To start this blog, I wanted to begin a series looking at how starting pitchers age by tracking a variety of statistics by age, over a fixed period of time. The time period I’m looking at is 1955 to 2012, with starting pitchers’ ages 20-40. I chose 1955 as the cut off because that is when IBB started to be tracked, allowing me to calculate FIP per age over these years. The data is just raw right now, and I’m certain that normalizing it may help a little since I am looking at very different run scoring environments, but for now, I’ll just use the raw data.
Games Started
The first metrics are very simple and will give us an idea of how much pitchers are pitching in the league at a given age. This way we can find the peak(s) and see when it is reasonable to expect a pitcher to be most prolific in innings pitched and games started. The first graph is looking at the games started per age:
From the graph, we can see that the true peak is actually at age 25 with 23,567 games started; however, age 26 looks to be very close at 23,539 games started. From this, we can reasonably infer that the most games started per age group is around 25-26 years old. Yet, this doesn’t tell us that much because there are probably more pitchers pitching in those age groups, resulting in more games thrown. The graph below confirms this:
The peak is now clearly at 25 years old, with a steady decline after. The rates of attrition nearly double every 5 years after the age 25 peak. Below are the rates of attrition for various age groups:
| Age Group | # of Players | Player Attrition | Attrition Rate |
|---|---|---|---|
| 26-30 | 5651 | -751 | -13% |
| 31-35 | 2253 | -512 | -23% |
| 36-40 | 543 | -227 | -42% |
Innings Pitched
I would expect the innings pitched per age to have the same distributions shape as games started and number of players per age. If more games are started in an age group, naturally one would expect more innings pitched for that age group. Here is a graph illustrating the total amount of innings pitched per age group:
In this chart (right click and open in new tab/window to enlarge), we see that the most innings pitched per age has shifted to age 26, indicating 26 year old pitchers are becoming more efficient and pitching longer into games. This also makes sense. As a pitcher matures, he gains the strength and experience necessary to pitch longer into games. I then looked at the number of IP per game started for each age group to find at what age a pitcher is pitching deepest into games:
This chart tells a little bit different of a story. Pitchers are most efficient if they are very young, or as they mature around ages 30-31. I’m guessing that the higher IP/GS for ages 20-21 are because pitchers those ages are usually phenoms who can last longer in games based on elite pitch quality alone. Once the pool of pitchers enlarges at ages 22-23, then IP/GS drops significantly. I think this is due to the pool of pitchers at 23 years old includes many good, but not elite pitchers. And from there we see a steady increase in IP/GS until age 30. After age 37, the data shows that pitchers drop sharply until what looks like an outlier at age 40 (we are dealing with a really small sample of pitchers at age 40 too).
Another thing I noticed when looking at the data is that older pitchers are still pitching as many innings per game started as younger pitchers, and in some cases even more. So, I decided to look at the average number of innings pitched per age group, and found that older pitchers are generally pitching more than younger pitchers in a season. 
This perplexed me because my thought was that older pitchers break down and pitch less. I also looked at the average number of games started per age group and found that older pitchers (over 30) start on average 3 to 6 more games than under 30 pitchers. I think some of this is due to bias – the older pitchers are still in the league at age 35 are good, and therefore pitch more. I also wonder if younger pitchers get injured more often, or sent to the minors more than an established veteran who had “earned” his place in the rotation, regardless of pitching quality. I think I’ll need to look into this more. There could also be a generational bias, in that before the days of pitch counts and 5 men rotations, older pitchers pitched more often.
Conclusions
The distribution of starting pitchers show that most starting pitchers in the league are ages 24-27, which is the age group that starts the most games. However, that age group is not the most efficient in terms of innings pitched per game started. I think a big factor in this is that due to attrition starting at age 26, the least talented pitchers get weeded out. And by age 30-31, we are looking at a group of pitchers who have beat attrition, and gained experience and strength to last longer into games. From there, we see a slight decline likely due to aging and loss of “stuff.” And finally, older pitchers are still pitching as many, if not more, innings per game started as younger pitchers.
Working the Count 