Free Allegiance Forums

FreeBeer wrote:QUOTE (FreeBeer @ Aug 22 2009, 08:23 PM) Dear Dr. MrC,

First time caller, long time listener. I have this nagging problem. It seems that my moo has regressed linearly and my sigma, well, let's just say I can't get it up. Do you think Viagra is right for me?

See thats the way Free tells me to relax, and your being a bit of a wiener cake. Now you know why I got a man crush on him

Hey! Where's my diagnosis? You're not supposed to diagnose yourself, wiener cake.

FreeBeer wrote:QUOTE (FreeBeer @ Aug 22 2009, 08:33 PM)

Hey! Where's my diagnosis? You're not supposed to diagnose yourself, wiener cake.

Oh right right

Viagra and a dance from Ethyl should do the trick

Ah.. ye olde "Saturday Night Special". Gotcha.

Well, a linear regression basically finds you the best straight line that fits a collection of points. The problem is that rank is not a simple linear function of time. Instead, rank is a function of skill, and skill, although somewhat correlated with time played, will be different for everyone. There will be episodes when people's skill gets worse and episodes when it makes some unexpected jump.

Like they say in the bottom of a company's financials, past results are no guarantee of future performance and although perhaps in the short term you can draw a line that says "I've gotten 0.3 points since last May, so if this rate continues I'll be the next level in a couple months". You can't extrapolate that over the infinite horizon, to say "at this rate I'll be better than aarm in three years". If your skill has been going down in the short term, perhaps you should be predicting what day you'll be hitting 0 again!

I suspect that the best curve is either logarithmic or one that has some asymptote, every person has their limit beyond which they'll never get better, something like rank = 15 - 100 / (time + 80).

On another note, sigma is clearly inversely correlated with time played, because sigma essentially captures the uncertainty that one's rank is known, and that uncertainty goes down the more you play.

cashto wrote:QUOTE (cashto @ Aug 22 2009, 09:24 PM) On another note, sigma is clearly inversely correlated with time played, because sigma essentially captures the uncertainty that one's rank is known, and that uncertainty goes down the more you play.

sigma can and does go up.

Fwiffo and i spoke on TS about both subjects and other then the above I think your right Cashto

thanks all for input and suggestions. yea, i got a chance to talk briefly about the process w/ mrc and it helped me a lot in understanding what's going on. the long and short of it, regarding ranking predictions, is that it's pretty tough (practically impossible) to predict rank changes since it's depended on not only how well you do in the future but also how well everyone else does also. it's like trying to predict who's playing on what team in order to factor in their effect on your rank.

instead, we can keep track of things like "this month's leading drone killer" which is not only easier to do, but a heck of a lot more accurate!

my head exploded somewhere between the 2nd and 3rd posts, but its good to know there are maths ppl behind the scenes making sure the miner dies if i put x amount of gat bullets into it!

MrChaos wrote:QUOTE (MrChaos @ Aug 22 2009, 08:14 PM) sigma can and does go up.

Hey, I figured that out!

(it happens when you're doing, say, consistantly good and then you have a small stretch where you start doing really poorly. Alleg skill is like "bwuh? does he suck now? I dunno." Then when you start getting back to being good, Alleg Skill goes "ah, glitch in the data. Why'd you go giga exp in a squadgame anyway?"

you use gats?