[ View Thread ] [ Post Response ] [ Return to Index ] [ Read Prev Msg ] [ Read Next Msg ]

BGonline.org Forums

think me answering on this topic is biased by my statistics but you brought it up.. here are some data

Posted By: Christian Plenz
Date: Sunday, 11 September 2011, at 1:02 a.m.

In Response To: GridGammon dice analysis (leobueno)

only 264 Matches from when I started over more or less seriousely playing on Gird

gained 1700 experience points (my statistics start at old ELO 1800 in June whenh I received log-in details)

only 51,200 rolls

dice stats:

Player below 95% confidence (red bar in XG dice section)

1 Doubles in a row (red)

Entering from the bar: 2 points board (red and passes 99% confidence)

Entering from the bar: Total Enter (red)

(1,3,4,5 points board all below expectation but not red)

Race Dice

Average Pip 8.013 (red)

Double Ratio 14.764% (red and passes 99% confidence))

Big Double Ratio 7.238% (red)

Opponent's Dice:

Doubles in a row: 3 doubles (GREEN)

Race Dice:

Average Pip 8.281 (not green)

Doubles Ratio 17.612% (not green)

Big Doubles Ratio 9.209% (GREEN)

Ok this is only from 51,200 rolls so you can say not enough.

But following would have let me throw my computer off the window and stop playing there if its not only for entertainment and knowing the good reputation :

win: 131 loss: 133

should have won according to XG: 155,9 (59,05%)

PR Player 4,89 PR Opponent 9,17

Somewhere its stated that ca. 1300 players are active on Grid so one of them has to be unlucky ;) Right now I am on a winning streak and the probability for my performance to occur went down from 1/6000 (after 217 matches) to 1/800 (today). Douglas Zare has published some very good articels on the Long Run (Gammon Village) and when reading it yesterday I questioned myself if maybe there must be some other definition..

The 'should have won' is taken from the XG luck (4ply) player account. I as well have exported all matches to Open Office and summed up the total (wasted) mwc derived from the bot analyzis. This result/expectation is closer to the real result by .5% or 1.3% (and my density function is just 1/200 instead of 1/600) if I weight it the same way as the 'Should have won' does (I think this comes from different match lenghts). Example: the luck result taken from the Export is 59.9% but only 155,9 (59.05% are displayed in the player a/c)

should have won: 59.05%

Luck sum of all matches: 59.8%

mwc sum of all matches: 58.54%

mwc weigted as luck in player a/c: 57.81%

Messages In This Thread

 

Post Response
Your Name:
Your E-Mail Address:
Subject:
Message:

If necessary, enter your password below:

Password:
 

 

[ View Thread ] [ Post Response ] [ Return to Index ] [ Read Prev Msg ] [ Read Next Msg ]

BGonline.org Forums is maintained by Stick with WebBBS 5.12.