16 replies to this topic

[mabrowndog]

As usual, Jnai (Dr. Dan Brooks) puts together another fascinating piece of analysis, this time on arguably the club's biggest disappointment and lightning rod. The graphics alone are compelling enough, but the entire article merits a read from start to finish.

Check out Beckett's pitch usage patterns over the past 5 years:



I'd been aware of a drop-off in his use of the 4-seamer, but I never noticed it was this severe. Nor did I notice the gradual increase in his throwing the cutter or the sudden jump this year in deployment of the sinker.

The rubber really hits the road when Dan looks at his usage by inning:



Brooks writes:

On the one hand, it’s an example of a classic pitching approach: establish the fastball early, and then go to your other pitches. On the other hand, it’s confusing: What’s the use of establishing the fastball, which is still a decent pitch for Beckett, if it’s simply going to be shelved for the rest of the game? It’s possible that this is a mental thing. Without a 70 fastball/curve, Beckett could be worried about using the four-seam and attempting to pitch to contact with his two-seam and cutter instead.

In the first inning, when Beckett has been absolutely hammered this year (10.42 ERA), you see an almost obstinate reliance on the fastball, with very few curves. By the third inning (and throughout the rest of the game), the cutter has supplanted the four-seam fastball. Such an identity crisis from the first inning to the later innings is truly odd.

Later in the comments section, he notes:

I won't lie: the takeaway is a bit schizophrenic. He's been hammered in the first inning when he throws the most fastballs, and yet, his fastball is - from a results point of view - his best pitch. Paradoxical! Is he being hammered on his OTHER stuff in the first inning, or perhaps on his two-seam fastball, which he also throws a ton?

The answer here is yes by the way, the other stuff is getting hammered in the first inning. His first inning curveball, cutter, and sinker are a disaster.

Then there's his eroding velocity on all his fastballs, which have lost 2.5 to 3 mph across the board:



Brooks then takes a look at pitch location. Against RHH, Beckett's missing horrendously on his sinker, leaving it up and over the plate. Against LHH, it's the cutter that's left in batters' wheelhouses.

Next Brooks assesses swing rates generated by each pitch type and discovers some oddities:

In 2012, Beckett’s four-seam fastball whiff percentage ranks 178th of 246 pitchers who have thrown at least 100 four-seam fastballs, but his cutter comes in at 12 out of 59 pitchers who have thrown at least 100 cut fastballs.

When it comes to swings and misses, Beckett’s curve currently ranks 21st out of 55 pitchers who have thrown at least 100 curves—right in the middle of the pack. It’s too bad we don’t have any data on the very early Beckett 70-plus curve for comparative purposes.

Using linear weights Brooks gets down to brass tacks on each pitch, first as rate-based run valuations, then on a conventional 20-80 scouting scale:

Suppose you were to take every starting pitcher in baseball and compute this same number for their four-seamers. Josh Beckett would rank sixth out of 188 starters who have thrown at least 100 four-seam fastballs. And if you were to normalize all their scores into z-scores, multiply by -10, and add 50 (which creates a sort of ad-hoc 20-80 scale), Josh Beckett’s four-seam fastball is a 73.6.

The curveball ranks 42nd out of 128 qualifiers, good for a 56.8 rating. The cutter ranks 50th out of 71, good for a 44.7. The two-seamer ranks 83rd out of 159, good for a 49.2. And the change is 67th out of 125, good for a 50.1.

IMO, the 4-seamer being so highly valued reflects how batters are sitting on his other offerings, knowing damn well he's going to miss his spots. The loss of his curve, once the most devastating pitch in his arsenal in terms of keeping hitters off-balance, is probably the biggest factor in his disrupted approach.

Simply put, even without his ugly stat lines, Beckett's approach and execution have been all over the place and a complete mess, and unless they find a taker the Sox are committed for another 2 years at ~$16M per annum.

Yes, pitchers should be expected to decline as they approach and pass their 30th birthday. My thoughts on Beckett have long been that since he's been a power pitcher in the Clemens mold (using his legs to generate forward drive with a smooth and compact delivery), his loss of velocity would be delayed and not as precipitous as the average MLB starter. Instead, the decline began at around age 28-29. I've also viewed him as a pretty smart guy, one capable of making adjustments to offset certain elements of physical decline. So I'm not quite ready to chalk up the entirety of his epic shitfest to simple attrition and the effects of aging.

I've also been among those carrying the torches and pitchforks decrying his attitude and conditioning, and I still believe the latter has been a factor in his poor performance. But doesn't there really have to be something else at work here? After having a successful 2011 in spite of his continual losses in velocity, why the sudden shift in pitch mix in 2012? And why the consistently abysmal location?

I'd be curious to see if there's been any subtle change in mechanics, such as arm slot or release point, though I suspect Dan would have already ferreted this out if there was anything relevant.

What has been the effect of all the instability in pitching coaches and catchers, if any? I'd like to think that his makeup, experience and intelligence would have made him immune to such things.

Kudos to Jnai for his tremendous work.

Edited by mabrowndog, 21 August 2012 - 10:48 AM.

[Harry Hooper]

I don't think he's establishing the fastball early. It's that he can't execute throwing anything else most games until his stiff shoulder/arm loosens up. Once loose, he de-emphasizes the 4-seamer that simply ain't what it used to be (in terms of velocity) in favor of other options on the menu.

[Sprowl]

I am baffled by the last section:

Suppose you were to take every starting pitcher in baseball and compute this same number for their four-seamers. Josh Beckett would rank sixth out of 188 starters who have thrown at least 100 four-seam fastballs. And if you were to normalize all their scores into z-scores, multiply by -10, and add 50 (which creates a sort of ad-hoc 20-80 scale), Josh Beckett’s four-seam fastball is a 73.6.

Why is this the right method to use to evaluate a four-seamer, especially when it directly contradicts the pitch value analysis that precedes it? How can a pitch have a negative linear-weighted value per 100 pitches, yet be considered a good pitch?

I think the additional piece of the puzzle is the change in Beckett's arm angle from 11 o'clock to 10 o'clock -- that has occurred progressively over his time with the Red Sox, and is responsible for the loss of bite on his curveball, as well as the increased dependence on the cutter, which is the classic resort of a fastball pitcher who loses velocity and has too much run and too little jump on the fastball. Power pitchers generally come overhand (eg, Clemens and the early Beckett).



The late Beckett looks more like this:

[Jnai]

Why is this the right method to use to evaluate a four-seamer, especially when it directly contradicts the pitch value analysis that precedes it? How can a pitch have a negative linear-weighted value per 100 pitches, yet be considered a good pitch?

Sorry if it wasn't clear. Negative is better in this context.

[drtooth]

Given his lack of 1st inning success, it looks like he tries to use the 4-seamer early. Once he gets hit hard with that pitch, he pretty much abandons his 4-seamer altogether. He troubles (once you get past the chicken/beer bluster) seem to be that he has not adjusted or is still adjusting to a needed change in pitching style away from being a power pitcher, especially given a fairly drastic drop in velocity this year.

[Buzzkill Pauley]

That's an awesome article, jnai.

It has certainly appeared to my eyes that Beckett has been adapting into a "let-them-put-it-in-play" starter as his 4-seamer fastball velocity has declined so that he can continue to go deeper into games. Finding that he factually gets more swings-and-misses with his cutter, change, and curve than either his 4-seam or 2-seam fastball, it starts to make sense.

That is, Beckett seems to me to be using the 4-seamer to test the boundaries of the day's strikezone in the first inning. But he knows he can't expect to go longer than 5 innings if he continues to rely on it, because at its reduced velocity MLB hitters can just keep fouling it off until they get one in a location they can drive, or he goes to a different pitch entirely.

What's really surprising is how rarely he goes to his other two-best pitches -- the change and curve -- in the 1st inning, rather than letting his mediocre 2-seamer get abused by the top of the opposing team's order. One would think this sort of trend would be readily obvious in pitchers' meetings.

[mabrowndog]

Given his lack of 1st inning success, it looks like he tries to use the 4-seamer early. Once he gets hit hard with that pitch, he pretty much abandons his 4-seamer altogether.

As Dan notes, he really isn't getting hit that hard on the 4-seamer in the first inning, despite relying on it so heavily. It's his other stuff that's getting plastered:

I won't lie: the takeaway is a bit schizophrenic. He's been hammered in the first inning when he throws the most fastballs, and yet, his fastball is - from a results point of view - his best pitch. Paradoxical! Is he being hammered on his OTHER stuff in the first inning, or perhaps on his two-seam fastball, which he also throws a ton?

The answer here is yes by the way, the other stuff is getting hammered in the first inning. His first inning curveball, cutter, and sinker are a disaster.

[wade boggs chicken dinner]

From Fangraphs, here is his pitch value for the CRV:
2002 = -1.9
2003 = 4.9
2004 = -3.1
2005 = 11.4
2006 = 6.0
2007 = 11.4
2008 = -0.5
2009 = 1.5
2010 = -3.1
2011 = 2.5
2012 = 0.9

To me, the issue is that Beckett always used to be a two pitch pitcher. Two really good pitches, but two pitches nonethe less.

Now he's a one pitch pitcher, and that's a hard way to succeed in this game.

[Sprowl]

Sorry if it wasn't clear. Negative is better in this context.

But why use that method? It is the rationale for why you crunched those numbers in that fashion that I don't understand. The result is non- or counter-intuitive (at least to me), and contradicts numerous other results that rate the fastball, and especially the 4-seamer, as a bad pitch. Fangraphs lists two different pitch valuation results for Beckett's fastballs: each one of them rates the fastball as negative -- which means not better but worse in each of those valuation schemes: PitchFX (as opposed to the manually reclassified data) puts the 2-seamer at +1.67, but the 4-seamer at -0.61; BIS lumps all the fastballs together and comes up with -0.17 runs per 100 pitches. Is manual reclassification alone sufficient to flip the sign and the evaluation? The way it is stated in the article, it appears that Beckett's fastball is subpar, but almost every other pitcher's fastball is even more subpar -- in other words, all major league fastballs suck. Is that really what's going on here? Huh is right.

[Alternate34]

Taking the article at face value, that Beckett relies heavily on his four seamer in the first while his other offerings are getting crushed, my initial suspicion is that he is having problems adjusting his pre-game warm-up as he ages. The 4 seamer has always been his pitch. This is not to say that his other pitches weren't good especially his curveball, but those other pitches generally require more feel than the fastball and I suspect that is more true for Beckett considering his fastball for most of his career was among the best in the majors. Of course, this is speculation because I have no idea what his regiment is and if it has changed or if soreness is forcing it to change.

[Kull]

I am baffled by the last section:


Why is this the right method to use to evaluate a four-seamer, especially when it directly contradicts the pitch value analysis that precedes it? How can a pitch have a negative linear-weighted value per 100 pitches, yet be considered a good pitch?

I think the additional piece of the puzzle is the change in Beckett's arm angle from 11 o'clock to 10 o'clock -- that has occurred progressively over his time with the Red Sox, and is responsible for the loss of bite on his curveball, as well as the increased dependence on the cutter, which is the classic resort of a fastball pitcher who loses velocity and has too much run and too little jump on the fastball. Power pitchers generally come overhand (eg, Clemens and the early Beckett).



The late Beckett looks more like this:

On the face of it, this appears to be just one more example of bad starting pitching mechanics that were allowed to fester for most of a season. First we saw it with Buchholz (for much of the first half), Lester for most of the season, and Beckett to this day.

[Buzzkill Pauley]

On the face of it, this appears to be just one more example of bad starting pitching mechanics that were allowed to fester for most of a season. First we saw it with Buchholz (for much of the first half), Lester for most of the season, and Beckett to this day.

Except that it's highly likely to be a "bad starting pitching mechanics" that developed because Beckett's shoulder and back can no longer physically withstand the additional strain of pitching over-the-top.

And I could be wrong, but I believe Sprowl put it there to show Beckett's gradually lowered-slot over time, not to suggest this is something that he dropped his arm angle strictly within 2012. If I'm wrong, I'm sure there's release-point evidence from pitchFX to show it.

Regardless, those specific pictures are about 7 years apart, if not more, and between them Beckett's rung up about 1500 innings for Boston (including postseason and spring training starts). That's a lot of miles.

[Smiling Joe Hesketh]

Except that it's highly likely to be a "bad starting pitching mechanics" that developed because Beckett's shoulder and back can no longer physically withstand the additional strain of pitching over-the-top.

And I could be wrong, but I believe Sprowl put it there to show Beckett's gradually lowered-slot over time, not to suggest this is something that he dropped his arm angle strictly within 2012. If I'm wrong, I'm sure there's release-point evidence from pitchFX to show it.

Regardless, those specific pictures are about 7 years apart, if not more, and between them Beckett's rung up about 1500 innings for Boston (including postseason and spring training starts). That's a lot of miles.

Yeah, nearly every pitcher lowers his arm slot over time as the shoulder wears. Pedro lowered his over his time in Boston to a remarkable degree.

[Sprowl]

And I could be wrong, but I believe Sprowl put it there to show Beckett's gradually lowered-slot over time, not to suggest this is something that he dropped his arm angle strictly within 2012. If I'm wrong, I'm sure there's release-point evidence from pitchFX to show it.

Yes, Beckett's arm slot has been dropping over a long period of time -- it was quite noticeable by 2010, but I think it began much earlier. Fangraphs doesn't have the time series charts of movement on the curve and 2-seamer that it used to offer, but iirc, the loss of effectiveness on the curve was concomitant with more slurvy motion (and less vertical drop). It was the subject of several threads 2-3 years ago.

[Kull]

Yeah, nearly every pitcher lowers his arm slot over time as the shoulder wears. Pedro lowered his over his time in Boston to a remarkable degree.

Good points Buzzkill & Joe. Serious question - are certain pitches more ineffective from a lower arm slot? For example, if the 4-seamer needs the high release point to be effective, then it wouldn't be possible for Beckett to rely upon it.

[Buzzkill Pauley]

I don't want to sound like every pitcher is the same, but generally a 4-seam "rising" fastball or a 12-6 curve is harder to square up when it's thrown over the top, as that pitching motion accentuates vertical movement. OTOH, a 3/4 or sidearm delivery is generally better for a slider or cutter that relies on a more horizontal break.

Bronson Arroyo's slider from 2004-05 was great for noticing this -- he was someone who deliberately changed arm angles to get more horizontal or vertical movement, as opposed to someone who tried to get every pitch from the same angle to accentuate deception on the pitch. His frisbee slider, especially, was thrown pretty-much sidearm, so that the spin on the ball and the arm angle swept the ball away from RHH (or into the breadbasket of LHH).

[Alternate34]

Good points Buzzkill & Joe. Serious question - are certain pitches more ineffective from a lower arm slot? For example, if the 4-seamer needs the high release point to be effective, then it wouldn't be possible for Beckett to rely upon it.

Sinkers can be more effective from a lower arm slot. Derek Lowe for example:

http://a57.foxnews.c...1_dereklowe.jpg

I was going to put a picture of Mike Hampton, but I got different images for his arm slot and wasn't going to go crazy. Kevin Brown also threw a sinker from a lower arm slot I believe.

Cutters are a more difficult diagnosis. Mariano Rivera throws his from a higher arm slot:

http://upload.wikime...g_ball_2007.jpg

However, Buzzkill is right that a lot of pitchers have success with the cutter from a lower slot. I would say especially lefties, but I don't want to be held to that. I'm trying to find a good picture of Al Leiter at release point because I think he had a lower arm slot.

On the slider, again, Buzzkill is generally right. Sliders usually rely a lot on horizontal movement and benefit from a lower arm slot. However, Francisco Rodriguez's arm slot was similar to the earlier Beckett. It looks a little lower than early Beckett, but much higher than later Beckett:

http://blog.silive.c...large_krod.jpeg

To reiterate though, good sliders like Bronson Arroyo's and great ones like Randy Johnson are often thrown from a lower arm slot and get a lot of horizontal movement.

There are a few issues about using arm slot as a general rule. One is the problem of cutters, slutters, sliders, slurves, and curves. We know a 12/6 curve when we see one. Beyond that, categorization becomes difficult. You can't use release mechanics and have it match up with the way the terms are used in history. Sliders are sometimes thrown like a sidearm curve ball with a lot of wrist breaking and sometimes thrown like a cutter with little wrist breaking. It's never consistently identified. Going by breaking direction doesn't work consistently either. K-Rod's slider had a lot of perceived downward movement, while other sliders have a lot of horizontal movement, and there is a wealth of ground in between. What does this have to do with arm slot? Some sliders are better thrown from 3/4 to sidearm, some are better thrown from a higher arm slot. Some are variable and can be thrown from changing arm slots, though that seems to be a lost art these days.

As for what this has to do with Beckett, it could be an explanation for his increased sinker and cutter usage. With his lower arm slot, he sees more action on those pitches relative to his four seamer. Here is some speculation, but maybe something to check. When considering lower arm slots, we usually think of the effect on the spin of the ball. However, the release point gets moved to the right from the pitcher's point of view. Additionally, the release being lower requires the pitcher to put just a hair more upward force (or if you like, less downward force) on the pitch to get it to the strike zone. The only reason this occurs to me is that the location charts from the article indicate that Beckett misses with his sinker vertically. Those red spots are both middle of the strike zone in height. The cutter is being missed horizontally and vertically. The location of the cutter is left of where you want a cutter to be from the catcher's point of view. He wants those cutters bearing in on lefties he gets their bat handle, not over the plate. There is a red spot outside the strike zone where he does want that cutter, but the three glaring zones are the middle, middle top, and low inside to lefties.

Of course, the problem with the theory that these misses are the result of a lower arm angle is that he is locating his 2 seamer pretty well against lefties. He might want to catch the plate more, but low and away tailing off the plate would seem to be what you want. Why does he have decent location of that 2 seamer against lefties and poor location against righties? One explanation could be that he has more problems commanding the right side of the plate from the catcher's point of view. He can't backdoor that side effectively with his 2 seamer and he can't get his cutter inside enough. You could argue that arm slot has something to do with it. He throwing farther away from that side with a lower slot. However, that could be one of many explanations.

I mentioned a few issues with using arm slot as a general rule for pitch effectiveness. Beckett demonstrates another issue with this. His lower arm slot has not led to a more effective sinker or cutter, which would contradict the general rule. There are a whole host of reason why this may be true. He has lower velocity so any marginal increase in effectiveness of those pitches would go unnoticed. Beckett's physiology may favor a higher arm slot for some reason i.e. his finger length may favor a higher arm slot or he has built his musculature to make him more effective from a different arm slot. He may still be getting used to a different arm slot. Of course, I believe his arm slot has been getting lower as he ages, so that last one may be fishy.