In April of '09 Kelley became a perceived wunderkind for the bullpen. Then, he got hurt. When he returned, his Ks and walks were still solid, but his return brought a bunch of dingers that weren't there in early. This Spring, a couple of quick jacks is enough (for me) to raise the question -- "does Kelley have a gopher problem?"
This is perhaps the biggest issue with young pitchers - yet tends to be the one that gets the least screen time. EVERY one of the BOR guys last year that didn't work out had gopheritis. The guy that defied his eye ratio numbers, (Snell), didn't.
Mind you - I'm not making a prediction here. I'm just saying -- Kelley doesn't have remotely the MLB track record to be certain of anything. Yes, he might have what it takes to be an SP. He also might not have what it takes to stick in the pen -- IF his late '09 gopher problems are more than a transient blip.
IMO, the perception of how great Kelley *IS* (versus how good he "might" be) is likely overstated. Felix had oodles of talent from day one, but it's taken him a few YEARS to fully harness that talent. What was his problem a couple years ago? Homers. The analysis was that he was over-challenging. REGARDLESS of the "why", the end result is that Felix tweaked his game, and the HR problem vanished, and he's a CY arm.
Well -- Felix is not the only guy capable of correcting a HR problem. Plenty of pitchers have done it in the past. And plenty of uber-talented pitchers have also FAILED to do so even years later.
========
As for 2010? I think moving Kelley at this point would be galactically stupid, if you have any actual desire to make the playoffs during 2010. The double hit of moving a bullpen guy into the rotation - where he will not be ALLOWED to pitch deep - (or throw 180 innings) -- is going to put major pressure on a bullpen that wasn't that great in 2009 -- and that is topped by a pair of guys with ONE good season each.
I really hope Aardsma and League are for real. But this is not a case where either one is a "sure thing". Both are 1-year wonders at the moment. Both could succeed. Both could fail. One thing is certain. The offense is not good enough to recover from a pourous bullpen, (should it develop).
Then again - the track record thus far for Z has been far better with hitter judgement than pitcher judgement. Vargas, Olson, French are Z's choices ... and Aardsma.
Personally, I think Vargas has a decent shot at "developing" into a decent BOR SP. It's all about the HRs.
.
Q. How much worse would Kelley be as a starter?
A. He ain't gonna post no 8 blinkin' K's a game, I can tell ya that. Nor will he walk 1+ men a game.
Relieving is easier than starting, a lot easier, and if Shawn Kelley were really going to strike out 8.0 men and walk 1.8, well... it reminds of when Jim Lefebvre put Bill Swift into the starting rotation for a couple of weeks. Swift fired three games like 1, 0, 1 ER.
Reporters asked hopefully, "if he keeps pitching like this, will he stick in the rotation?" Lefebvre replied brightly, if he keeps pitching like THIS, he'll be in the Hall of Fame.
If Kelley were to finish the 2010 season with 8.0 strikeouts and 1.8 walks, he would outpitch Felix Hernandez.
It's a given that Kelley's numbers go down in the rotation. How much so? I'm thinking like 6.5 strikeouts, 2.7 walks, something like that. Which makes him well above average -- and easily worth $10m per year.
.
Q. Is Kelley a high-upside guy? Are you really standing to gain much from him? A guy like Brandon Morrow or Phillippe Aumont, at least you get an All-Star if you're right.
A. I grew up watching (and hating) Catfish Hunter. Catfish must have thrown 70%, 80% fastballs, and all but three fastballs in his career were thrown into the catcher's mitt. Maybe it was four, check me on that.
Ron Luciano talked about ump'ing Catfish: "He'd open the game with a fastball one inch off the plate. You called that a strike, and he threw the next one another half-inch outside. You called that, and by the third inning you were leaning into the dugout calling strikes."
Kelley's a throwback. In my day, the Catfish Hunters and Ferguson Jenkinses and Dennis Eckersleys (and later, Maddux and Schilling) threw dead-red to hitters sitting dead-red. They had the location to get away with it...
Ben whatsisname, the ex-M's catcher, always said "the very best pitch in baseball is a located fastball." Right after Gaylord Perry's spitball and Erikkk's hook, naturally.
I've never seen a RHP come up through the Mariners system with better command of a good fastball than Shawn Kelley. Never. Well, Scott Bankhead, but he threw 88.
I don't know if Kelley will bring that location back to 2010, but the Kelley of 2009, he offers a shot at a poor man's Catfish Hunter or Ferguson Jenkins. (Notice, Taronator, that Catfish and Fergie gave up HR's, too. It's part of the family heritage.)
.
Q. You really think Kelley is that good?
A. I think he's in that template -- guys with plus-plus command who don't throw marshmallows.
I don't think the difference between 94 and 92 matters much with him, no. It would to Brandon Morrow.
.
Comments
Of course Catfish Hunter also had gopheritis...as did Bert Blyleven. You can win with gopheritis as long as you're not putting many guys on base.
But yes...that is Kelley's biggest impediment to stardom...bullpen OR rotation. He'll never be a dominant closer or set-up man if he's getting jacked once every five appearances. And he'll never last in the rotation if he gives up 1.4 HR/9.
high gopher counts for a located fastball guy like Schilling or Kelley or Radke or Johnson (don't know their HR counts, don't know if it matters much).
The first thing I'd say is the pinpoint fastball guy is likely to surrender a few of those, simply because he's throwing straight so much of the time, and if the hitter does guess location, he's already got timing in his pocket. So it *seems* to me that those guys should give up a few more than the average mix pitcher.
Secondly, I would think that it's not a harbinger of anything else, simply because the located fastball guy (velocity and movement consistent year-to-year) wouldn't suffer from much volatility in this department. If he's a high HR guy, so what? If it coincides with a big LD% and too many flyballs, ok. But if it's just a function of a higher percentage flyballs end up in the seats, I could live with that.
Are going to give up home runs. Kelley's HR rate of 1.8 last year is way above the 1.4 demarcation line, but of course two homers more or less make the stats look much different for him.
IMHO you should be able to watch Kelley toss one game and realize that it isn't a Doug Fister-level concern here. Guys who chronically give up 1.5 homers do so for specific reasons, none of which exist with Kelley.
I think your statement about watching ONE GAME completely negates your argument. Hitters, pitchers and managers have understood for a century that the vast majority of HRs are off "mistake" pitches. Now, Vlad or Howard is going to do more with a mistake than Yuni or Jack Wilson, of course.
But, HR issues are not - and never have been - about competency. They're about "consistency". What is the frequency rate of a given pitcher's mistakes. Rivera probably throws 20 mistake pitches per season, (and 4 get crushed).
And, from my point of view, it is completely and utterly impossible to judge the frequency of a given pitcher's tendency to throw mistakes based on one game, (especially if that game is actually only 1 inning).
Kelley throws 753 pitches and 9 go over the wall. How many mistakes were thrown in those 753 pitches. 20? 40? 90? I don't know the ratio of mistake-to-HR -- but the HR stat isn't capturing JUST the HR events. It captures the frequency of mistakes - which I suspect can be traced back (for large groups) to worst-than-norm BABIPs and higher occurences of other XBHs.
In general, yeah, good control numbers are a nice proxy for capturing mistake pitch frequency -- but some high-control pitchers still have horrible HR rates. Some poor control pitchers have better. I don't believe in the least that there is a 'magic template' for determining which pitcher is going to allow more HRs, (especially based on a single inning).
I've often stated that the one area of pitching that gets the least (i.e. non) attention is "consistency". Being Washburn every game is better than being Nolan Ryan 50% of the time and HoRam 50% of the time. But, that guy who can be Ryan 50% of the time is the one that is viewed more positively. The pundits tend to overview from "either" the upside (Felix having 3-hitter stuff 34 times a year), or downside, (Olson is effectively HoRam 34 times a year).
The focus is (IMHO) too squarely on how many 'dominant' pitches are thrown, and not nearly enough on how many 'mistakes' are thrown. The perception is that weaker pitchers throw more mistakes. The reality is that (especially with young guys), weaker pitchers may throw fewer mistakes, (else they wouldn't get a chance at all).
I agree mostly with what you say here but I choose to look at things a bit differently, Sandy. Where I draw the line in the sand with the Saber community is when it is assumed that a % of effectiveness vs. luck can be put on a pitcher. I've read up a bit on DIPS theory and can respect it for what it is but there is a key element that gets lost in translation between the study and how its interpreted. Really, at its heart (correct me if I'm wrong) based upon season to season averages, it has been determined that pitchers don't control the results of balls in play. I even disagree with that to some extent but that's a different point. How that information seems to have been translated from what I read from analysts now is that they don't have any control over balls in play. IMO, the key word that is forgotten is RESULT, or whether or not that ball goes into a glove. However, given that there are many variables to that result, by far the pitcher controls the most inputs into the direction and velocity of a batted ball.
This applies to all varieties of batted balls. I agree that most hr's for example are deemed mistakes by the pitcher and that all pitchers make mistakes in a game. Some make more than others and some are penalized more often than others which IMO boils down to stuff as much as anything.
No disagreement really ... maybe some refinement.
My take on the current mass SABR-think on pitcher stats is that the subtlety that gets lost in the static is that HR are actually capturing a LOT more data than simply the HRs. As you note, the general belief is that pitchers don't control "balls in play". That excludes HRs, (BBs and Ks).
The BBs and Ks are pitcher-obvious. The HRs are slightly more complex, in being given completely to pitchers credit/blame-wise -- with a nod to the reality that parks impact HR totals.
My belief is that the DIPs numbers are HEAVILY influenced by HR rates, (a lot more than 5 or 10 events over an entire season could possibly explain). But, this is where I think the HR category is actually capturing a lot of that bleed-over "control the ball" aspect that pitchers really do have.
Some real low-K pitchers can have success, (Silva's good years - Wang). But, a high HR rate (especially a change for the worse) might be the mathematically obvious trait that explains a jump from 10H/9 to 12H/9. He doesn't have to give up two extra HRs per game, for the HRs to capture the fact that he's getting knocked around much harder than normal. The HR/9 stat, for pitchers, has essentially become the proxy for *ALL* the control a pitcher has over the balls in play. (Not to say this is completely accurate - just pegging the current landscape).
There are sabes who continue to "try" and factor in GB/FB rates and other pitching stats to refine the basic theory of DIPs -- but mostly they haven't advanced much -- and I think it is because they are missing the reality that the control over ball-in-flight that pitchers have is likely "already" factored into the DIPs calculation via the HR/9 stat.
I'm thinking the HR/9 stat is capturing that mistake-rate ... but complete admit that this is REALLY putting a lot of importance on a stat that can swing a great deal with very, very few events during a season. Since the scale is so small - the chance of random fluctation, (luck), is much, much higher.
It's no easy task to try and compute "mistake" pitches. In theory, every ball 4 is a mistake. But, a ball 4 to YuBet is a definite mistake, while a ball 4 to Pujols is obviously a 'mistake' of drastically different proportion. This is why walk-rate alone isn't a perfect stat to judge contorl. Batista routinely would nibble against Pujols and attack Yunis. The situational strategy complicates the picture. Every 1.5 BB/9 guy has great control. But, every 4.2 BB/9 guy does not necessarily have horrible control.
To calculate true mistakes from unintended results. My thought though is that despite all of the noise surrounding the ball in play stuff, pitchers do for the most part control direction and velocity. Now perhaps its true that they can't finely control either but as input to a mythical equation, there's is the largest.
Some of the biggest arguments I've gotten into about baseball stem from sabes disbelief that Roy Halliday is causing lazy grounders hit to his middle-infielders. In fact, he has that ability, its what he and every other pitcher are taught to do and their varying rates of success to me shows talent differences, not random events.
We seem to be quite willing to declare Ichiro's ability to hit .350 as a skill, unique from others. At least from what I can glean from many of the neo-sabes, Roy Halliday to generate ground ball outs at similar %'s is entirely un-acceptable to their community. To sum up, there is information that they don't (and will never) have. That's not luck, its a repeatable skill, perhaps a bit esoteric, but its there.
Matt's system actually identifies a small group of pitchers who have shown extended consistenty ability to produce better BIP results. But, the number is very small - and the impact is not very large.
That said - what is Halladay's defensiveBABIP? For his career: .294
Standard AL BABIPs are around .300 to .301. So, yes he skews just slightly to the good. Dunno what his aggregate defenses were behind him. But, compared to the average defense, Halladay is "barely" better than average at inducing in-play outs.
Let's compare him to say ... Washburn:
Career BABIP: .275.
Do you see that? .275 is Washburn's career BABIP. That's a full 20 points better than Halladay. Washburn is very likely this generation's absolute best pitcher in terms of 'beating' the BABIP gods. Halladay isn't even in the same league with him in regard to in-play results. Almost nobody is. But, Halladay fans 1.5 more hitters, walks 0.7 fewer hitters, and has a 0.8 HR/9 number, while Wash is at 1.2.
The numbers say Halladay is a great pitcher. Because he's got solid, (but unspectacular) Ks. Otherworldly walk numbers. And plus-plus HR numbers. Halladay doesn't NEED great BIP numbers to explain his success, and compared to everyone else ... he doesn't have them.
I'm really not referring to BABIP when I state that Halliday has an ability that eclipses most others. So his having similar #'s to the norm wouldn't really surprise me. My point is his (and others, depending on their expertise) ability to steer the ball towards certain fielders with a controlled velocity. The % of success need not be sky-high to consider that a skill.
The ultimate arbiter for every defensive-looking stat (pitcher or fielder) is pretty basic. Was an out recorded -- and if not, how many bases were advanced?
Imagine we could definitely measure that Halladay does indeed produce lower than normal mph for ground balls -- and that said groundballs tended to go more toward the left side of the mound. And that compared to other pitchers, he was the best at this. Yes. That could be considered a skill. But, having that skill doesn't mean anything (baseball-wise), unless it produces more outs.
Perhaps there is a break-point where he's soooo good at this that he ends up surrendering a higher percentage of infield hits. What's the result? He gets the same number of outs as the next pitcher on groundballs.
This is why we NEED the Sabes -- to take a theory (like Halladay is better at getting GB outs than everyone else), and attempting to quantify it. Well, BABIP doesn't answer that question specifically. But, given his BABIP -- I think it does pretty clearly state that if he does have an edge there -- he's giving it back elsewhere.
Ultimately, offense, defense, pitching can all be viewed as bases-per-out. We can't easily segregate pitching/defense, but we can count the same things the same way for each pitcher and see where players thrive or flail compared to those norms.
The what happened can be (largely) quantified. The why it happened will always be subject to debate.
Good points. One quibble again with BABIP. I'm not sure in the Halladay example it means too much, I don't see him giving back much to make up for what I believe is a skill. Extreme control pitchers by nature are going to have a higher hit total. In his case, he is at the top of the game in WHIP, k/bb ratio, HR's, and merely upper middle class in hits/9. One could look at his BABIP exclusively (not that you would) and come to some false conclusions. One could also decide that he operates outside of statistical norms in the other ratios described and call for heavy regression. My point is that's a false conclusion using the available stats. I don't think anyone expects him to crumble any time soon.
A couple of seasons ago, I recall a broadcast where someone had interviewed Halladay who stated that he was done with the SO and was content with pitching to contact. I believe that was done for his 06, and 07 seasons. Looking at his reference page, you can see he did precisely that, and while successful, he wasn't nearly what he had been. Then, the last 2 years, he brings the K rate back up to near the top of the league and he's unhittable. To me, that shows exactly who is in control of the situation.
Can't we take this the other way as well. Why is HR given up a good arbiter? Why do we assume a HR given up was "caused" by the pitcher, or was on a mistake pitch? There IS a guy in the batters box too you know.
One very quick example is the two 450 foot blasts that Tui hit the other day. Both where GOOD pitches. Both were "Pitcher's pitches". Neither one was a mistake, and both were in the exact strategic location where the pitcher wanted it, yet Tui hit one for a moon shot double and the other for a HR. And like you said in a stat that can be greatly impacted by one event I think it is mostly impossible (except on the extremes) to really tell ANYTHING about a pitcher by his HR given up.
In reality HRs given up are just as random and luck driven as BABIP (Look at Aardsma last year for instance).
In fact I would more quickly believe the idea that Halladay can influence when, where and how hard a hitter hits a ground ball, than believe that a pitcher has any great control over how many HR's are hit off of him except for perhaps just in sheer number of Fly balls given up.
In fact that is WHY Washburn has such a good BABIP because he DOES give up a ton of fly balls, and for the most part is able to suppress a blizzard of HRs. Is that a skill (High FB% plus Low HR/Fly), luck, ball park, good defense, fewer than usual mistake pitches? I don't think we have any inkling.
I'm not trying to come off as too sure of myself here. There are a ton of influences over batted ball direction and velocity and in no way would I imply that the pitcher stands in complete control of the event. Just as easily as he may be able to execute his game plan, the batter can also enforce his will on events to some degree. Upper-cut swings, cheating on inside fb's, looking for a location etc. all can play a major part in results. For the most part though, at game speed one won't see a ton of quality pitches hit 450ft. Can it happen? Sure, but not many pitchers would feel as though they executed their game plan well if those events happen with any regularity.
I still won't yield on the HR being a function of luck exclusively. By and large, a mistake has been made or an opportunistic guess has been made by the batter to be able to react and load up sufficiently to hit it out.
As to the ability to create left-side grounders, it really is planned for. Its why the (true) slider was developed. Guys who didn't have a good curve ball have historically been taught this much easier to master pitch to play off the fb, a well thrown slider has the batter slightly out in front with the contact towards the end of the bat causing the ground ball in the general direction of the SS. So in essence, having mixed and located pitches well enough to get ahead of a batter, the executed slider has both the ability to get the K, or a grounder aimed at the best glove on the field. Does it happen every time? Absolutely not. But as I stated earlier in the thread, we don't hold a 65% failure rate against a batter, in fact, we call him the best pure hitter in the league.