April 2, 2009
What’s a Fifth Starter?

If you haven’t read Pat’s piece on the Mets rotation situation, I highly recommend it. Whenever I read anything along the lines of analyzing a rotation, I’m reminded of perhaps my favorite sabermetric article, entitled “How Good Is Your Fourth Starter?” The article was written in 2006 and the author, Jeff Sackmann, attempted to define just how good each starter is. His methodology was taking the best 32 starts from a pitcher in order to calculate the first starter and continue down the line for each team. For example, Johan made 34 starts of 2.53 ERA ball; therefore, the Mets’ “First Starter” would have 32 starts and an ERA of 2.53. The “Second Starter” would consist of two starts of Santana’s 2.53 ERA combined with 30 starts of our second best starter (Pelfrey) yielding a 3.65 ERA, and so on. The reasoning behind the latter methodology is to simplify a team’s rotation into five “Starters” while still using the data from all but two of the 162 starts over the course of a season. The results of Sackmann’s study (again, 2006) were the following:

Lg     #1     #2     #3     #4     #5
MLB    3.60   4.14   4.58   5.10   6.24
AL     3.70   4.24   4.58   5.09   6.22
NL     3.51   4.04   4.57   5.11   6.26
Mets   3.72   3.97   4.41   5.02   6.55

Overall, the rotation put up a 4.73 ERA as compared to 4.70 NL average, making the rotation, well, average. But, despite an average rotation, the Mets were pretty good in ’06, thanks to a great lineup, solid defense, and an outstanding bullpen. But what really stood out to m—and why I loved the study—is just how “awful” all starters are. We as fans see a guy with a 4.00 ERA and we deem him a solid three starter, but in reality he’s a number two. I mean a number four that puts up a 5.11 ERA and a number five putting up a 6.26? It just looks awful, and it’s not just the bad teams bringing up the average either. The Dodgers had a very good pitching staff in 2008, and I calculated their fifth starter as putting up a better but still awful 5.61. Which brings us to Livan Hernandez.

Pat went over the numbers in his article: Livan is pretty terrible. The chart I made seen at the top of the article is also telling, as Livan is hit hard and everywhere. Livan is projected by CHONE to have 4.91 FIP next season. Livan’s fastball has sat at 84 MPH three years in a row. That’s enough: Livan is bad, plain and simple. The only thing he has going for him these days is durability, so the question is this: is a bad pitcher who will always make his starts worth a rotation slot? My answer is a resounding maybe. I can think of two benefits: one, he saves the bullpen, and, two, by being healthy he prevents even worse starters from pitching. To those of you who think there’s no one worse I submit these pieces of evidence:

Year   # starts   Pitcher       ERA
2006   4 starts   Pelfrey      5.48     
2006   5 starts   Williams     5.79 
2006   8 starts   Soler        6.00
2006   7 starts   Perez        6.38
2006   5 starts   Zambrano     6.75
2006   3 starts   Gonzalez     7.71
2006   4 starts   Lima         9.87
2007  13 starts   Pelfrey      5.58
2007   6 starts   Lawrence     6.83
2007   5 starts   Many*       14.60
2008  20 starts   Pedro        5.61

*starts were made by Humber, Vargas, Park, and Williams

I also see two unique negatives with Livan. One, if he pitches alright, he might convince the Mets that they do not need to look for an upgrade in the rotation either via trades or the minors. Two, he’s still a bad pitcher and there is little upside. Additionally, the other problems with the Mets rotation remain as well. Santana and Pelfrey’s previous seasons’ workloads, Maine coming back from injury, Ollie being Ollie. So while I argue that Hernandez has a good chance of being an adequate fifth starter, if something goes wrong with a pitcher ahead of him on the depth chart, he simply is not capable of filling the void.

With all that being said, I should mention that I’m partially biased: I’ve always liked Livan. It’s hard not to like a Cuban refugee that throws every type of pitch with an effortless motion. I’m sorry but watching a pitcher throw a 84 MPH fastball, 77 MPH slider, 74 MPH changeup and a 65 MPH curveball (as well as every pitch in-between) is fun to watch for me. Although, I’ve never watched him pitch wanting him to succeed as much as I will in this upcoming season and if he posts a 6.00-plus ERA again this year, I’m sure I’ll think differently.

7 Responses to “What’s a Fifth Starter?”

  1. Comment posted by Danny on April 2, 2009 at 8:58 am (#943431)

    First off, hilarious chart. Just fantastic. I’m lolling, as the kids say.

    I’m also a little partial to Livan, as I was to El Duque. Their back story is fascinating to me and I love how they attack hitters with deception and movement. When they are on, it’s fun to watch when it’s on your side (and utterly frustrating when your team is the one flailing away, see the Mets against Moyer). And while I always clown on the Innings Eater™ thing, it’s really just because I was horrified the Mets would pay a bounty in prospects to get someone like Joe Blanton. But on a low-contract fill-the-last-spot-in-rotation way, I don’t mind a guy like Livan. The risk, as you mention, is definitely that the Mets stick with him all year, or that one of the 1-4 guys comes up limp and all of a sudden the Mets have 2 below average guys in the rotation.

    The Phillies got lucky with Moyer last year, maybe we can coax a league average season out of Livan and ride those 200 innings into the postseason.

    Fuck it, it’s time to be optimistic.

  2. Comment posted by Kneel Before Zod! on April 2, 2009 at 9:29 am (#943437)

    so boring

  3. Comment posted by Ramon on April 2, 2009 at 10:00 am (#943448)

    LOL! Great article, Joe!

  4. Comment posted by Dave in Spain on April 2, 2009 at 12:11 pm (#943539)

    I love the article! Great to see a dose of statistical reality to offset the unwarrented expectation that our 5th starter is only acceptable if he can hold down a 4.50 ERA. Nobody has that.

  5. Comment posted by littlefallsmets on April 2, 2009 at 6:54 pm (#943843)

    He’s a warm body, he’s a lot cheaper than Pedro’s delusions of grandeur would be and he’ll be healthy enough TO pitch poorly, which is more than we could have said for Mr. Martinez…

    Maybe it all sifts out that in the end the Mets have to just put Niese in there and stick with him, as they are so loathe to do with young players, and the dude will eventually shine.

    If Livan’s inadequacy is the path to that, maybe it’s worth it.

  6. Comment posted by dogcatcher on April 2, 2009 at 11:17 pm (#943927)

    I have long thought that the 5th starter is the most underutilized potential method of improving a baseball team. I think that chart cemented that impression.

    Imagine a normally distrubuted team with a 1 through 5 as the mets have. Lets say Santana, pelf, Ollie, Maine and Livan.

    Now lets say the mets sign Lowe as the new 2, and each pitcher slides down 1 spot.

    Thats an improvement of .25 x 32 games (8 runs) = .44 x 32 games (14 runs) + .61 x 32 games (20 runs) + 1.5 x 32 games (48 runs). That’s 90 runs on the margin…thats like signing Manny ramirez and putting him at 2nd base for castillo!!!

  7. Gravatar
  8. Comment posted by Joe Sokolowski on April 3, 2009 at 2:35 pm (#944286)

    Thanks for the comments guys.

    Dogcatcher, thanks for running that math, that’s some pretty dramatic stuff. I just want to point out that your calculation would assume perfect health from the entire rotation which is, obviously, very hopeful.

    I think the reason most teams don’t improve the staff as a whole is based on supply and demand. There’s simply a lack of quality arms on the market each year. In addition to that, there are payroll constraints. The rotation of Santana, Pelf, Ollie, Maine and Lowe would cost ~54 million and would only go up when Maine and Pelf are eligible for free agency. Devoting that much money to five players is impossible for most teams, and would be difficult to build a team around for the rest sans the Yanks. Therefore, it makes sense to use the fifth spot to save money for other positions while trying to catch lightning in a bottle with retreads or weaning young pitchers into a rotation. Your point still has its merits though.