Over the years I have read many different types of series previews and to me they have always seemed to be lacking something. After a while I decided I would create something that different that would incorporate what I was looking for and something that would frame the expectations for each game.
Thus my series projection model was created.
The model has gone through several iterations, with minor tweaks to try to incorporate more data. However, this was originally born during the 2010 playoffs because I wanted to get a better idea of what the odds were of the Giants advancing, as I found the “expert” predictions to be lacking.
Here is a brief description of how the model works:
- I start by estimating the runs scored and allowed for each team given the starting pitcher, bullpen, defense and each team’s offense.
- The data used in the projection model is based on the current season’s statistics to date and ZiPS projections, with the weighting shifting more toward the actual stats as the season progresses.
- The estimated run differential is then converted into a projected winning percentage using the pythagorean expectation.
- Then, it’s converted into an odds of winning the game using the log5 method developed by Bill James.
I’ve been using this at Crazy Crabbers for a while now and even used it to help handicap the first round of the MLB Playoffs last year. Hopefully you enjoy this and it becomes a way to get some more A’s-related content here on Bay Area Sports Guy. (Editor’s note: yes, please!)
So without further adieu, here are the odds for the upcoming series between the Oakland A’s and Detroit Tigers.
|3 of 4||25.0%||25.0%|
|1 of 4||25.0%||25.0%|
|0 of 4||6.1%||6.0%|
The odds are remarkably even here, which isn’t exactly what you would expect given where these teams finished last year and the overall perception of each team. The only game where the Tigers are clear favorites is when Verlander takes the hill on Sunday, but in every other game the A’s are slight favorites (in Game 3 the odds are 50.25 vs 49.75 but rounded to 50-50).
There are some caveats with the projection. First, I have assumed Brandon McCarthy will be able to start on Saturday. Hopefully that is the case and it seems to be likely based on what Susan Slusser posted after yesterday’s game. Also, I made no adjustment for Yoenis Cespedes not being in the A’s lineup. But again, per Slusser hopefully he will be able to come back sometime in this series.
The A’s should have a decent chance to take three of four from a very good team, which I am not sure many people would have thought looking at this series before the season started.