Guys, I was wondering about something. I just tested a system for NBA totals that hit around 56.5% over 600 picks. What would my percent error be?
| Poster's Sportsbook Poll: 2011View Poll Results | ||
| # 1 5Dimes | 450 total points | 5Dimes Review |
| # 2 Pinnacle | 408 total points | Pinnacle Review |
| # 3 Heritage | 227 total points | Heritage Review |
| # 4 Bookmaker | 138 total points | Bookmaker Review |
| # 5 BetIslands | 129 total points | BetIslands Review |
| SBR Top-Rated SportsbooksRecommended List | ||
| Pinnacle Sports | SBR Rating A+ | Pinnacle Sports Review |
| 5Dimes | SBR Rating A+ | 5Dimes Review |
| BookMaker | SBR Rating A+ | BookMaker Review |
| Legends | SBR Rating A+ | Legends Review |
| Bodog | SBR Rating A | Bodog Review |
-
11-08-05 07:54 PM #1
Testing a System
SBR Founder Join Date: 8/21/2005
-
11-08-05 08:40 PM #2
it would probly depend on your system.
SBR Founder Join Date: 8/10/2005
-
11-08-05 08:48 PM #3
Assuming you're dealing with out-of-sample data:
For a sample size of 600, you'd need a success rate of 55.93% (335+ correct) or better to be confident at the 95% level that you have a profitable (52.38% or better) system. At the 99% level you'd need 57.17% (343 correct), at 99.9% you'd need 60% (360 correct), and at 99.99% you'd need 61.17% (367 correct). With 339 correct there is a 2.37% chance the system is not profitable.
The 95% confidence interval for the true success rate is 52.67% - 60.50%.SBR Founder Join Date: 8/28/2005
Reply With Quote