The Unwritten Book is Finally Written!
An in-depth analysis of: The sacrifice bunt, batter/pitcher matchups, the intentional base on balls, optimizing a batting lineup, hot and cold streaks, clutch performance, platooning strategies, and much more. Read Excerpts & Customer Reviews
You should coordinate this with MLBAM, as they've done some work doing these mappings already. You may also need to decide how to handle DOB and name changes.
Forecasts should provide confidence intervals; given this, there are standard scoring methods in meteorology that take into account sharpness of projection as well as calibration.
You wrote:
As for calculating the mean, most people will just take the straight mean.
Is this referring to mean number of PAs until you reach the 0.70 correlation point? It'd be better to use random subsets for cross-validation than odd and even PAs, and not any more difficult to code.
-Chris
I was walking through an arcade a few months ago and came across an old _World Series '99_ machine. The Rockies were playing the Cubs, and wouldn't you know it, Neifi Perez comes up to bat and goes down with 3 swinging strikes.
If you're looking at hitting, fielding and baserunning see Cristian Guzman, 2005.
-Chris
Tango, why not just calculate the partial derivatives to find the rate of change of the individual components? It should be simpler and close enough.
-Chris
I'd try fitting a generalized logistic curve. You could do it easily in R using the nls routine. I'd give it a go but I'm away from my workstation for the weekend.
In the .500 universe Tom Tango would be writing articles where he hypothesizes a universe where the players and teams are all unequal.
The Universal Player ID and Biographical Data project (Chris Long) —
You should coordinate this with MLBAM, as they've done some work doing these mappings already. You may also need to decide how to handle DOB and name changes.
Forecast Evaluations (Chris Long) —
Forecasts should provide confidence intervals; given this, there are standard scoring methods in meteorology that take into account sharpness of projection as well as calibration.
Reliability of statistics (Chris Long) —
You wrote: As for calculating the mean, most people will just take the straight mean. Is this referring to mean number of PAs until you reach the 0.70 correlation point? It'd be better to use random subsets for cross-validation than odd and even PAs, and not any more difficult to code. -Chris
Neifi Perez, 2002: Worst Hitting Season of All-Time? (Chris Long) —
I was walking through an arcade a few months ago and came across an old _World Series '99_ machine. The Rockies were playing the Cubs, and wouldn't you know it, Neifi Perez comes up to bat and goes down with 3 swinging strikes. If you're looking at hitting, fielding and baserunning see Cristian Guzman, 2005. -Chris
SuperVORP (Chris Long) —
Tango, why not just calculate the partial derivatives to find the rate of change of the individual components? It should be simpler and close enough. -Chris
Forecasting Playing Time (Chris Long) —
I'd try fitting a generalized logistic curve. You could do it easily in R using the nls routine. I'd give it a go but I'm away from my workstation for the weekend.