juventino wrote:This is my contribution to the age old discussion about the points and scoring system. My solution is planed to work sidy by side to todays scoreboard. As an extra feature and statistic.
I have tried it my self for 100 games and I think it is a fair way to look for skill.
The first thing I do is to divide every game type. I have made it in 1v1 games for my self since it was the easiest and quickest (I am lazy).
Then I calculated my winnings and losing. ( I have another 50 games with a 90% win rate but I havent done the calcuations)
Wins 82 %
41 Games
Avarage gain: 10.15
Avarage opponent: 1159
Total gain: 416
Losses 18 %
9 Games
Avarage loss: 39
Avarage opponent: 1171
Total loss: 351
Gain over 50 games: 65 or 1.3 points per game.
The score according to my system would be 2487. It is my ""break-even" when if I play i statisticaly wins 0.0 points per game.
Of course this cant be done so quickly. There must be a sufficient amount of games to make this system fair. Perhaps should there be a lower limit of 100 games to make this type of score.
But in this way you take into calculation the opponent too.
interesting concept BUT it has a BIG disadvantage. it is too damn specific.
judging by your math you'd be a player worth 2487 point BUT ONLY in 1v1 games on a certain map with the same settings against players averaging 1100-1200 points.
if you play against people averaging 2000 points your winning % will surely drop but your gains would increase and perhaps you'd be worth 2800 points.
then if you switch other data like the number of players, the map and so and play at least 100 games with each setting you'll end up with probably 40-50 thousand games (or even more, don't have time to do the math) and in the end you'll get results that will probably be very very different.
let's say in 1v1 with certain settings and opponents you'll be worth 2800 points.
on 3 p terminator no card adjacent games on doodle with players of 2500 or higher you'll be worth 1600
and in assassin 6p games flat rate freestyle unlimited fog of war on age of merchants you might be just a simple cook.
so you see this is a great way to find out what game type is your best and what would be your value if you played only a certain game type but it's impossible to do this kind of experiment because it involves many thousands of games.
could somebody do the math for all the possible combinations?
“In the beginning God said, the four-dimensional divergence of an antisymmetric, second rank tensor equals zero, and there was light, and it was good. And on the seventh day he rested.”- Michio Kaku
