck being 67%.

I think you might be right monte. Maybe I shouldn't be recording the luck like that at all, although still more accurate than the one on the stat page.

I was recording mean luck before as (1 / probability). This would give a number between -/+ infinity. 0 being the expected value. This would work perfectly but the problem is converting it to a percent.

bump