Inspired by Low Country Joe's odd devotion to Quality Starts -- I decided to study the issue, based on the premise of Quality Starts, but using a worthwhile measure.
The measure is based on the odds of winning when a team gives up a given number of runs:
According to Baseball Prospectus and using my math, here is the league's winning percentage with each score allowed:
0 100%
1 91%
2 79%
3 69%
4 53%
5 43%
6 31%
7 22%
8 18%
9 11%
10 6%
11+ 0%*
* 11+ is not actually 0, but for simplicity and due to my desire not to give a pitcher points for giving up that many runs, I set it at 0.
Now the key measure is what I call "Team Runs". The formula to calculate this is ER+(9-IP)*4.5, rounded in the mathematically appropriate direction for individual games.
So if a pitcher gave up 3 runs through 6 innings, Team Runs for the game would be 5. 5 runs gives a team a 43% chance of winning, so the pitcher gets 43 points for the game.
Then you just average the score for each game and you get the percentage of games that a pitcher would win with an average bullpen, an average offense and average luck.
Simple enough? I think so. Here are the results for select pitchers:
Santana 62.82
Oswalt 61.30
Carpenter 60.53
Halladay 59.97
Webb 59.24
Zito 55.47
Capuano 52.56
Haren 51.94
Millwood 49.76
Blanton 46.72
Loaiza 46.58
Alternately, using SalB's Pythagorean model, we get:
Santana 65.79
Carpenter 64.61
Oswalt 64.33
Webb 63.87
Halladay 63.07
Zito 59.78
Haren 56.55
Millwood 53.53
Blanton 51.01
Loaiza 50.99
Poll
What do you think?
Great ... (8 votes)
Good ... (11 votes)
Interesting ... (37 votes)
Poor ... (2 votes)
Bad ... (0 votes)
I think I saw something like this at _____ (4 votes)
62 total votes




There are 18 Comments. Load Now.
Shortcuts to mastering the comment thread. Use wisely.
C - Next Comment
X - Mark as Read
R - Reply
Z - Mark Read & Next
Shift + C - Previous
Shift + A - Mark All Read
Comment Settings
Live comment alert: Hide it!
Comments for this post are closed.