Deception, or Lucky

John Lannan recorded a season high five strikeouts last night, rolling through the Arizona lineup like Halladay. Obviously he has to be doing something different, right?

I dropped in a few links below to the pitch f/x data of a couple of his previous starts to contrast to this one.

Arizona, Aug 13

Colorado, May 13

Chicago, April 26

And for good measure, his start against the Yankees last season.

One thing is for certain, by that Colorado start, the velocity on his fastball was down 2 mph. Another thing that makes Lannan unique is that he really uses the whole strike zone. He almost never throws to the same location twice. Is it by design or lack of control, hard to tell. The release point last season was much higher, but earlier this season he had drawn it. Most pitchers drop the arm slot to compensate for an injury, and he has, but the horizontal point of release is also in, meaning he was bending and dropping his elbow more.

The key to his success last night was the swinging strikes on the changeup. If he can maintain these mechanics, he is disguising his release point a little better than earlier in the year. The key is to keep the motion "over the top" so that the ball is hidden behind his head as long as possible, making it tougher for righties to pick up. When the elbow is bent, the ball is coming from off his ear, and it is much easier to see (ironically, his opponent last night, Joe Saunders, has had similar issues with his changeup). Whether or not this is something he and Tomlin worked on is unknown, but he is throwing the ball slightly different than he had earlier this season. The increased velocity (89 mph avg) leads me to believe the adjustment is working.

It is still too early to tell if John Lannan is back. That was the Diamondbacks, and they tend to get themselves out a lot. He has about eight to ten starts to resurrect his season and get back into the rotation for next.

Now about that Marquis guy starting tonight...

Finding Hidden Value- the Cutter

The key to Rizzo's philosophy with pitching has been to keep the ball in the park. He has done this by bringing in ground ball pitchers and finding guys who can pick up the thing behind them and chuck it around to record some outs. This has worked for several clubs, so Rizzo has tried to get a leg up on the competition with his own model.

The problem with sinker ball pitchers is that they do not generate many strikeouts, as noted that they are dead last. (Pittsburgh, the second worst team in the NL, has 11 more than the Nats) The theory is simple: a swinging strike rarely generates a base runner, let alone a run. The Nats had Stephen Strasburg fall into their laps, and he is likely to generate a few of his own. But what about the rest of these guys? How can the Nats maximize the effectiveness of their staff, both current and future players?


Sky Kalkman provides some interesting data on the cut fastball. While the data he provides has many shortcomings, it illustrates several things. First, given the infrequency in which batters see the cutter, it immediately becomes a weapon, almost as much so as the knuckleball. Second, while the hitting numbers for the cutter appear weaker in an 0-2 count, where a pitcher would be using it strike a batter out, the numbers it produces in a 2-0 count are staggering. 2-0 cutters are thrown in the zone 66% of the time, swung at and missed 15%, and generate a BABIP of .273.

2-0 Counts % Zone Swing Whiff BABIP HR CON
Curveball 2% 57% 19% 35% .217 4.0%
Cutter 3% 66% 44% 15% .273 1.8%
Fastball 76% 62% 40% 12% .333 3.4%
Off-speed 10% 56% 39% 28% .248 4.9%
Slider 9% 61% 36% 29% .343 2.4%

The cutter, especially in hitters' counts, does exactly what Rizzo is aiming for: keeping the ball in the park. Overall, the cutter trails only the curveball in home run rate. The biggest advantage is that the cutter can be used inside the strike zone much more effectively. The result is fewer hitters' counts, and an available "out pitch" in a hitters' count.

All Counts % Zone Swing Whiff BABIP HR CON
Curveball 9% 46% 39% 28% .296 1.6%
Cutter 3% 56% 49% 18% .289 1.7%
Fastball 59% 55% 44% 14% .308 2.1%
Off-speed 12% 47% 48% 27% .288 2.2%
Slider 17% 50% 47% 29% .285 2.0%

Obviously Steve McCatty just can't have ALL the guys pick up the cutter, but when looking at some pitchers with great location (Craig Stammen) that just don't have enough in their arsenal to be anything more than a number 5 guy, why not give it a shot.

Sky mentions numerous success stories, but he omits what I find to be the most interesting case. Mark DiFelice of the Brewers has owned righties, absolutely owned them. In 51 innings, he has walked 15 and struck out 48, throwing his blazing 83 mph cutter exclusively. There are theories as to why his cutter works, but the applications should be the same. Work in the zone early, then expand the zone to get outs. Too many pitchers on the Nats do not throw first pitch strikes for fear of giving up too many hard-hit balls.

The cutter may be a fad, it may be too difficult for most pitchers to master, but the slider and the splitter both changed the game. For a team looking to make up some ground in the standings, the Nats need all the help they can get. Rizzo believes sabermetrics is easy... yes, reading numbers off a sheet and choosing the best ones is easy. But will he be the one to go off on his own and try something new, generate his own model for success?

** I will work out the chart formatting soon... stupid blogger formatting.

Craig Stammen... Is He Part of the Future

Wow... took a week to get this posted. Work has been getting in the way of all this blogging and crap.

Craig Stammen finally got back on track last week against Atlanta. He only went 6 innings, but didn't get knocked around like he had in his previous starts. Some may think that he has regained the form he showed in his early July hot spell. However, start in Atlanta was marred with inconsistent location, and the Rockies game exhibited a steady decline in velocity from pitch 1 to 81. It should also be noted the consistency of the strike zone last night... in a season of lousy calls, the His last two starts, he has avoided the fly ball. Basically, Stammen doesn't have the velocity to beat anyone above the belt. However, if he continues to throw nothing but fastballs low and away, teams will adjust and hit the ball the other way. At some point, he needs that fourth pitch to keep hitters off balance the third time through the lineup.



In any event, with his innings getting up there and the bullpen relatively fresh, do not expect another performance as to what he showed in July.

His last good start in July shows the ball staying down and away with a consistent release point. In the Atlanta game, he appears to have gotten away with a few pitches up in the zone that had been spanked recently. The thing that bugged me most about Stammen's recent slump though was this:



I'm not exactly sure what shape that is, but clearly the flattening of his release point had a huge negative impact on his ability to keep the ball down in the zone. It continued from the first Milwaukee game into the Pittsburgh game... honestly, the Florida game was too short to get a good feel, but it looked like his release was significantly more erratic.



The conclusion is that Craig Stammen and Coach McCatty need to fine tune his mechanics so that he doesn't have that "inverted T" on his release chart. This reflects the fact that he locates his pitches better with a more "over the top" delivery. In the offseason, he will have a chance to polish his 2-seam fastball to make it more of a weapon driving the ball down in the zone.

Update: The second Milwaukee game yields better results with a tighter release point. Clearly the further he strays side-to-side, the more danger he finds.

Craig Stammen: Sometimes Lucky is Better Than Good

For the first time in his 9 game career, Craig Stammen survived the 7th inning. He was the tough luck loser in a 1-0 game, putting just seven runners on base in seven innings (5 hits, 1 walk, 1 error), doubling three of them up.

The causal observer would believe Stammen pitched his best game. Someone who watched the game would argue that Stammen finally escaped a couple jams with two line-out double plays, plus prevented the big inning in the first by pitching around masher Brad Hawpe and getting the ground ball out of Tulowitski.

Who's correct? That is why pitch f/x is around. First, Craig Stammen has marginal major league stuff. His fastball struggles to find 90, and his breaking pitches are not unhittable. Stammen can survive as a decent back-end pitcher by working the outer quarter of the plate, getting one or two favorable calls per game, and learning to keep the ball down. Greg Maddux made a career of this crap, and a few other righties have followed suit, but with less success. Below are a couple of example of Stammen working batters away this season:





Almost all of his starts, he has shown similar location, regardless of how the other team hit him. In Colorado last night, however...



Jesus.

He was all over the place. The Rockies were not patient enough to wait for a mistake, especially when Stammen becomes more vulnerable the third time through the lineup.

Now for what I like to call "The Coors Effect", and no, that doesn't involve anybody shotgunning "A Taste of the Rockies". Even as far back as his first start against Pittsburgh, Stammen featured three distinct pitches:



Last night however, the thin air had a noticeable effect on the movement of all three of his pitches. His curveball and changeup were reduced to the same pitch and the movement on his fastballs were drastically reduced. Combine that with the struggling location, and you'd think that we are looking at Daniel Cabrera.



The Rockies failed to execute with the bat when Stammen made a mistake. It is that simple. While I hope Stammen rebounds and drops another 7 inning, 1 run performance on the Astros, I am not holding my breath. Expect continued struggles later in the game, once batters get a read on his fastball and his location starts to fail.

More on Olsen

Chico was all over this.

Looking at the pitch f/x data, the fastball velocity is back up two mph. That is good, but great will be if he can add a little bit more as his arm gets extended this summer. The most notable thing that sets this start apart from those he made earlier this season is that he was able to locate three distinct pitches, and even threw a couple 2-seamed fastballs for good measure. The keys, though, are graphs 5 and 6, which show three distinct pitches, staggered by several mph each. Earlier in the season, these plots were erratic, with the three pitches blending together.



The problem is that Olsen has teased with his talents before. Don't buy in until all the evidence is on the table.

Cabrera's Last Stand

I know you all can’t wait for the next Daniel Cabrera start. I know I can’t. Seriously, the sky is the limit for Cabrera. He could actually join Rick Vaughn in the Hall of Shame. Seven wild pitches in 24.1 innings, adding 17 walks, 2 hit batsmen, 255 strikes on 459 pitches (55.6%). Those numbers don’t really even begin to tell the story of how poorly he has actually pitched. He has managed to do this over five starts as well, so management is coddling the giant and are not being rewarded. It would be unfair to say Cabrera is the worst starter in the NL because the Mets look like they are going to have paid 2.2 million for Oliver Perez’s body of work. He has probably been the second worst starting pitcher in the NL, though.

The sad, sad thing is that management had to know what they were getting into; it is not like he suddenly lost command. He never had it. Period. Randy Johnson didn’t begin to harness his unique anatomy and stuff until he was 29, in fact appeared to be regressing upon arriving in Seattle before breaking out in ’93. Now he’s first ballot to Cooperstown. The difference is that Cabrera broke down at the end of last season after a decent start and really needed to be fully examined before coming on board.

As noted by numerous accounts, Cabrera’s velocity is way down. Everybody wants to cover their asses and blame it on tinkering with his delivery. Yet a closer look his struggles in the second half of last year reveal similar trends. He produced fewer swings and misses as the season progressed, generally resulting in fewer strikeouts and more balls in play. This season, he has produced a total of 22 misses. That number is fine for Derek Lowe or Jamie Moyer, pitchers who pitch to contact, but a power pitcher gives up more line drives and fly balls. Power pitchers, who don’t have pinpoint control, generally rely upon beating average hitters.

A look at Cabrera’s pitch history shows that even when he has been successful, he has had fairly miserable control. He actually has survived this long because of the perception of his lack of control. First pitch fastballs resulted in 79 balls and 61 called strikes out of 193 pitches. Basically 75% of the time the batter is taking all the way. Most of the taken strikes were hittable pitches; there is no grouping on the fringes. His overall first pitch results are about 50% strikes. Yes, very poor. His slider has gotten him out of jams before, but without the velocity on his fastball, hitters now lay off it and watch it miss.

There is little hope for Cabrera to figure this out in one start or one month. Randy St. Clair doesn’t seem to have the first clue of where to start. Sending him to the minors won’t accomplish anything, except maybe further some bad habits by getting a few extra cheap punch outs. The key is to shorten his windup and delivery as much as possible, eliminate the wasted energy. He will never make the leap Johnson made, but getting him to the 7th inning would be encouraging. It's not happening and he will likely be out the door by June 1st.