THE BOOK cover
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

Buy The Book from Amazon


SABR101 required reading if you enter this site. Check out the Sabermetric Wiki. And interesting baseball books.
MOST RECENT ARTICLES
MAIL : You ask | We say

Advanced


THE BOOK--Playing The Percentages In Baseball

<< Back to main

Tuesday, November 21, 2006

Forecasting Playing Time

By Tangotiger, 11:21 AM

Ever wanted to figure out how many more plate appearances a batter will get to the end of his career?  There are several parameters to consider, with current playing time and age as two key pieces.  Other important information would be the quality of the batter, past playing time, fielding talent, positon, and expansion among others.  What I’m going to show you is how to do a basic forecast using two paremeters (age, and PA in most recent season):

Future PA = [PA]*((67/[age])^2) + (67/[age])^6.7 - 650

For example, a 30-yr old, with 575 PA in that season will forecast for 2436 additional PA to the end of his career.  In reality, all 30 year olds (74 of them) born between 1930 and 1960 who had between 550 and 600 PA actually averaged 2384 PA. 

I will post the actual data on my site soon, and I encourage others to come up with something nicer than mine, a better fit, and does away with possible negative PA.


#1    Tangotiger      (see all posts) 2006/11/21 (Tue) @ 12:28

Here’s the raw data:
http://www.tangotiger.net/files/forecastpa.csv

age: year minus birthYear
PAclass: “150” means “150 to 199”
avgPA: average for that age and PAclass
n: number of players in that age and PAclass
futPA: actual future PA
expFuturePA: based on the formula posted above

Someone do me proud!


#2    Chris Long      (see all posts) 2006/11/26 (Sun) @ 07:49

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.


Page 1 of 1 pages


Name (required)
E-Mail (optional; WILL be published)
Website (optional)

<< Back to main


Latest...

COMMENTS

Feb 11 23:23
Reader Mail of the Day: Why do we need X years of fielding data?  And what about outliers?

Feb 11 22:49
Clutch analogy

Feb 11 22:08
Who is Jeremy Lin?

Feb 11 20:11
Fighting leads to goals?

Feb 11 19:55
Why do players get crappy caps?

Feb 11 19:12
Hero of the month: Brittney Baxter

Feb 11 17:59
MGL: Today on Clubhouse Confidential

Feb 11 10:29
Dwight Evans

Feb 11 02:12
Performance through the ages

Feb 10 23:01
For Your Soul