as mentioned above this article is kicking ass xD nice one indeed
Comment by AC_Butcha_AC — January 20, 2010 @ 12:03 pm
1) It’s hard to compare the College game. Sure, the fielders are worse, but balls are also hit, on average, harder than the lower minor leagues in particular, due to metal bats.
2) Is part 3 going to be a list of guys in the minors that fit this mold? That would be pretty cool . . .
Comment by The Nicker — January 20, 2010 @ 12:17 pm
You are absolutely correct about the college game. I believe I said as much in my article that’s linked, but if I forgot, shame on me. Still, I don’t think it’s a stretch to say that college fielders are decidedly worse on average that Minor League, professional, baseball players.
I’m hopeful that Part 3 will be an analysis of the Minor League performance of the guys outlined as my sinkerballer sampler yesterday, and then on Friday, we’ll knock out a list of underrated minor league sinker pitchers using the week’s findings.
Comment by Bryan Smith — January 20, 2010 @ 12:23 pm
This is one of the most interesting articles I’ve read in awhile. I’m looking forward to the analysis down the road.
Comment by Stripesjr — January 20, 2010 @ 12:43 pm
Byan, where can you find batted ball data for minor league pitchers? I can only find sites with GB%.
Comment by Matt Walsh — January 20, 2010 @ 1:23 pm
A friend of mine pointed out the following:
“Wasn’t Porcello forced to give up his CB in the minors and thus a “different” pitcher from a repertoire standpoint? How much can we look to Porcello as an example?”
So what is the benchmark to look for in minor league pitchers in terms of projecting future success as sinkerballers? I use a GO/AO of 2 for a quick and dirty cutoff. That number can be easily found on the player’s bio page on milb.com. I know a ratio of 2 is pretty extreme, but I figure hitters at higher levels will be better able to elevate pitches and you need a high ratio in the low minors to predict that the pitcher will keep a positive ratio in the future.
There isn’t really a great source for the kind of batted ball data we have for Major Leaguers, which is why analysis like this still takes some guesswork. But Jeff Sackmann has built a very nice database at minorleaguesplits.com, and I think that’s the ground zero for batted ball info.
Comment by Bryan Smith — January 20, 2010 @ 1:30 pm
I just assumed this to be true, nice to see it quantified. Great read.
At the end you suggested that the bebwfit that a pitcher like Porcello gains from a jump from the low minors to the MLB (better defense) balances out with the negative result (tougher batters). Would you equally suggest that MLB clubs use minor league sinkerballers more quickly in emergency situations (injured starter) than “rush” other pitching prospects up to make a few spot starts, even if the sinkerballer is in the low minors and the other prospect is polishing his game in the high minors?
I briefly discussed this in the comments section yesterday. Porcello was a different pitcher in Lakeland than he was as a senior in high school. I have always believed what was said about his move away from the four-seam fastball — that he was taught that he could do a lot more damage with the two seamer. I’m just not sure I believed the strictness of the breaking ball restriction that was reported. This year, he threw his curveball 8.1% of the time. I know for a fact that the number wasn’t zero in 2008, because I know some scouts caught a glimpse of it. Are we supposed to discount a year’s worth of performances (which were compiled using the same fastball-slider-change combo that he used 92% of the time in the MLB in 2009) because of a half-reported “restriction”? I didn’t think so.
Comment by Bryan Smith — January 20, 2010 @ 1:35 pm
I purposely did not say this in the article, but I’m glad someone picked up on it. I’m not sure I’d say “low minors”, but I do believe the spot-start pitcher that teams should be looking for to fill a quick void should be given to sinker pitchers more often. I think they are better prepared — and stand more to gain — by the call-up.
Comment by Bryan Smith — January 20, 2010 @ 1:37 pm
Does anyone here know the average percentage of a team’s errors which are actually a ‘Reached on Error’?
Example: a team has 350 Errors at the end of a season. How many of these would be ‘ROE’s?
I guess the percentage is around 85%-95%
If someone out ther knows this that would be awesome.
Best wishes from Germany =)
Comment by AC_Butcha_AC — January 20, 2010 @ 1:39 pm
I sort of do the same, and like you, I don’t know that benchmark. I’m hoping by pointing out that we don’t know this, that we can work towards figuring it out. This series is just the tip of the iceberg in changing the evaluation of these players, I hope.
Comment by Bryan Smith — January 20, 2010 @ 1:39 pm
Still looking for a team is Joel Pineiro, who would fit the worm burner mold and match up best with a good defensive team. Is he being under valued?
Is the value there for a minor league reliever like Josh Papelbon who has career GB%/LD%/FB% splits against RH batters of 61.0/12.1/24.4 with his submarine style delivery?
Would having Seattle’s defense behind him or with Boston’s move to improve theirs give him a better shot at a call up?
It does seem this way. Sean Smith’s being doing work with TotalZone for the minors and he’s noted that there needs to be some adjustments made to TotalZone numbers at any given level if you wanted to make them into “defensive MLE’s.” The adjustments generally decrease as the level increases, which goes along with the path of the DER’s shown here (they increase, meaning defense gets better as we move up).
I’d be interested to see what the difference between a 64% fielding efficiency and 69% fielding efficiency translates into. An extra baserunner over 6 IP? More?
I guess I don’t see how undervalued this cross-section of MiLer truly is. Bergesen was always considered a potential back-end guy, as was David Hernandez and Jason Berken. Hernandez is a hard-thrower that evaluators have downgraded due to limited arsenal (FB/BB) and command issues. Bergesen was a groundball guy evaluators have downgraded due to limited ability to miss bats.
The numbers look nice and make intuitive sense, but I’m not sure there’s much predictive merit we have without a more comprehensive look at the player. Is he getting ground balls on out-of-the-zone contact? Is the movement late enough to neutralize the better strikezone command/pitch-ID of more advanced hitters? Does the pitcher have other offerings that come on the same trajectory, making ID more difficult?
I think it’s an interesting read, and I am eager to see what the next analytical step is. But I organizations seem to already do a pretty good job of identifying which “soft-tossers” are most likely to succeed at higher levels, and standard attrition of MiL arms seems to take care of the rest.
I’m certainly not saying we need to reinvent the wheel with prospect evaluation here, or that this series should affect great change in that regard. But, we need to demand that we (and the organizations, especially) always do better in identification. I certainly agree with “pretty good”, but why not look on the periphery and try to find reasons that we can get better?
I’m not suggesting a huge number of players are being underrated, but I guarantee there’s 20 pitchers in the minors right now that will have some success in the Major Leagues that aren’t talked about in prospect circles. And I believe the job that we’re trying to do is to make sure that every player is talked about and identified before he succeeds in the bigs.
Comment by Bryan Smith — January 20, 2010 @ 2:45 pm
Very fair — and I apologize for not reading the first article prior to posting, as that would have greatly clarified things for me. I’m very interested to see what your work (and others’) ultimately turns up. The object is certainly, as you state, to do the best job possible of identifying future ML contributors. Kudos for pushing the conversation in this area. Creating dialogue is the most important by-product of analytical writing (I think) and it’s always nice to see interests sparked.
Wouldn’t it be helpful if we had ground vs. air defensive efficiency? It is possible to track it from milb’s logs. Maybe Jeff Sackmann’s database could be used.
Good article. Bill James used to write about the Tommy John family of pitchers, and occasionally included right-handed pitchers like Dennis Leonard. In the game of the late 70s-early 80s, controlling the running game for a sinkerballer was somewhat more important than it is today because of prevailing HR and SB rates. .
Comment by Mike Green — January 20, 2010 @ 3:46 pm
Using minorleaguesplits.com, and comparing major league pitcher rates, anything above 50% groundball rates are above average. If the pitcher is inducing 60% groundballs (like Joel Pinero was this past year), he is throwing/pitching exceptionally well. After spending the past (near) year writing about minor league players at Razzball.com, I am becoming more and more suspect of just the high strikeout pitchers. If a pitcher is inducing 55% ground balls and the remaining 45% of balls in play are split between line-drives, fly-balls, and infield flies, the pitcher, although not as hyped as his flame throwing teammate, seems to produce well at each level. By not giving up many homers (due to low fly ball rates), and having good control, these pitchers are reaching the majors a wee bit later but with more success in the beginning.
Hmm, I was already high on Jennry Mejia, but with a k/9 of almost 9, throwing near 100 mph and a 2.5 go/ao rate per milb.com, he seems like a great prospect. Of course, he doesn’t fit the mold of underrated groundballer.
minorleaguesplits has Mejia at 59% over all, over 60% in St. Lucie, almost as high as Porcello’s 64% 2 years ago. Lost his command a bit in AA.
Jon Niese had a 54% groundball rate per mlsplits. I suppose he isn’t much of a sleeper, as he is fairly well-known, but a tad underrated nevertheless I’d say. Type of guy who is especially hurt by what i believe was a pretty poor Buffalo fielding team.
Here’s a potentially naive question: doesn’t the concept that sinker pitchers are underrated contradict the opening quote of the article? The idea is that sinkerball pitchers get lots of ground outs, but fewer strikeouts. If pitchers really cannot influence the outcome of a ball that is put in play, it would be a disadvantage to allow more balls in play vs. strikeouts, and therefore sinkers would be bad. It seems to me that the quote is mistaken.
Not a bad question. Let me try to explain. The reason I think these players are underrated, ultimately, is because I believe we read too much into their hit rates. I think scouts and statisticians can see a guy give up a lot of singles, that lead to a lot of runs, and immediately discount the player. But, at the end of the day, this thought process forgets McCracken’s DIPS theory, because these singles aren’t the pitcher’s fault. He’s done his job — initiated the ground ball — it’s just that his defense didn’t get there.
Yes, you can make up for this just by striking everybody out. And from now until the end of time, the great K/9 pitchers will (and should!) dominate prospect rankings. This is just a call to not forget that there’s another route to success.
Comment by Bryan Smith — January 21, 2010 @ 12:52 am
One thing that surprised me was that at almost every level of the minors and MLB, just about 18% of groundballs went to the outfield for base hits. This would suggest that there’s not much difference in infielder’s range. However, infield hits and ROEs consistently increase as you go lower into the minors.
The second thing was that even though ground ball hit rates were very stable, fly ball hit rates in MLB are only about half of what they are in Rookie ball (17% to 35%). If infielder’s range doesn’t change, why does it for the outfielders? High fly ball hit rates suggest lower vertical angles (more towards LDs and away from popups), but there’s probably a bias towards advancing HR hitters, who hit the ball higher in the air and are easier to catch. So here, the difference is likely in the batters and not the outfielders.
Comment by Brian Cartwright — January 21, 2010 @ 1:48 am
Awesome indeed, Bryan.
Keep it coming.
As for Porcello, he’s going to need to get significantly better next year just to stay the same, as the difference between Polanco and his replacement (be it Scott Sizemore, Ramon Santiago, or someone else) is likely to hurt him, badly. The guy was a rock.
That’s a helluva good question. My completely unsubstantiated theory:
The infield data makes sense without much too much thought. Younger players tend to have better range (but less awareness and positioning) and “clankier” gloves, especially in the lower minors. Hitting grounders for base hits is not a skill, and thus remains constant. Bunting for hits certainly is a skill, although one wonders why hitters would be ahead of fielders early on.
As for fly ball rates, that’s a tougher one. Certainly, tracking the more difficult flies gets polished as a player ages (up until physical limitations set in), but that’s probably not it.
My guess involves higher K rates in the low minors — the lack of plate discipline and adjustment to wooden bats tends to favor the pitchers early on, so fewer batted balls are getting to the outfield. The flies-closer-to-LDs you mention is probably true, so the balls that DO make it out there tend to be hit more solidly, i.e. less can-o-corns and more ropes to the gap.
I’m not sold on that last explanation, though. Could it really be as simple as that? Good high heat and movement that turns into fly outs at the major league level is more likely to be a K in the minors?
As you climb from the low minors up to MLB, the HR% increases while the FBH% decreases. The high HR, low FBH balls have a higher vertical angle, while low HR, high FBH balls have a lower angle. HRs avg about 32 degress, according to Hit Tracker, while the April 2009 Hit f/x sample shows 12-13 degress as the highest rate of hits.
My hypothesis is that if you have two roughly equal batters in A ball, the one with more HRs (higher vertical angle) gets promoted. Therefor the outfielder in MLB and the high minors is not chasing the same type of FBs that the outfielders in the low minors are.
I plan on testing this by looking at the FBH% of batters who have been promoted to see if their rates as batters change or are stable as they are promoted, and if the guys with the higher FBH% tend to not get promoted.
On the GBs, I wasn’t talking about bunt hits. Infield grounders where the batter is safe at 1b can be scored as either an error or an IFH, depending on the judgement of the scorer. Therefor, I like to lump them together as a measure of the infielder’s ‘hands’. What I do know for sure is what pct of infield grounders the defender converted into an out.
Comment by Brian Cartwright — January 21, 2010 @ 3:45 am
Sounds reasonable to me, although guys that have a large FBH% don’t seem to be unfairly looked over for promotion — one thing personnel directors seem to look for in determining minor league promotions is extra base hits in general, and a guy with a higher FBH% is likely to have more doubles and triples to compensate for a lower number of homers.
As for bunts, oops on me. If the same % of ground balls make it to the outfield for hits, but a much lower percentage of grounders are turned into outs (from division I’s 61% to minors 64-66% to majors 69%), then yeah, we’re looking at errors and infield hits of all varieties. The hard charge from third and the throw from deep in the hole at short are difficult, and you definitely see a much lower % of Zimmermans and Furcals in the minors.
Very interesting stuff, Brian. I certainly don’t think you should attribute all the change in FB DER to the batters, but it’s part of it. Your hypothesis about the more powerful players being promoted is generally correct, and let’s not forget that a lot of these players don’t develop power until their mid 20s.
I also think a reason you’re seeing those numbers is backspin. This is a skill that a lot of these players have barely heard of before professional baseball, but one that is preached to them constantly by hitting coaches. Backspin, as my tennis playing comrades can tell you, will make the ball go further than hitting the ball with topspin, and makes home runs more likely. It also, though, will keep the ball in the air longer, so if the ball doesn’t go over the fence, it’s probably easier to catch. I think these players learning and applying backspin as they progress up the minors might be reflected in your numbers.
Comment by Bryan Smith — January 21, 2010 @ 10:08 am
I agree that sinker pitchers are underrated, I guess I just don’t understand how the DIPS theory relates to ground ball pitchers. Does the value of a pitcher who relies primarily on sinkers come from allowing fewer extra base hits, or is it that he will actually have more of the balls put in play turned into outs because those balls are typically hit on the ground, and therefore easier to field?
Also, perhaps you can answer a few of my questions about sinkers. What is the difference between a sinker and a fastball that is down in the zone consistently? And why is it that I have heard commentators say that a sinker has more “sink” when the pitcher’s arm gets tired?
Jeremiah: I think I’m using DIPS theory in direct conjunction with DER, because I’m saying a pitcher that can’t control his balls in play in Low-A is going to have less turned into outs than one that can’t control them in the MLB. But to your first question, it’s the latter — groundball pitchers are commodities because groundballs are turned to outs a lot and become home runs never.
A sinker is generally a mis-nomer, as usually we are talking about guys that get good sink on their two-seam fastball, and command it well down in the zone. Some players grip the ball in a couple different ways, but thinking about my sample, I believe most of those guys just throw “heavy” two-seam fastballs.
The reason you hear about the tired thing is because, with some of the guys, throwing harder flattens the pitch out. If they have nice, easy arm action, their grip and their motion does the sinking for them. If they rear back and throw hard, it flattens out a bit. I’m not sure how true this is (I feel like someone did a pitch F/X study of this, but I can’t find it), but it’s certainly an ardent belief of coaches … and thus, commentators.
Comment by Bryan Smith — January 21, 2010 @ 10:36 am