What does it really mean to be a 'number five' starting pitcher?

Often in baseball conversations, pitchers are spoken of as being a "true number one," or a "solid number three," or something along those lines. Most recently at Let's Go Tribe, it came up in regards to Trevor Bauer, who was moved back into the starting rotation earlier this season, and has often been described as having No. 1 or 2 starter stuff, but No. 5 starter results.

I don't recall whether I've written this post before, but at the very least it's something I've dug into in the comments before (because it comes up every few months), and I know others before me have written about it at LGT too. It's been a while, though, and I think it's worth revisiting.

One a very basic level, because there are 30 teams, we can consider the 30 best pitchers to be No. 1 starters, pitchers 31-60 to be No. 2 starters, etc., leaving us with pitchers 121-150 as No. 5 starters. Whoever the No. 30 pitcher is, I doubt he's someone who feels like a No. 1 starter, which perhaps means we should make a distinction between "No. 1 starter" and "ace," accepting that not every member of the former group is also among the latter.

The dissonance between what feels like a No. 1 starter and what sort of production actually makes someone one of the 30 best starters explains some part of why pitchers often feel disappointing: They don't need to be as good as most fans think in order to rate as a No. 2, or No. 5, or whichever category they're being placed in.

What sort of production does each group of guys really give you? The first thing to decide is how to get your 150 pitchers. The most straight-forward way is to look at the 150 guys who've thrown the most innings. Those are the 150 guys who've done the most work as starters. We're not quite halfway through this season, and I think that's too small a sample to work with, so I'm going to use all of 2015 and what we've got of 2016 so far. By innings pitched, the top 150 starters are the guys who've thrown 115+ innings during that time.

Now we can take that group and sort them by whichever metric we like. I'll show the range of numbers for each of the five groups, and I'll also show the median because we're not really hoping to have the 30th best pitcher as our favorite team's No. 1, we're looking for someone who's at least in the middle of that group.

ERA+

• The No. 1 starters range from 121 to 220, with a median of 135.
• The No. 2 starters range from 104 to 120, with a median of 110.
• The No. 3 starters range from 97 to 104, with a median of 100.
• The No. 4 starters range from 86 to 97, with a median of 91.
• The No. 5 starters range from 66 to 86, with a median of 79.

Not especially impressive, is it? A starter with a league-average ERA+ rates as a very solid No. 3 starter, and someone just 10% better than average rates as a very solid No 2.

FIP

• The No. 1 starters range from 1.88 to 3.44, with a median of 3.06.
• The No. 2 starters range from 3.48 to 3.90, with a median of 3.78.
• The No. 3 starters range from 3.94 to 4.26, with a median of 4.10.
• The No. 4 starters range from 4.26 to 4.64, with a median of 4.40.
• The No. 5 starters range from 4.65 to 6.12, with a median of 4.92.

Again, what's required to rate as a solid member of any of those groups isn't as impressive as I think most fans would imagine.

There is a significant flaw with using the top 150 pitchers in innings pitched, which is that the top 150 pitchers aren't evenly distributed among the 30 teams, and injuries, midseason call-ups, and things of that nature conspire to keep a number of the top 150 guys off that innings list. If we're trying to find the top 150, it's probably reasonable to drop the innings qualification a bit, and then look at the 150 best pitchers. If you lower the innings-pitched bar all the way to 75, you get 181 starters, allowing you to throw out the worst 31 of them, roughly one per team.

Now I'll redo the groups, based on this larger sampling of pitchers...

ERA+

• The No. 1 starters range from 122 to 220, with a median of 139.
• The No. 2 starters range from 107 to 121, with a median of 113.
• The No. 3 starters range from 100 to 107, with a median of 103.
• The No. 4 starters range from 91 to 100, with a median of 97.
• The No. 5 starters range from 82 to 91, with a median of 87.

The median for each group jumps up 3 to 8 points, which is significant, but anyone with an above-average ERA+ is still a solid option for the middle of the rotation, and someone 10% better than average still rates as a decent No. 2 starter.

FIP

• The No. 1 starters range from 1.88 to 3.36, with a median of 3.00.
• The No. 2 starters range from 3.38 to 3.81, with a median of 3.63.
• The No. 3 starters range from 3.82 to 4.12, with a median of 4.02.
• The No. 4 starters range from 4.12 to 4.34, with a median of 4.26.
• The No. 5 starters range from 4.35 to 4.75, with a median of 4.53.

Again, things change somewhat substantially (especially for the No. 5 starters), but a FIP of 4.00 makes you a solid No. 3, and anything below 4.50 makes you a completely acceptable part of the rotation.

The Indians have gotten almost all their starts since the start of 2015 from six guys. Here is each of those pitchers with their ERA+ and FIP during that time, along with which tier each of those metrics puts them in, using the second set of group I created, with the higher bars to clear:

• Cody Anderson: 98 ERA+ (top-half No. 4), 4.64 FIP (bottom-half No. 5)
• Trevor Bauer: 101 ERA+ (bottom-half No. 3), 4.20 FIP (top-half No. 4)
• Carlos Carrasco: 121 ERA+ (bottom-half No. 1), 3.16 FIP (bottom-half No. 1)
• Corey Kluber: 117 ERA+ (top-half No. 2), 2.96 FIP (top-half No. 1)
• Danny Salazar: 141 ERA+ (top-half No. 1), 3.48 FIP (top-half No. 2)
• Josh Tomlin: 141 ERA+ (top-half No. 1), 4.41 FIP (top-half No. 5)

By ERA+, the Indians have three No. 1 starters, and Corey Kluber isn't one of them. They also have six guys who rate as a No. 4 or better. Things aren't quite as rosy by FIP, but the team still has two No. 1s and six guys who belong in an MLB rotation. Trying to average the two different metrics out, Carrasco, Kluber, and Salazar all rate as No. 1 starters, Tomlin and Bauer as No. 3 guys, and Anderson and a good No. 5.

It's a really good group.