Hi everyone.
I've been working for quite some time on an NBA sides model, which is based on team's efficiencies (as its core), and reflects other potentially significant variables, such as rest, injuries and other situational factors.
I decided to give it a go for the remainder of this season as it has shown some promising results in the past.
The stakes of the system bets will be based on one of the fraction kelly criterion, probably rounded to integers or integer + a half. It will not be units, rather a certain fraction of a current bankroll. I know it can devastate one if the edge is miscalculated, but if you are able to sustain similar winning rate from the sample in which it was backtested, you will be far better off betting kelly (or its fractional derivative) than flat units. I might write more about this tomorrow.
The first bet is chicago +3. The line movement does not favor me but i will try to not defy the model.
2011-02-24, Chicago +3 (-107), 5,35% to win 5%
Let's have a nice run to end the reg. season