SABR-Toothed Triber: Vetting the Other Carlos

It’s not like we have anything else to think about in Cleveland sports, right?  Let’s talk Indians’ prospects and math!

The Indians have had only six starting pitchers this year.  Depending on your perspective, that’s either a promising development or one of many reasons we’re in last place.  Either way, the consistency in the rotation lends itself to some fairly straightforward assessments.  For example, not a single starter has a K/BB ratio of 2:1—typically the benchmark for MLB success.  Masterson leads the team with 1.78 K/BB.  Aaron Laffey (of all people) leads the rotation in K-rate at 8.16 per nine innings pitched.  Carmona leads the starters with 3.69 ERA that he has parlayed into an All-Star appearance.

But I think most of us know that our starting pitching is a bit suspect.  Talbot has posted a sub-4.00 ERA largely using smoke and mirrors: some models suggest his ERA should be nearly a full run higher than it is based on peripheral stats.  Jake Westbrook has stayed healthy, but can hardly be considered a viable front of the rotation starter.  This season has borne that notion out, with a 4.59 ERA and his typical groundball-inducing ways looking fine as a middle-of-the-rotation starter, but a bit questionable as the “ace” of the staff.  Even more, there’s considerable reason to believe that Jake will soon be traded, or at least not signed this off-season, so the questions around him just seem to multiply.  Then there’s Masterson, who’s made some great strides as a starter in the past month, but around whom significant doubts should remain.  Anyone who allows a .325 batting average against lefties compared to only .253 against righties will have to face a considerable amount of skepticism in the rotation.

In other words, we’ve got quite a few question marks moving forward.  Take a look at the following chart:


There’s no one there who looks to be a dominant force in 2011.  Both Carmona and Talbot are sporting unnaturally low BABiPs (batting average on balls in play—typically .300-.310 for groundballers), suggesting that some of their success this season has been a little lucky.  Unless you believe that Laffey can keep striking out nearly a batter an inning as a starter—something he’s never done in his career—we really only have one “strikeout pitcher” in Justin Masterson.  And a 7.32 K/9 is nothing to write home about, especially considering his high walk rate.  There’s just not a lot of upside in this group.

Enter Carlos Carrasco.  Check out these numbers:

2010 (AAA):7.633.352.281.404.380.304
AAA Career:8.562.912.941.024.18

Ignore the ERA for just a second—we’ll get back to that shortly.  Let’s start by looking at his peripherals.  The first thing to notice is that he’s a strikeout pitcher: he’s struck out nearly one batter per inning over his entire AAA career.  Considering that we’re currently dead last in the AL in both strikeouts AND defense, those extra K’s would help immensely.  Second, Carrasco has posted some pretty decent walk-rates: his 2.91 BB/9 rate for his AAA career is better than any of Cleveland’s current starters and his 2010 rate of 3.35 is bested only by Westbrook and Carmona.  (That’s right: Fausto Carmona is walking fewer batters than everyone but Westbrook.  Success!)

So you combine Carrasco’s strikeout and walk rates, and he’s posted a career K/BB rate of nearly 3.00.  Remember that no current Indian starter has a rate even close to 2.00!  Yes.  He could be useful.  Could be…

So what gives?  Why is Carrasco’s AAA ERA above 4.00?  That doesn’t seem to jibe with what we know about pitchers with high strikeout rates and low walk rates.  While I wasn’t able to find Carrasco’s career minor league BABiP, in 2010 it’s a perfectly reasonable .304, so it’s not like he’s getting seriously unlucky with balls finding holes in the infield or anything.

Well, obviously it’s going to come down to the one column of the chart I’ve not yet addressed: HR/9.  For Carrasco’s AAA career he’s given up more than one home run per 9 IP, and during the 2010 campaign it’s jumped to 1.40 per 9 IP.  In layman’s terms, that means that this season Carrasco is giving up a home run about once every six innings.

That’s a remarkable amount of longballs.  I know that Columbus is a hitters’ park, but that’s still a lot, and not all of them have come at home.  Comparing him to the current Indians’ staff is perhaps not fair, since most of our guys are groundball specialists, but no one in the current Indians’ rotation has topped 0.88 HR/9, and Carrasco has allowed home runs at nearly twice that rate.

So for comparison’s sake, I looked up the average HR/9 in the AL.  This season, it’s 0.93 HR/9—or slightly fewer than one home run allowed per nine innings pitched.  This figure is slightly down from the previous five seasons (as are most offensive numbers) bouncing from 1.13 in 2004 to 1.01 in 2008.  In other words, if Carrasco can keep his HR/9 rate at around 1.00, he should be an average AL starter in regard to allowing home runs.  And if he can couple that with a 2.00 to 3.00 K/BB rate, he’ll be an above average pitcher—maybe even a front-end of the rotation starter.

But it’s all going to come down to his home run rates, and these things can be darn hard to predict since they tend to jump around so much.  For example, during Sabathia’s epic 2008 season, he posted a HR/9 rate of 0.68, and was roundly recognized as the best pitcher in baseball (for both the Indians and the Brewers).  This season Sabathia’s HR/9 is 0.93, and he’s just another guy.  Cliff Lee went from 1.57 HR/9 in 2007 to 0.48 HR/9 in 2008, and, not coincidentally, he went on to win the AL Cy Young.  Not only are these swings in home run rates unpredictable, but they dramatically affect a pitcher’s overall performance.

So I decided to run a few simulations on Carlos Carrasco.  I let his strikeout and walk rates vary slightly—from 8.73 and 2.70 on the “best case” end to 7.02 and 3.38 on the “worst case” end.  I held his hit-by-pitch numbers consistent (and low) based on his minor league numbers.  But I let his HR/9 range widely, from 0.81 (best) to 1.40 (worst).  Then I plugged those figures into an FIP calculator to get a feel for what his ERA would be in the various scenarios.  Here are the five scenarios I came up with*:


*If you’re interested in running a simulation with different numbers, leave the Ks, BBs, HRs, HBP, and IP—within reason—in the comments, and I’ll run it for you.


Yes, I recognize that my “worst” scenario is pretty rosy: I still have Carrasco striking out a decent number of batters and walking fewer than Justin Masterson.  It’s just that in looking at his AAA stats, I couldn’t justify making them a whole lot worse.  And in that scenario, he’d still have an ERA around 5.00.

But that’s not what I’m really interested in when talking about prospects.  I like to look at upside, and boy does Carrasco have that.  If he can keep his MLB strikeouts and walks close to his AAA rates, and slightly cut his home run rates, he can be a sub-3.50 ERA pitcher.  This was the sort of upside that the Indians’ front office likely saw when they asked that Carrasco be included in the Cliff Lee deal.

One more note of interest: in the simulations above, I used 200 innings pitched—mostly to make the math easier on me.  But 200 IP is a pretty decent sample size for a season.  Guess what the difference is in home runs allowed between the “Best” scenario in the “Worst”.  It was 13 home runs.  That doesn’t seem like a lot.  Over the course of a 200 inning season, only 13 home runs took his HR/9 from 0.81 to 1.40.  Which just serves to remind us, I guess, how much a home run really inflates a pitcher’s ERA.  The difference between the “Best” simulation and the “Good” simulation was only two home runs!  Everything else was held fairly constant, and the ERA jumped from 3.48 to 3.70!  Two home runs in 200 innings!

Yes, home runs are really important, and if Carrasco is ever going to have success, he’ll need to figure out how to limit them—a skill he hasn’t demonstrated to any degree in his minor league career.  But if he does, we might have a front-end starter on our hands.  If he doesn’t, well, jeez, I don’t really know.  We’ve still got David Huff, I guess.

But for some reason, I don’t think we’ll have to wait much longer to see Carrasco in a Tribe uniform and witness first hand whether he can cut those home runs down.

  • Stinkfist

    That was a long post. I thought it needed some love. You always do a convincing job crunching those numbers.

  • Scott

    So, this is what we have to look forward to if LeBron sticks to what pundits are saying… Excellent!

  • mgbode

    Ok, the numbers support it. Huff bumped from my 2011 projected starters.


    Though I’m not really sold on Laffey watching him pitch.

  • subadai


    Please use your influence in the blogosphere to begin a campaign in the state of Ohio, and beyond, to donate (and match or exceed) the amount of money that, He-Who-Shall-Henceforth-Not-Be-Named (Lebron James), has donated to the Boys and Girls Clubs of America.

    There is no reason that he should be allowed to use his access to the media conglomerates of our nation to salve the pain of the sin he has committed against the good people of Ohio or to implicate the Boys and Girls Clubs of America in that same sin.

    It would be just and right if those of us with the means donated to the same charity with an amount they can afford. Not one of us has the means that He-Who-Shall-Henceforth-Not-Be-Named has, but there are more of us. Our voices should be heard. The people who will continue to LIVE in Ohio should support the Boys and Girls Clubs of Ohio.

    I will make the first contribution of $200.00.

    WFNY, just tell me where to send it.

    I urge all other WFNY posters to ask the same question of our moderators.