So my assumption on conservative players is probably right. They are likely to have higher luck simply because they don't lose as many rolls. The player that tries to 4v4 connect in round 1 two out of three games will lose those rolls more often than not and be affected with less roll wins and probably more early round flags, which saves the "bad luck" you've had for that short amount of time.

I'd say that the people who have higher luck (49.5% or higher) are probably fairly conservative at the tables.

Not true. Luck is computed on the difference between the result and the _expected_ result.

Conservative players win more of their rolls, but they're expected to win more of their rolls, so the luck would still average out.

What makes the difference is ties, which is why (almost) everyone is slightly below 50% over time.

If a certain player rolled even rolls more often than another, they should see a slightly lower luck percentage over time.

And I realized I partially misread what you said, so my 'not true' doesn't actually apply. However, hopefully what I've written is still useful.

luck isn't influenced by your conservativeness directly (I think)

in other words, if the random number generator is perfect, it doesn't matter if you rol risky or not

Let me try and explain intuitively. To make it simpler, let's forget about defender winning in case of a tie. My reasoning works with that included too but then my % are more complicated. So here we assume that you reroll all dice in case of a tie until the result is not a tie.

Let's say there are only 2 types of rolls to do: a risky roll with only 40% chance of winning and a safe roll with 75% chance of succes.

If the random generator works perfectly, the risky player will win 40% of his attacks. With the formula that luck is calculated currently: you get +60% for a succesfull attack, -40% for a loss so luck%= "50% +1/2*((40%*60)+(60%*-40))" = "50%"

Same for the safe player: +25 for a succesfull atack -75 for a loss combined gives a luck% of "50% +1/2*((75%*25)+(25%*-75))" = "50%"

basically, if random numbers are generated correctly, any type of rolling will yield the expected value for any roll type if you play enough games so every players luck would en up with the same luck%

I am not 100% sure my assumptions are right so if anyone can prove my reasoning wrong, pleas do so because I wanna know what the deal is myself.

Since Thrax is such a coooooool mod who will check with Ryan in a few weeks if my mod opproval is either yes/no, I will say he is right in a way too.

Conservative players should have a slightly higher luck% compared to risky players, simply because they play more turns. If a risky player loses a big roll, he will keep taking the same risks to come back from the key loss. A more conservative player will still play conservative after the same big loss, maybe aim lower , and will in general survive more rounds compared to the risky player.

if say a +3 roll fails with a -0,90 effect on the luck avg, the risky player will average that out with maybe 10 extra rolls before he must flag out whilst the conservative player might be able to average that big loss impact out by 20 extra rolls before he flags

Vermont, don't the expected values already include the results of the tie too? that's why a 2v2 odd is not 50% but 46% or something

if your randomness is correct you should have 0,54*+46%-0,46*54% = 0% influence on your 50% basic luck%?

I don't remember exactly what but I believe the 49,5% had to do with something else? I can be wrong here too

there were two theories about the not50% luck. first, that the luck was not weighted per rounds/game, and since early flag outs are always on horribly bad luck games, then the total luck for the month would always appear low. this has been debunked, and proven that the luck is actually weighted per rounds/game. second, something to do with attacking greys. but i never quite understood that theory. i mean, if the luck is computed against expected luck, then it should average to 50 no matter whom you roll against.

perhaps it has something to do with the same reason that our percentage wins for each place never add up to 100. Ryan either rounds down his fractions, or doesnt use enough digits in his floats or something, and therefore a 49.99% luck game will get reported as 49.9% luck. i dunno.

no, that doesnt make sense, cus then the max effect would be 0.1%, and we are seeing an average of about 0.8% off of 50%.

all i know is the way the plugin does it it differs a bit with the results form ryan.

Someone do my account, my luck sucks right now I need a scapegoat!

oh, i do not mind someone using the fcuku account as a guinea pig, as long as its approved by the powers that be.

do me jurgen!