Fixed some issues
Status: Beta
Brought to you by:
ululuth
Sourceforce downloaded revision 1 when I clicked files and download, spent a good hour fixing all your bugs in rev. 1 and come to find out you have a rev. 2 and rev. 3 - none the less. I -think- your code still has errors, and mine does not. I updated the example link with one that actually works, and applied all the sample data to match their example.
I am not sure why you allow a $M to be passed in. I guess it makes sense if you want it to have a win/loss belong to no one.
Online Diff: http://diffchecker.com/s3870Jtw
View and moderate all "patches Discussion" comments posted by this user
Mark all as spam, and block user from posting to "Patches"
View and moderate all "patches Discussion" comments posted by this user
Mark all as spam, and block user from posting to "Patches"
View and moderate all "patches Discussion" comments posted by this user
Mark all as spam, and block user from posting to "Patches"
Oh, and the main error is according to http://www.glicko.net/glicko/glicko2.pdf in step 3
v = 1 / summation(foobar)
you have:
v = summation( 1 / foobar)
----------------------
Also the delta and first mu_p are the exact same calculation - they both are based around the other player's score (s sub j). You have mu_p based upon the players (s)