Rookie Phenomenon… Huh?
Posted by Bill Baer on Friday, June 19, 2009 at 8:31 pm
I was linked to a blog entry by Rob Neyer (thanks, BBTF!). In it, Rob talks about Mets broadcaster Gary Cohen claiming that nearly 15% of all starts made by pitchers this season have been from rookies, more than at any other time since “rookie” became an official adjective in the baseball vernacular. The discussion at BBTF attempts to answer Rob’s question, “Why all the rookie starters?”
Good points were made throughout the discussion, but that 15% kept eating at me — it seemed too high. So, I went to Baseball Reference’s Play Index and I found all the starts made by rookie pitchers in each season dating back to 2000. For transparency, I’ll tell you exactly how I got my numbers.
For the Play Index, I went to the “Pitching Season Finder”, set my years from 2009 to 2009 (repeated this for previous years, such as 2008 to 2008), selected the player’s First to First season of his career, and further sorted by the total amount of starts made that season. Lastly, I added a qualifier (under “Select Additional Criteria to Match”), which was GS greater than or equal to 1 (to remove the events involving rookies who did not start the game).
I found the following data:
If my data is correct — and there’s always the possibility that I slipped up somewhere — then Cohen’s claim is false and this year is not unlike years past.
Maybe Cohen meant the number of rookies who have made starts instead of the number of starts made by rookies (there’s a subtle, yet significant difference). Let’s have a look.
Closer to Cohen’s number, but as you can see, this year’s crop of rookies has accounted for the fewest percentage of total starts in the entire decade.
There is no rookie phenomenon, at least not based on the data.
Not sure if this would make up the entire difference but wouldn’t a pitcher who had a cup of coffee in 2008 not show up on your list because 2009 would be the second season of his career? For example, I would assume your first list above does not include Josh Outman of Oakland, who pitched 25 innings in 2008. But I believe he would still be considered a rookie and therefore his 11 starts in 2009 could be considered “rookie starts.”
Jeff,
That’s a really good point — thanks for that.
However, since the criteria is the same across the board, I don’t think it causes any kind of a significant difference in data. Additionally, I’m unaware of any other way to filter out rookies using BBRef’s Play Index.
EDIT: Just got your whole point (yeah, I’m slow): if I could find a list of some sort of rookie-eligible pitchers who have started games this season, it would be very easy to calculate the percentage, but you’re right — it could be enough to get up to the percentage Cohen cited.
Still, it’s the lowest of the decade, so it’s not the story it’s being made out to be.
Oddly enough, I used ESPN’s stats page to break out Rookie Pitchers from all others and Rookie Starts only added up to 161. One of the tools is obviously inaccurate but what’s interesting is that in the case of comparing this number to B-R’s Play Index, the number actually goes down not up.