Indians rotation preview (part 3)

by APV on Mar 5, 2009 9:20 PM EST Comment 32 comments

Jeremy Sowers

We are all painfully familiar with Jeremy Sowers at this point. Top draft pick thought to be a short trip away from bigs even his ultimate ceiling was not the caliber of some prospects. And that was pretty accurate, as he debuted in Cleveland with quite a bit of success in 2006. That success was built on good control (5.6 BB%), decent HR rates (1/9IP, 52 GB%), and more than a little luck (.259 BABIP, 4.57 FIP, 3.57 ERA). Since then the picture has been decidedly more bleak. He has coupled strong performances at Buffalo (FIPs under 3.65) with disastrous performances in Cleveland (FIPs over 5).

Sowers fundamental problem seems to be that he simply doesn’t have a very good fastball. I don’t think this is primarily a velocity problem. Last summer he showed the best velocity of his professional career (89.7, up from 85.4 and 88.5), but the results were obviously not great. The club has talked a lot about the separation between his fastball and change – essentially that his changeup isn’t slow enough to keep hitters off balance – but I’m not sure I totally buy that line. Like just about every pitcher, Jeremy’s fastball is his primary pitch (69% last season), used far more frequently than either his slider (16.9) or change (14.0). Despite throwing far fewer of each of them, Jeremy actually got more swinging strikes in 2008 off both his slider and his change than he did off his fastball. Only about 10% of Jeremy’s fastball strikes were of the swinging variety. That’s bad (Laffey, not exactly a stud fastball guy, got 17.5% swinging strikes). I don’t know if it is a deception issue, or the need for more movement, or the need for more separation with his secondary pitches, but I think we should all cheer every time Sowers gets a swinging strike off his fastball. We don’t have pitch f/x data for 2006, but Sowers did use his fastball more in 2006 than he has in any season since. But batters didn’t really get any more or less contact or swing any more or less frequently at it in 2006 than in 2007 and 2008. I think there is good reason to be pessimistic about Sowers unless he starts getting better results, particularly swings and misses, from his fastball.

Scott Lewis and David Huff

Lewis and Huff are the next two reasonable options for the 5th rotation spot (I would be very surprised if Zachson gets serious consideration as a permanent member of the rotation). Both Lewis and Huff have been very successful, when pitching, in the minors. Scott Lewis had a successful four-game debut last season in the bigs – although that success had a lot of luck on its side, in the form of a .225 BABIP.

Lewis is a flyball pitcher (GB < 40% in 2007 + 2008), which means in the absence of a lot of luck he is going to give up HRs at the big league level. Because of this, his success is going to be predicated on keeping guys off base as a way of minimizing the damage. In the minors, Lewis did a great job of this by showing exceptional control (minor league career 5.6%), very good strike out rates (24.8%), and by giving up less than a hit an inning. These are great numbers. Along with concerns about his health, the knock on Lewis has been that his stuff won’t translate against better competition (apparently an 87mph FB doesn’t scare scouts). I say at this point you have to give Lewis the chance to fail. But Lewis is still going to have to couple good control with a decent K-rate…better than he showed last September.

We’ve got no major league numbers on Huff, but he was exceptional last year. Very solid K-rates (~25%), very good BB-rates (~5%), and decent GB-rates (52%). That’s a lot of good. Scouts don’t like him for the same reason Lewis doesn’t get talked about, they question his stuff. I think in Huff’s case the scouts are wrong. Huff has gotten better as he has advanced – I have a lot of optimism about what he’ll do in the bigs at some point.

1 recs | Comment 32 comments | Share

Story-email Email Printer Print

Comments

Nice write-up. I’m guessing that there is something of a consensus among us here, simply that Huff > Lewis > Sowers, whether this year and over the next few years. (Seriously, anyone disagree?) Laffey’s a little more difficult to place there, and Jackson is Columbus/Cleveland filler. Not that there’s anything wrong with that.

by dgcambridge on Mar 6, 2009 11:45 AM EST reply actions 0 recs

I think you have to put Laffey on an equal footing with Huff, based on the success he’s already had, and Laffey is eight months younger, too. Jackson does have some upside but is probably what you say he is.

Well, that's just, like, your opinion, man.

by Jay on Mar 6, 2009 11:56 AM EST up reply actions 0 recs

Agreed. I give Laffey the edge over Huff right now because of higher floor, despite lower ceiling.

by Ryan Kelsey on Mar 6, 2009 12:27 PM EST up reply actions 0 recs

I don’t disagree, but to be fair you’d have to go back to when each was on the cusp of pitching in the majors – ie, having a good AAA sample.

Sowers put up great #’s that pushed him to the big leagues, where he did well initially. Huff is at that point now. If you lined each up at the point where they were to make their first ML start and compared past performance to project future performance, it might seem a little different.

by mcrose on Mar 6, 2009 1:11 PM EST up reply actions 1 recs

Not if you look at K/BB. Huff is way ahead of Sowers, even by your proposed standard.

Well, that's just, like, your opinion, man.

by Jay on Mar 6, 2009 1:30 PM EST up reply actions 0 recs

Not so much at the time, when they were all poised for the ML step. Of Sowers, Laffey, Huff and Lewis, they all had good control, and Lewis probably had/has better K/BB ratio than Huff. Sowers at that point had struck out over 7 per 9 during his minor league career, etc.

My main point, of course, is that we haven’t actually SEEN much of Huff outside of his stats, and unlike Sowers and Laffey, he has yet to success or failure at the ML level. They have had success, then failure where they had to make changes (still trying), while Huff is still bright and shiny.

I’ll admit, due in large part to watching the previous guys, that Huff may have a better arsenal to fortify him for ML duty, but I really don’t see a huge spread within that group when they were all “so fresh and so clean”.

by mcrose on Mar 6, 2009 3:19 PM EST up reply actions 1 recs

No. Annual p46. Huff had drastically better K/BB in 2008 than Sowers at the same point in his career — same age and level. SLewis’ K/BB is not much better than Huff’s.

I’m not looking at Huff’s “sunny-ness,” I’m looking at his performance.

Well, that's just, like, your opinion, man.

by Jay on Mar 6, 2009 3:44 PM EST up reply actions 0 recs

His performance in the minor leagues. I understand that the ability to strike people out is a great advantage for a pitcher. But so is putting the ball where you want to. How the ball moves is another. There are several ways to be an effective major league pitcher, some with narrower margins for error than others, but being able to repeat what works is crucial no matter what.

I just don’t think the initial success of Sowers and Laffey was a complete illusion. When they could control their pitches and execute what gave them great results in the minors, they had ML success.

To me, Huff is in a similar situation to them as they were poised for the final step. I have no idea if he will have initial success, and still less if he will be able to stick to what works. He’ll get his shot this year I’m sure.

by mcrose on Mar 7, 2009 9:54 PM EST up reply actions 1 recs

I have not been dismissive of Laffey and Sowers’ initial success in the majors. I don’t dismiss it at all.

Fact is, though, Huff’s K/BB numbers are far, far betters than there’s were, so the “similarity” of their positions basically ends with ERA. K/BB is far more important than ERA in terms of projecting success in the majors, and therefore, Huff appears to be likelier to succeed in the majors than Sowers was, both in the short term and over the long haul.

Laffey doesn’t lend himself to the same kind of direct comparison, partly because he got to the majors at the youngest age, but mostly because he’s an extreme groundballer, so one can’t blithely apply the normal K rate standards.

Well, that's just, like, your opinion, man.

by Jay on Mar 7, 2009 11:26 PM EST up reply actions 0 recs

It makes no sense to lump Laffey and Sowers together, in any context. They are very different pitchers and their success is predicated on a different approach. Of the two, I think Laffey’s initial success had more substance behind it.

Here’s an interesting set of numbers. Sowers BABIP has hovered around .300 whether he has been at AAA or MLB since 2007. In AAA, Sowers has been very good, posting FIPs under 3.70. In the majors he has sucked, posting FIPS well above 5. Despite the similar BABIP numbers, what batters have done with those balls has been very different. In the minors, Sowers has kept most of the damage to singles, allowed SLG% well under .400 in both seasons. In the majors batters have slugged nearly .500 off of him over the past two seasons. Sowers seems to be a great example for that mysterious “stuff not translating to the majors” phrase. Laffey not so much. Laffey has always maintained his best skill, inducing ground balls, regardless of whether he is in the minors or the bigs.

by APV on Mar 8, 2009 2:35 PM EDT up reply actions 0 recs

I like when you know stuff.

Well, that's just, like, your opinion, man.

by Jay on Mar 8, 2009 11:18 PM EDT up reply actions 0 recs

Of course there’s a context to compare and contrast them. The context is transitioning from a successful minor league pitcher to a servicable (at least) major league pitcher, and that’s the only context I’ve been writing about in this thread.

Hell, Clay Bucholz and Phil Hughes fit into that context, and the scale of difference between them and Sowers/Laffey/Huff/Lewis as far as type of pitcher is dimensional.

Read from the top – the first response to your excellent writeup postulates a consensus that Huff is better than Lewis is better than Sowers. I’m not part of that consensus because Lewis had a relatively insignificant cup of coffee and Huff and hasn’t shown me anything other than his Akron and Buffalo boxscores (euphemism for online perusal of minor league splits data).

And like I said, if you hypothetically had all our young pitchers mentioned above toeing the ML rubber for the first time in ‘09, with their minor league ascent and success the most recent impression on our malleable brains, do I think it would be an obvious consensus on this site who would end up more successful in ’09? No, I don’t – and that’s all I was saying.

I have no doubt that you and Jay, for example, might sift through the available data at that point and post a great screed on why Huff edges out his cohorts, but if you (say, right before that post), had a site poll predicting IP, ERA, and yes, wins for each, the results would be a little more mixed.

by mcrose on Mar 9, 2009 12:07 PM EDT up reply actions 1 recs

I have no doubt that you and Jay, for example, might sift through the available data at that point and post a great screed on why Huff edges out his cohorts

Well, they keep doing that. They’ve pointed out the differences in K and BB ratio several times now. That’s what we’re looking at. (And treating Laffey different because of his significantly different GB rates.) If you have something to counter that, I’m all ears.

And even though it doesn’t change things here, I will say it again, the relevant question is “Who do we think is going to be the best pitcher this year?” (or career) It’s not “Who was the better prospect at an equal point?” (which is a fine trivia question) We don’t throw out the new data on Sowers.

by dgcambridge on Mar 9, 2009 12:55 PM EDT up reply actions 0 recs

So, we’ve finally figured it out! All you have to do is look at the K/BB ratio and you know who’s the better pitcher and who will have the better ML career! Wow, I don’t ever have to actually watch a pitcher again…

So, we’ve nailed it – Lewis> Huff>Sowers>Laffey. Oh wait, if you’re a groundball pitcher, you’re in a different category. What is the cutoff for that? At what point does groundball ratio lessen the importance of K/BB? How would Carmona have ranked in the above hierarchy if you hadn’t seen him but looked at his AA and AAA stats when he was getting his first taste against ML opposition?

I think we’re shooting at different targets somehow. I’m merely pointing out that going from the minors to the majors is a big step, and whether you’re a control lefty with an 88 mph heater or a flamethrowing righty, figuring out ahead of time sight unseen who has the physical and mental makeup to successfully make that jump, not only to have immediate success but to, when facing failure for the first time, make changes or stick to what brung him, actually is a little more mysterious than boiling down minor league stats.

by mcrose on Mar 9, 2009 2:00 PM EDT up reply actions 1 recs

I’m up for discussing it further. Just quickly, I was wondering if there was consensus on how were projecting these guys going forward, which is not the same as saying that anyone knows exactly what the results will be. Think of them as real people or as statistical probablities – but with way, nothing is certain. If your point is that the uncertainty levels are high (“from the minors to the majors is a big step”), I don’t disagree. But we don’t have the luxury of giving all three guys a full year; decisions have to be made now.

Anyway, it’s not all about strikeout and walk and groundball ratios, but those are important. They are some of the better predictors we have for minor league success. Sowers isn’t striking out enough people, and isn’t forcing an usual amount of groundballs, so his odds of success are low, and that assumes he gets back to an insanely low BB total.

My point was simply that Huff ranks ahead of Lewis who ranks ahead of Sowers, based on the data we have. If you disagree, great, let’s hear it. If there’s a makeup issue, or some mechanical issue that we don’t know, share. But I don’t accept that we have to throw our hands in the air and say that we have now idea. (not that you are necessarily saying that). I just thought it was strange that you were asking for a breakdown, after Jay and APV pointed out that the breakdown is there to see.

All of your groundball questions are good ones. It’s tricky, which is why I didn’t put Laffey in my original lineup. It’s going to be the subject of discussion for some time.

by dgcambridge on Mar 9, 2009 3:00 PM EDT up reply actions 0 recs

I would rank them that way right now too. Sorry if I’m coming across like I’m missing something, I’m just saying that, if that’s a clear consensus right now, its in large part due to Sowers’ inability to sustain the success he had after what seemed to be a successful transition to the major leagues.

We could throw other pitchers in the mix, but to continue the point let’s focus on Sowers vs Huff. At what point would you and, say, most posters here, agree on a clear preference for Sowers over Huff for the fifth spot? Right now you would, and I agree I’d give Huff the nod.

But not because I trust him so much, I really don’t know how he’d fare, and I’ve only actually seen him throw a few pitches. I just know he hasn’t failed yet, am very curious and hopeful to see how he’ll do, and the biggest reason is that Sowers isn’t very good right now. Known recent ML suckiness vs hope for a continuation of proven minor league success.

Let’s stagger the timelines and say that Sowers was the same age and he and Huff anchored the Buffalo staff second half of last year with their respective pre-ML AAA stats intact, both vying for the fifth spot this year (let’s say that was the only spot open). The consensus might be a little mixed at that point. An analysis of the stats would give the edge to Huff because he misses more bats, but it would be close enough for other factors to come into play. Who’s more mature, who has the better health record, and of course, since its a competition, who has more effective starts in these ST games.

Stagger the timeline again, so Sowers has his first ML year under his belt. He showed much the same attributes that marked his success at AAA – controlling his pitches, keeping hitters off balance, being intelligent and consistent with his game plan. At that point, you could still wish he had better stuff and more K’s, but it would be hard to give Huff, despite coming off a stellar season at AA and AAA, the edge in the battle for the fifth spot.

And that’s really all I was pointing out in my first response – as much as it seems obvious right now what the pecking order might be, it might be less obvious if you compare them at different points in their careers.

by mcrose on Mar 9, 2009 11:28 PM EDT up reply actions 1 recs

I’m not sure what the point of your sarcasm is. Are you … what, ridiculing us for being more knowledgeable than you are? Or gloating about your lack of understanding on this topic? It has, in fact, very well established that K/BB makes a huge difference in terms of being able to carry run prevention forward from the minors to the majors; and for that matter, ERA is pretty flimsy compared to FIP. Yet you persist with this rough-equivalence argument that is based wholly on ERA (from what I can tell). Have you actually even bothered to look at the peripherals?

There are actually pretty clear cutoffs for extreme groundballers. There are many very good pitchers who exceed 50, but only a handful who approach or exceed 60. Look at this table, and I’m sure you’ll have little difficulty figuring out the extreme groundballers.

60-65% — 2
55-60% — 1
50-55% — 4
45-50% — 10
40-45% — 13
35-40% — 7

If you graphed this, you’d see a nice bell curve. The median is 44.9%, the great majority of established starters are within 5 or 6 points of that number. And then, way over on the right, you have the extreme groundballers.

We may be getting to a situation where the Indians roster has distorted our perceptions a bit. Extreme groundballers are pretty uncommon; it just doesn’t seem that way because we’ve had at least one — and sometimes three — for the last five seasons running. When we ended 2007 with Westbrook, Carmona and Laffey all in the rotation, that was truly a historic level of extreme groundballitude. Having one of those guys is unusual, two is quite rare, three completely unheard-of. In 2008, there weren’t three qualifying starters with groundball rates that extreme in the entire American League.

Well, that's just, like, your opinion, man.

by Jay on Mar 9, 2009 4:58 PM EDT up reply actions 0 recs

Neither ridicule nor gloating. I just don’t think you can isolate one metric and use it across the board to rank pitchers. Nor hitters, for that matter.

As far as groundballs, I have no problem understanding what % makes someone a groundball pitcher. I was pointing out that if you were to treat them as exceptions to the K/BB rule, what % would be equivalent to a given K/BB rate? Are there other metrics or attributes that are exceptions, that can make a pitcher as effective as one with a better K/BB?

Sowers had a GB and K rate fairly similar to Carmona in their pre-ML AAA stints. Are they the same kind of pitcher?

And FIP is a great stat since it tries to be fielding independent. Perhaps that’s a better overall measure, regardless of the type of pitcher. Again comparing AAA just prior to ML exposure, it would be hard to see much daylight between, say:

Sowers 102 IP 3.23 FIP
Laffey 99 IP 3.18 FIP
Huff 82 IP 3.14 FIP
Lewis 75 IP 2.55 FIP

I’m not anti-Huff, pro-Sowers or stat-Luddite – just saying it has yet to be shown that Huff is clearly a better pitcher than the others.

by mcrose on Mar 10, 2009 12:32 AM EDT up reply actions 1 recs

Neither ridicule nor gloating. I just don’t think you can isolate one metric and use it across the board to rank pitchers. Nor hitters, for that matter.

But no-one is doing that. Who are you talking to?

by Voltaire on Mar 11, 2009 3:20 PM EDT up reply actions 0 recs

dg and Jay, obviously.

I made a rather general point that even though right now it might seem that Huff is better than Lewis is better than Sowers, that might not seem so clear if you compared them at similar points in their careers.

Jay’s responded that, no, Huff’s K/BB pointed to a clear difference. I replied that… well, heck – read the thread! Maybe you have an opinion to share on the subject.

by mcrose on Mar 11, 2009 3:43 PM EDT up reply actions 0 recs

You can’t boil it down to one metric, but if you were going to look at only one, it’s probably K/BB. When Adam and I devised the Progress Score formulas, it was very difficult to come up with something we liked to summarize the production of the pitchers. We incorporated K rate, BB walk rate, GB rate, Games Started, ERA and neutral-FIP. For hitters, the production formula is basically just neutral-OPS. And yet for all that extra effort, I think Adam would agree that we’re happier with the hitter scores than the pitcher scores.

Anyway, Huff ended up with the highest Progress Score of any of our prospects, 10.2, and none of our other pitchers scored as high as 5.0. Maybe if we ask real nice, Adam will figure out scores for Sowers-2006 and Laffey-2007. (Adam, if you do this, remember to give them credit for all Games Started including in the majors, but use minor league numbers only for everything else.)

Well, that's just, like, your opinion, man.

by Jay on Mar 11, 2009 9:01 PM EDT up reply actions 0 recs

Yeah, I can see how hitters would be easier – and I guess if you boiled down my point, that would be it, that there are more factors and you’d have to decide how to weight them. Different emphasis, different result.

I’d certainly be interested if Adam wanted to do that exercise. What was Lewis’ Progress Score, out of curiosity?

by mcrose on Mar 11, 2009 10:28 PM EDT up reply actions 0 recs

Lewis’ was 4.9 last season. He gets hurt by only starting 21 games (injured to start the season), and even more for having a sub 40% GB%.

I’ll have the Sowers 06 and Laffey 07 numbers shortly. My guess is they’ll go Huff > Laffey > Sowers…my guess (I haven’t looked at the numbers yet).

by APV on Mar 11, 2009 10:52 PM EDT up reply actions 0 recs

So I was a little off…and this also brings out some of the ways i’d like to tweak the formula for the future.

Laffey 2007: 11.7
Huff 2008: 10.2
Sowers 2006: 8.4
Lewis 2008: 4.9

Laffey gets a bump for being very young (22.3), having an excellent GB rate (62+), and also putting together his best K (18.5) and BB (5.6) numbers that season.

I like Laffey a lot. I like guys who do at least one good thing exceptionally well (in Laffey’s case, induce ground balls). I’m still more positive about Huff, though. Huff does everything pretty well, and has a K% better than any on this list (24.9…Lewis is second at 21.7), and striking out guys is pretty much the best thing you can do.

by APV on Mar 11, 2009 11:02 PM EDT up reply actions 0 recs

adding…Laffey’s 2007 really was awesome. I talked about it then and I still reference it. It was the kind of season that raises the baseline projection for a prospect.

by APV on Mar 11, 2009 11:06 PM EDT up reply actions 0 recs

Interesting results, thanks. Have you guys posted the formula you use? Not trying to pry anything out, just don’t know if I missed it.

Anyhow, you mentioned tweaks – are there other metrics you’d include, or just change the weighting of some you’re already using?

by mcrose on Mar 11, 2009 11:44 PM EDT up reply actions 0 recs

We posted it, but Lord knows where. The basic idea for hitters and pitchers is this:

Progress score = Production + Level + Position – Age + constant

Level is the same for both hitters and pitchers — 6 for AAA, 4 for AA, 2 for High-A, 0 for lower than that. We’ll sometimes assign 1, 3 or 5 based on split seasons.

Age is the player’s age as of July 1 of that season, with one decimal point. (More precise than the usual whole-number “seasonal age.”)

Constant is just an arbitrary but consistent number added in so that most prospects end up somewhere between 1.0 and 10.0. It’s 10 for hitter and 20 for pitchers.

Position is a positional value score from 0 to 4 for hitters. For pitchers, it’s just Games Started divided by 5. This awards bonus value to starters over relievers, and it also rewards pitchers who have been healthy and more durable, which is a huge issue for pitchers specifically. The asterisk, as noted above, is that you do want to give the pitcher credit for major league starts in this part of the formula, if he had any, since the goal is to measure durability.

Production for hitters is neutral-OPS divided by 60. (Yeah, really, it’s that simple. You could mix in NetSB, but the effect would be miniscule.)

Production for pitchers is:

[(K100 – 2*BB100) / 3] + [GB100 / 5] – [2*(neutralFIP + ERA)]

K100 is K per 100 PA.
BB100 is BB per 100 PA.
GB100 is GB per 100 PA.
neutralFIP is the park-neutralized and luck-neutralized FIP as reported by minorleaguesplits.com

So anyway, yeah, we like Huff.

Well, that's just, like, your opinion, man.

by Jay on Mar 12, 2009 1:58 AM EDT up reply actions 0 recs

the biggest difference in the numbers between Huff and Lewis isn’t in K% or BB, but is in batted balls. Huff gets a lot of grounds >50, Lewis gives up lots of fly balls (GB% <40). Given equal K and BB numbers, I’ll take the guy with GB tendencies every time. Also Huff has more speed on his fastball – which isn’t the only thing, but it is a thing.

by APV on Mar 9, 2009 2:55 PM EDT up reply actions 0 recs

I agree with Jay, it’s a moot question here: the order is the same either way.

The question of approaches is interesting. I know it doesn’t seems fair," but I don’t think you do throw out Sowers ML numbers, even if we don’t have comparable numbers for Huff et al. If the goal is to pick the best starter now, we use all the data we have, no?

by dgcambridge on Mar 6, 2009 1:42 PM EST up reply actions 0 recs