Yuku free message boards

First off thanks for the reply, second before I explain what and how I use the information on your page, I want to reiterate that I have no formal backround in mathematics, but do possess a good understanding of odds and percentages and that I have the utmost respect for your abilities and the work that you put in on it.

First, I create an average, just as you have done. However I eliminate Ashby and ARGH, only because their numbers are not consistantly posted. I don't look at the standard deviation of the predictions because each has their own methods of achieving the same goal. I do however look, watch and use the deviation of the prediction average versus the line.

My friend, who worked for the Vegas book, once explained to me that the line is created by public preception. Vegas gives a line that Team A is 5 points better than Team B. Not that Team A "IS 5 POINTS BETTER", but that "THE PUBLIC PERCIEVES THEM TO BE". Oddsmakers want to create a line where 50% bet on the favorite and 50% bet on the underdog. The line is determined by the money bet.

I am assuming that you know all of this. But this is how I view the number (the prediction average). This is the TRUE number. The gentlemen that come up with these numbers are the best, and most respected, in the world at what they do. There is nothing but pure numbers forecasting the difference therefore I view it as the actual difference between the two teams.

The next step I use is the deviation between the line and the prediction average. I also view this in terms of Vegas linemakers in that if the predection average is higher than the line, the line isn't giving enough points to the favorite to cover. Contrastly, if the line is higher than the prediction the favorite has enough points to cover the spread.

The third step I use is the difference between the line and the prediction. This number I chart. For example, through over 600 games if the difference is .01 to .99, the winners are 79-40-3 (or just about 66 1/2%). But breaking it down further, if the number falls between .67 and .79 the winning percentages increase to just over 77% (17-5-3).

The more games, the more I am able to define the playable parameters. Right now out of 603 games, 279 games under the tighter parameters are 199-72-8 (73%).

Tonight (2/7)...these games fell into the parameters:
Maryland (-7) W
E. Michigan (+8 ) W
Ball State (Pk) W
Baylor (+2) W
DePaul (-2) L
S. Illinois (-4) L

I hope this answers your questions. Not being real savvy with computer programming what would I need to be able to merge the stats into something similar to what you have done. My thanks again and look forward to future communications.

Thanks, Doc

I'll just post the picks that fit within parameters and the line I played them at.

526) UMass (-5') L
538 ) Wisconsin (-5) W
543) S. Carolina (+11') W
556) Richmond (+6') L
568 ) Northwestern (+5) W
569) Minnesota (+2) W
572) Michigan St (-18 ) W
574) Wis-Green Bay (-2') W
575) Vanderbilt (+3) T
581) Arkansas (+6') W
603) Manhattan (+1) W

I also have been a little "ballsy" playing a particular "reverse" trend. (The selections that fall within this parameter are hitting 37%. The very worst of this system)...so I pick just the opposite...

552) Rhode Island (-13') L
558 ) Geo Washington (-12') W

I'll try and post these everyday...just for verification's sake.

Later Doc

I'm gonna have to start layin' off the reverse plays for awhile...

Last Night : 9 - 6
Week to Date : 23 - 11 - 1 (68%)
On a per Unit basis (if you track like that) +10.9

Overall record Last night: 18-16-2
Overall for the week: 54-35-4 (61%)

Of the 4 games being played tonight (Just luv the Ivy ) 2 fall into play parameters...

529) Princeton (+8 ) Parameter win ratio ( 89% )
532) Yale (-10 ) Parameter win ratio ( 65% )

Unfortunately I will not be able to post this weekend. Heading out of town for a bowling tournament (if I don't injury myself too badly )...but will be back posting Monday. I will go back into the archives and add the numbers to my chart but will not post wins and losses, since I don't believe in the "hindsight is 20-20" stuff. I hope these threads have been helpful...if not interesting to read . Everyone have a great and safe weekend.

Later Doc

PS. As the number of games grow I will be tighting my selection parameters. It has been suggested that I may take a look at Win %'s of 56% and above. I will look into this as I compile stats on Monday.