Now that the trade deadline is almost here, baseball fans are excited to see who might be dealt, and what prospect might be had. Invariably, people are looking for the "impact bat" or the "ace pitcher." When we talk about an ace, though, people can have different definitions of the word. For some people, it might just be the best pitcher on their team:
"Bud Norris is the ace of the Houston Astros."
This implies that there are exactly 30 pitchers in baseball that qualify (and this season, Verlander isn't an ace because he's not the best pitcher this year).
For other people, it's just the 30 best pitchers in baseball.
"I wonder if the Orioles might deal for ace pitcher Eric Stults."
For most fans and pundits, though, ace is more of a nebulous definition. There are generally only a few players that qualify (say, as few as 5 and as many as a dozen), and the criteria is a blend of stuff and results. I thought it'd be interesting to use a few parameters and try to actually quantify "ace" material.
There are two main factors to determine an ace: you have to be durable and you have to be great. How do we quantify those numbers, though? First, I looked at the innings totals for 2012. I defined (and will define) durable as being 15th or better among innings pitched leaders (for 2012 [my focus for this article], it was 211 innings). That would mean that (for the era), you were as durable or better than the average pitcher that was the most durable on his team. The problem here is that you aren't always replacing those innings a starter doesn't pitch with replacement-level value. If you make 32 starts and pitch 160.1 innings, the innings you DON'T pitch are picked up by your bullpen (which may or may not be replacement level). To get around that, I made the following adjustment:
Any inning before 188 innings (the average number of innings a starter with 32 starts would log) was taxed at replacement-level starter FIP (which is 5.12), and any inning in between 188 and 211 innings is taxed at the reliever-average FIP (which is 3.79, actually).
So, what FIP actually defines greatness? I'd say it's whatever makes average look replacment-level (that is, whatever FIP carries an expected win percentage of 62% when set against the league average). In 2012, league-average FIP was 4.01. That left the value you'd need at 3.14. So, if you can provide 211 innings of 3.14 FIP baseball, you're an ace. You don't have to do exactly this: 199 innings at a 3.1 FIP is still an ace (the 11 innings you lack are made up for by the slightly better FIP you sport)
HOWEVER, this doesn't address something important. Every inning over 211 innings that a player pitches is another inning that your bullpen doesn't have to pitch. So, you should get a credit for pitching more innings than 211. In that case, you should be credited the difference between the pitchers' FIP and the average reliever (again, 4.01).
So, who was an ace in 2012? I have 8.
*was an "ace when healthy (not enough innings to provide necessary FIP)"
That makes sense to me. The two closest non-ace pitchers were Wade Miley (194.2, 3.15), who was almost good enough FIP-wise but 16.1 innings short, and R.A. Dickey, who pitched a ton of innings (233.2!) but just wasn't good enough in those innings (3.27 FIP, still very good).
Let's take this a step further. What defines a #2? Let's say it's 45th in innings (so the average of the second best pitcher), which for 2012 is 191 IP. Furthermore, we don't expect him to make the league-average starter replacement level; let's make him 56/44 instead of 62/38. That brings the FIP to 3.55. If we plug that in, we get the following "true #2"s:
That makes sense to me too. There are 8 true aces, and about 18 number 2s. The list isn't perfect, but I think it's very close and a worthwhile distinction to make (it also puts Shark into perspective).
One last thing; there is definitely an quality to an ace pitcher that is consistency (being excellent over a multi-year period). I'll perhaps look at that later; this is more of a snapshot sort of thing.