The Data Mining Project
Archive for June, 2009
5 Inning Plays
Jun 3rd
There were 2 good questions left in the comments of yesterday’s results post and I thought it was a good idea to highlight them and make a separate post for them since the question is probably on a lot more people’s minds.
Question(s):
What are you doing when the indicators point to a game play rather than a 5 inning play? Are you passing the game or playing in 5 innings anyway, even though it is counterindicated?
Are you only playing 5 inn plays now? And what if a team looks strong for the game and has a strong pen but is going against a team like Wash with a weak BP?
I am only doing 5 inning plays right now. I have never liked my bullpen projections because it is so variable every day. For example, the Dodgers use a very different bullpen when they are up by 1 run vs down by 1 run. Also, high variance games (think LAA and NYY earlier this year) where the bullpen gets shelled skew the current year statistics and make their bullpens look worse than they may actually be.
Right now I am ONLY looking at 5 inning plays. I am not even looking at my full game projections because I just don’t trust them at this moment. If there is a strong bullpen vs weak bullpen I am not taking the full game line of the strong bullpen either, mostly because I still don’t trust the numbers enough to put my money on it yet. As the year progresses, I anticipate adding the full game lines back in, but I haven’t made any real decisions.
I hope this turns out to be a good for our bankrolls and, as of yesterday, it has looked good.
Currently we’re up about 10 units since Sunday. So far, so good.
If you have any other questions, keep them coming.
NBA Finals Projections
Jun 1st
Updated on APBRMetrics.com
| NBA Finals | ||
| Orlando Magic vs Los Angeles Lakers | ||
| Orlando Series Win % | 43.92% | |
| Los Angeles Series Win % | 56.08% | |
These include a lot of season stats with Jameer Nelson and with regular home court advantage (HCA is higher in the playoffs) so I estimate more of a 35% chance for Orlando (maybe a tad lower), but then again I haven’t run the numbers.