Jeremy Bonderman has had a very long road back from the surgery he underwent to correct thoracic compression outlet syndrome. There was hope he’d be back in the rotation this year, but his spring became very extended and he didn’t make his first appearance until June 8th – and he was crushed. It was back to the minors and more rest and rehab, but with the rosters expanded Bonderman is back as a reliever, and the pitch f/x data is promising…so far.

When Bonderman made his June start, his pitches didn’t behave in the way they had in the past. His velocity was down. His fastball didn’t have the same arm side run. His slider wasn’t as tight and was more of a sweeping pitch. The particulars (via Brooks Baseball) were:

Let’s fast forward to his latest 2 outings. The results have certainly been better, but what about the stuff?

We see that his velocity is up about 2mph across the board. The fastball has a couple more inches of arm side run and the slider has an inch and a half more drop. This all seems to be good and much more in line with Bonderman’s past pitch f/x numbers.

I’m going to skip the 2008 numbers where he was clearly hurting and couldn’t maintain his velocity. This clouded pitch classification data greatly. Instead I’ll put the numbers from 2007 up. These were harvested from Fangraphs.

The pitches from September (granted there are only 36 of them) look much more like the pitches from 2007 than the pitches from the June start. And yes, small sample sirens should be going off. And this is far from an ideal analysis. I’m certain there are pitch classification issues that are especially prevalent between the two seam (FT) and four seam (FF) fastballs. Plus the system was relatively new in 2007 and still being calibrated. Still, the results from this cursory analysis are encouraging.

On Tuesday night Joel Zumaya threw a ball that the Comerica Park gun, and the Fox Sports gun clocked at 104mph. That is freakishly fast and a little hard to believe, at the very least there was probably some rounding up. A check of MLB.com’s pitch f/x data had the pitch 50 feet from home plate at 102.2mph. So the stadium gun was likely a little hot, but that is still obscene. It did make me curious about the fastest pitches thrown this season.

By my check the pitch f/x system has recorded 91 pitches that topped 100mph this season (through 6/25/09). Of those 91 pitches, they basically all belong to Zumaya.

If you order the pitches in descending order, Zumaya actually has the 38 fastest pitches this season (Justin Verlander had the 39th) and 57 of the 58 fastest. Zumaya has 28 pitches at 101mph or faster.

As for the fastest pitch this season? It wasn’t the pitch that finished Bradley which came in a 102.2. But it did come earlier in the at-bat when Zumaya hit 102.7. For good measure he came back the next day and hit 102.6 against Mike Fontenot. Those are the 3 pitches this season top the 102mph barrier.

Pretty much everyone has beat me to this, but I still feel I should offer up something about Dontrelle Willis’s first start of the year. There are a couple ways to approach the assessment.

You can look at a final line that included 10 hits and 2 walks in just under 5 innings and come away unimpressed or even discouraged. Especially when you look at the fact he only notched one swing and a miss.

Or you can look at the start in the context of last year when the strike zone proved to be a foreign land to Willis. That he only walked 2 in his time on the mound could be a step forward. And even when he missed the zone, he missed by reasonable amounts, like you’d see in a regular start.

Figure 1 Pitch Location

I’m encouraged in that I don’t see him pitching scared. He came inside to both lefties and righties. I don’t know if that is by design or a lack of control in the strike zone, but the absence of unintentional pitch-outs is encouraging.

With so much of Dontrelle’s problems last year related to the inability to throw strikes (be it mental or physical) he had no chance of success. Efforts like Wednesday night where he’s in and around the strike zone, and throwing the ball in stead of aiming it. His fastball averaged a shade over 90 and he threw a number in the 91-93 range.

There was considerable mechanical work done on Willis last year with simplifying his delivery and trying to find something more consistent. It didn’t enhance things at all. The high leg kick is back and with it some inconsistency in release point. I don’t know how much of an issue that plays, but I think this graph of release point is interesting so here it is.

If we can say that the control problems are under control, at least the focus then turns to effectiveness. That is where Willis fell down in his debut start. He didn’t miss bats, even when he was getting ahead of hitters (59% first pitch strikes). Willis wasn’t particularly effective in 2007 and that was when he was right as he pitched a 1.6 WHIP that year. That his first start of 2009 looked to be an extension of that season wasn’t encouraging.

I don’t know if I’d call Dontrelle’s outing good, but I would call it progress. He can now throw the ball over the plate. He knows it. The coaching staff knows it. The fans know it. Now it is a matter of whether or not he can be an effective big league pitcher.

Some other takes:

• Bless You Boys: The Unbearable Lightness of Low Expectations
• Mack Avenue Tigers: Willis earns second look

Justin Verlander has turned in 3 remarkable outings in a row amassing 31 strike outs as hitters can’t catch up with his heater or their knees buckle with the curve. But very quietly Verlander has added a slider to his repertoire.

This pitch received significant attention from Rod Allen and Mario Impemba last night when he picked up a swinging strike with it against Kelly Shoppach. But he actually began throwing it as early as the April 27th Yankees game. The pitch was first noticed by Eric Cioe (who comments here on occasion) and he posted about it at Motown Sports.

Eric and others weren’t sure what to make of the pitch at first. It only showed up a handful of times against the Yankees and we weren’t certain rather it was in fact a new pitch, some pitch f/x funny business, or simply Verlander mis-throwing a pitch. After the pitch turned up again during last Sunday’s start against the Tribe it was pretty clear it wasn’t a fluke. But the pitch seemed to be used primarily as a “show-me” pitch. However last night it took on a more prominent role.

Let’s turn to some pitch f/x graphs to show you the characteristics. Each of these graphs are a compilation of Verlander’s last 3 starts (April 26th, May 3rd, May 8th).

Figure 1. Horizontal and Vertical Movement

There at least 3 distinct groupings here, but that doesn’t tell us much because we knew that Verlander’s repertoire included a four seam fastball, a change-up and a curve ball. We know that the lower right cluster is the curve ball, and that the fast ball is included in the upper cluster.

Figure 2 Pitch Speed and Horizontal Movement

Now we can visualize 4 different clusters. The fastball is of course the upper most grouping and the curve is again the lower right bunch. The change-up is centered around 84-85mph and has a little more right hand bearing action than the fastball. With this in mind you can look at Figure 1 and see what is actually a 4th grouping to the left of the fastball grouping.

But, look at that little cluster in the middle. The pitches are thrown 88-91mph and with a horizontal movement component that is completely unique. These aren’t underthrown fastballs or overthrown change-ups.

Figure 3 Speed and Vertical Movement

Taking a look at velocity and vertical movement the fastball group is still the fastball group and the curve is easy to identify as well as the only group with downward movement. The change-up can be identified by the velocity and vertical movement that is less than that of the fastball. But at that high 80’s range we see pitches that just don’t fit anywhere else. But they also are a fairly unclustered cluster so perhaps he is still trying to find consistency with the pitch.

Verlander has only thrown this pitch about a dozen times over the last 3 games total so this isn’t a frequent pitch or something that Verlander is likely to rely on. However, with Verlander’s fastball electric and sitting in the high 90’s, the last thing opposing hitters need is one more thing to think about. As long as Verlander keeps the pitch down and doesn’t hang his 4th best pitch, it can only help.

Justin Verlander has gotten off to a rough start in 2009. People are starting to question his “ace-hood” and with an ERA of 9.00 after 4 starts it is probably justified. What makes his start so perplexing is that his “stuff” appears to be back. His fastball velocity is over 95mph and he’s fanning 10.7 batters per 9 innings. The numbers that really need a deeper examination though are those when runners are on base. Verlander is stranding only 39.6% of runners (a normal rate is 65-75%) and hitters post a 457 OBP with men on and only 296 with the bases empty. What’s up with that?

Stuff and Selection

The disparity made me wonder if there was something mechanical that changed when Verlander went into the stretch. Did he have a harder time finding the zone, was his stuff not as good, less break, less velocity? I took a look at the pitch f/x data for his first 4 games looked at stretch and wind-up situations. The pitch mix, velocity, and horizontal and vertical movement for his pitches in both scenarios are displayed below.

Verlander actually throws his fastball harder from the stretch adding a full mile per hour. The extra speed on the heater doesn’t hurt his movement at all (those are inches, so a couple tenths of an inch is insignificant.

He throws his curve ball harder in those situations also, and in this case he actually is sacrificing movement. He gets over an inch less drop on his breaking ball. With his change-up he throws it a little harder with a wind-up, but still gets more sink with the extra mile per hour making me wonder if he slows his arm more with the change-up from the stretch.

But the bigger differences aren’t in the pitch path to the plate, but the selection. Verlander throws his fastball 75% of the time with the bases empty, but he starts throwing his curve and change much more with runners on base. The fastball is Verlander’s best pitch, his signature pitch, but he goes away from it when he gets in trouble.

Strike Throwing and Command

In terms of control, he actually throws pitches in the strike zone at a better rate from the stretch. But his rate of strikes is markedly lower. The table below is a break down of his strikes by type of strike.

Verlander doesn’t get swinging strikes or called strikes at the same rate when pitching with men on. A deeper look reveals that from the wind-up, 24 of the 28 swinging strikes that Verlander has  earned this year have come on the fastball. With runners on base, he goes away from the pitch and only had 16 swinging strikes with runners on and only 7 came on the fastball.

Release Point

One last thing I wanted to look at was release point. Does something change that makes Verlander more or less deceptive? Below is the ball’s location at 50 feet from the front of home plate.

At least at the 50 foot mark things are pretty consistent between the two deliveries. There is a cluster of pitches off to the right and a little lower than the main grouping. That cluster occurred in the first 3 innings of the Blue Jays game. I don’t know if his delivery changed that much or if it was a calibration issue. At the very least I think that this portion can be deemed inconclusive.

Based on this pretty limited study it doesn’t appear that being in the stretch itself disrupts Verlander too much. His curve and change-up are altered slightly, but being a geek instead of a ballplayer I don’t know how much those changes in movement effect results. What I can see though is that he becomes much more reliant on those pitches in that situation. So if the pitches are less effective from the stretch, by throwing them more he doesn’t help his chances.

PREGAME: The Tigers look for a series sweep of the Texas Rangers this Easter Sunday. The Rangers mashed Cleveland pitching but haven’t been able to muster much offense against the Tigers starters. Today it will be Edwin Jackson making his second start of the season. His first effort was wasted by the bullpen as he pounded the strike zone (67.4% strikes) en route to a 7.1 inning, 89 pitch start.

Kevin Millwood pitches for the Rangers. He held Cleveland to 1 run in 7 innings on 5 hits, a walk, and 5 K’s. He went with a fastball heavy repertoire (91MPH 4 seam, 90MPH 2 seam), and didn’t seem to get the ball down that well. Still, the results were impressive. Hopefully if he keeps the ball up against the Tigers, they’ll either layoff it or hammer it.

Game Time 1:05

Texas vs. Detroit – April 12, 2009 | MLB.com: Gameday

POSTGAME: It’s a good day to be named Brandon. Inge of course due to his 4th homer (and career number 100) to start the 8th inning comeback and his 2 out RBI single to cap it. Plus he threw in a slick play on a bunt to start the 9th. Part of the reason that the Tigers only needed to come down by 4 is because Brandon Lyon came on and retired 6 straight batters on 17 pitches. It was a very nice bounce back outing, and only his second since he blew Edwin Jackson’s last start.

As for Jackson, he wasn’t as sharp as his last outing, but mostly effective. He pitched out of a big jam in the 5th and in the 6th inning he retired the side on 10 pitches, mostly 95mph+ fastballs to put a nice cap on his day.

Kevin Millwood was frustratingly good. It was a steady diet of 89 mph fastballs and curves that had the Tigers completely fooled.

• Fernando Rodney picked up the save on 5 pitches (and the previously mentioned nice play by Inge)
• Placido Polanco seems to have his stroke going after a slow spring
• A couple of awful at-bats by Magglio Ordonez and Gerald Laird in that 8th inning. Fortunately the rest of the team picked them up.
• Speaking of which, how about the struggling Carlos Guillen lacing a two strike double down the line?

Table of contents for Strike Throwing with Pitch f/x

1. Strike Throwing – Part 1 – Lots of Tables

Way back in 2008 I started to run a series using pitch f/x data to look at strike throwing tendencies. Sadly this is the slowest moving “series” of posts ever. Nonetheless, it’s time for part 3 where we look at how teams do when they get strikes outside of the zone. For this exercise I’m not looking at those generous calls off the corners, but for those strikes when hitters go fishing.

The first table we turn to is the fish rate, or the percent of pitches outside of the strike zone that hitters swung at. This is presented by count. As for the pretty shading, red are lower numbers and green are higher numbers.

2008 Fish Rate: Percent of pitches outside the zone swung at

In what is somewhat of a surprise, it isn’t the 0-2 or 1-2 counts that get the most fishing. That occurs with a full count or a 2-2 count. My guess is that the pitchers are so intent on wasting a pitch, that it isn’t even tempting enough to swing at. The result is a truly wasted pitch. Maybe it sets something up, maybe it doesn’t (that would actually make for an interesting study – set-up pitches).

From the Tigers perspective, it is a little odd that they would get a relatively high chase rate on 0-2 but a really low one on 1-2.  As for the other staff I’m most interested in, the Twins had hitters fishing on pretty much every count. My inclination is that the Twins staff has control that is so good, that even when they miss, they aren’t missing by much.

Moving on, the next chart shows the rate at which fished pitches are put in play, as opposed to just resulting in strikes.

Rate of "fished" pitches put into play

On the whole, there isn’t a whole lot of variation by count. You can effectively throw out the 3-0 rate because only 3.3% of pitches outside of the zone on that count get swung at. The 0-2 whole sees the least success which isn’t surprising. The pitch probably has the least chance of being close to the plate, so if hitters are swinging their chances aren’t that great. And in the case of a 3-1 count it would seem that hitters may be looking for a pitch in a specific spot and would be willing to swing even if it wasn’t a strike if it was in their happy zone. Either that or they get so amped up for a fastball they swing no matter where it’s at.

Taking a Tigers centric focus, hitters didn’t seem to have a hard time putting the ball in play even when it was out of the zone. I don’t really no what to make of this.

Finally, it’s all well and good to get hitters to chase pitches. It would seem to work in the pitchers favor if they don’t have to throw the ball over the plate. But what do the results show? The table below is bases per fished pitch in play, of BPFPIP. I’m sure that acronym will catch on.

Bases per fished ball in play

The Dodgers, Cubs, and Blue Jays appeared to have the most success in limiting the oppositions success when they went out of the strike zone. What I don’t know is if that is a repeatable skill. But what does appear to be true is that getting hitters to fish is beneficial on the whole. The correlation between fish rate and balls per fished… is -.49 meaning that there is a relationship and that as fishing goes up, success goes down.

Now these numbers aren’t park adjusted at all and this is a fairly crude (okay really crude) analysis so take it with a grain of salt. Also the numbers aren’t the least bit surprising, but in my defense I didn’t know that before I started the analysis and it would seem a waste to throw it away.

Table of contents for Strike Throwing with Pitch f/x

1. Strike Throwing – Part 1 – Lots of Tables

In part 1 of the series I dumped some big tables in here that looked at the rate at which teams threw a)strikes and b)the ball in the strike zone. Today we’ll continue along the same path, but we’ll start to differentiate based on count.

Pitches In the Strike Zone by Count

In 2008 teams pitchers through the ball in the strike zone (as defined by pitch f/x) at the following rates:

• 0-0: 48.0%
• 1-1: 45.0%
• 2-2: 43.0%
• 0-1: 40.0%
• 0-2: 27.1%
• 1-2: 34.6%
• 1-0: 49.8%
• 2-0: 51.6%
• 3-0: 49.9%
• 2-1: 51.2%
• 3-1: 55.9%
• 3-2: 53.8%

The table below displays how each team’s pitchers did in each count(click on the table to make it bigger):

Percent of pitches thrown in the strike zone by count

My beloved Tigers are below the average across the board. Meanwhile the strike throwing Twins are considerably above average in terms of pounding the strike zone regardless of the count.

Let me pause for a moment to explain the coloring here. Redder colors denote higher values, greener colors lower values, and yellow in the middle. The deeper the color, the farther away from the mid point. (Excel 2007 can do this automatically under conditional formatting. I never had explored this int he 2007 version and I love it).

With the gradient explanation out of the way we can look for some trends.

• The Yankees and Mets tend to keep the ball out of the strike zone more often than not
• Cleveland pounded the strike zone unless they were behind in the count, and then it seemed as if they didn’t want to give in.
• The Angels were the opposite in that they threw it in the zone unless they were ahead in the count.
• The Red Sox pounded the zone when ahead, and stayed away when they were behind, but were fairly neutral with neutral counts.
• The A’s which have had a history of good pitching, didn’t really throw it in the zone too much in 2008

Pitches Thrown by Count

Of course knowing what happens in a particular count is only part of the battle. It’s probably worth looking at how often each team was throwing in a given count (and while we haven’t gotten to the “what happens when the pitch is thrown part” it probably has a bigger impact on performance I’d guess).

Below is the percent of pitches thrown by each team in a given count, with the same lovely shading.

Percent of pitches thrown in each count

Some stuff I notice here:

• The Twins have the highest percentage of 0-0 pitches, meaning that they are throwing fewer pitches in other counts and having the most short at-bats.
• Despite not throwing a lot of pitches in the zone, the Yankees seemed to work ahead more than behind.
• Baltimore, Oakland, Texas, Washington, Pittsburgh and Detroit threw a lot of pitches from behind in the count.  That’s really not surprising.

Conclusions

None yet to speak of, except that if you’re behind in the count your pitching staff probably won’t fare well – but that is kind of a “duh” conclusion. I’m just presenting data at this point. Up next I’ll continue to dive into this and look at which staffs do the best job of getting hitters to go fishing by count.

Table of contents for Strike Throwing with Pitch f/x

1. Strike Throwing – Part 1 – Lots of Tables

The Tigers walked a lot of people last year. Along the way they threw a lot of pitches, and many seemed to be ill advised. The performance cost Chuck Hernandez his job, jettisoned in favor of an instructor whose students have gone on to gain some renown as strike throwing machines. Armed with a season’s worth of pitch f/x data I’m ready to start delving into this whole strike throwing thing. We’ll start today with some general league wide information.

For those unfamiliar with pitch f/x I’ll have some additional links to more information at the end of this article. The short explanation is a couple of cameras measure the direction and speed a ball is moving shortly in front of the mound. From this the pitch’s path is calculated to within an inch of where it crosses the front of home plate. And it draws the trajectory in the MLB.com Gameday application. On to the data…

Starting at the most simple level, let’s look at the respective rates at which teams threw balls and strikes. The table below shows the breakdown for all 30 teams. (These data are from pitch f/x, which doesn’t capture every single pitch, but should be sufficient).

No surprise that the Tigers rank near the worst in the league in terms or throwing balls. But what is surprising is the difference between the Tigers and the best. The best as you may notice are the Twins, but that gap is 4%. When you have 125 pitches that is a difference of 5 pitches a game being balls instead of strikes or in play. Now it only takes 4 pitches to walk a batter, but those pitches are most likely not all going to one batter. And the difference between the Tigers and the Twins in terms of walks was 238 this year. Put otherwise, the Tigers walked 58% more hitters than the Twins.

You may have also noticed from the table that the Twins lead in another category besides fewest balls thrown. They were first by a substantial margin in terms of their pitches being put in play. It speaks to the organizational philosophy about early strikes and getting the opponent to put the ball in play. For this we’ll move onto the next table which simply looks at how often the ball is thrown in the strike zone.

InZone is simply whether or not the ball crossed the front home plate in within the strike zone. The strike zone is of course 3 dimensional so the ball could clip the zone later, but this is what we have to work with. The Tigers are in the bottom third when it comes to putting the ball over the plate. The Twins are first which explains the lack of walks and the high number of balls put in play.

One other column I added was a buffer zone. I added 1.5 inches to each of the 4 sides of the strike zone. I wanted to see how how often pitches were either in the zone or “close” to the zone. The types of pitches that might entice someone to swing because it’s too close to take. The Twins have a lead here too, but the rankings don’t really change to much across the board so it doesn’t seem that teams vary too much in their ability to throw pitches near the strike zone.

The last table we’ll turn to is the product of some work Dan Fox did with pitch f/x data and is called Fish-Eye. Fox did this to look at batter plate discipline, but I turned it around to see if certain staffs could render hitters less disciplined. The column headings are defined as:

* Square: This is the new metric, defined as the percentage of pitches in the strike zone swung at and made contact with. A high value here (relative to the average of over 86.3 percent) indicates that when the batter offers at a strike he usually makes contact. On the contrary, a lower value indicates hitters who, for reasons such as a long swing, are more apt to swing through strikes.
* Fish: Defined as the percentage of pitches out of the strike zone that the hitter swung at. A higher percentage here indicates that the hitter may have trouble recognizing pitches since he is offering at pitches that would likely be called balls. Average values here are between 29 and 30 percent.
* Bad Ball: Defined as the percentage of pitches out of the strike zone that were swung at where contact was made. This includes foul balls, although there is an argument to be made that a foul ball is not the intended outcome, and so should be discounted in some way. A higher value in this category indicates that, when swinging at bad pitches, the hitter is at least able to get the bat on the ball. Average values lie around 68 percent.
* Eye: Defined as the percentage of pitches in the strike zone on non-three and zero counts that were taken for strikes. A smaller value in this metric indicates a player who recognizes strikes and aggressively offers at them. I excluded 3-0 counts, since a hitter is much more likely to let a strike go by in this situation, and we don’t want to penalize them for that behavior. (I didn’t exclude these counts). Average values here are in the range of 35 percent.

I’ve also added Contact which is the percentage of time a hitter swings and makes some sort of contact.

Earlier we saw that the Twins got a high number of balls in play on their pitches but weren’t hurt too much because of it. Part of the reason is that there weren’t free passes putting guys on. But perhaps another reason is that they have a lot of hitters chasing pitches that are off the plate and not perhaps not making solid contact. It’s a recipe for success. The Twins throw more pitches in the strike zone than anyone, and they get hitters to chase more pitches off the plate than all but one other team.

But oh yeah, this is a Tigers blog. Forgive me if I look longingly at the Twins and with great hope given that the Tigers new pitching coach was heavily involved in development of the guys who did the work – Rick Knapp. As for the Tigers on these measures, they were pretty vanilla and close to the means. The exception was in Bad Ball where hitters made consistent contact with balls out of the strike zone. I don’t know if this is good or bad – yet.

Next steps are to look at the Tigers as individuals, and to look at location and results by count.

In last night’s Detroit-Arizona game there were several moments where I was completely befuddled by CB Bucknor’s strike zone. But I was going to just consider my bias as a frustrated fan and let it go. Then I saw Tom Gage’s game story and a ball-strike call was the lead for his story. So I thought I would go back to the pitch f/x data and post a couple of the at-bats in question. Except for one thing. The pitch f/x data that had been there throughout the game had disappeared this morning.

One of the moments in question was in the top of the 7th inning when Jeremy Bonderman appeared to freeze Justin Upton with a called third strike on a 2-2 count. Except it was called a ball and Upton would later go on to walk. I was IM’ing with Ian and couldn’t believe the pitch, but I chalked it up to being the frustrated fan of a floundering team.

The other egregious call was in the top of the 8th inning when Curtis Granderson was called out an a fastball that was borderline low, and at least 6 inches off the plate. The frustration of the one call combined with the crappy play of my team had me furious at this latest call. But I thought I would give it some time before posting. I try not to blog angry.

But when I went to post the at-bats in question this morning, the pitch f/x data is gone. There are still plots of the pitches, but not as tracked by pitch f/x and not where the actual pitches were. The system was working last night. It worked the night before at Chase Ballpark and the data still persists. It worked in other ballparks last night where the data still exists.

Perhaps it is some kind of data glitch where things were broken only at Chase only after that game. And in fairness I checked the other games that Bucknor has done this year, and all the data is there. So maybe I’m just irrational frustrated fan guy right now who is pissed because his team isn’t playing better, but I did want to graphically show how bad the calls were. Plus I have a long running disdain for Bucknor for other reasons.

C.B. Bucknor’s umpiring acumen has been well documented. In a 2003 Sports Illustrated survey of players he was voted worst umpire. In a 2006 Sports Illustrated survey of players he was voted worst umpire. So at least he’s consistent in his sucking.

Prior to Justin Verlander’s last start we discussed his dip in velocity and a change in arm angle is he was looking to make. Verlander had some improved results in his Tuesday night start against the Rangers. Looking at his pitch f/x data he averaged 93.9mph with his fastball and peaked around 97mph. A definite improvement to be sure.

As for the arm angle, I added his Tuesday night release point data to the previous graph. Now there may be some calibration issues because the latest data was from Comerica Park and the other data was from Chicago. But there appears to be a marked difference, and maybe even an over adjustment.

As has been widely documented at this point, Justin Verlander’s velocity is down significantly. Either it is a systematic change in measurement mechanisms, or it is a real dip. Given the remarkable disparity and the fact it doesn’t seem to be effecting other pitchers the same way, I’m inclined to believe it is real. The question is whether it is by design (slower pitches, more movement, more longevity is one theory), it is mechanical, or it is related to injury.

Verlander recently did a side session in which he was working on his arm slot saying that he had been throwing with a lower arm angle and was working on raising it. While we can’t necessarily look at his arm angle with pitch f/x data we can look at an approximation of his release point.

The pitch f/x system picks up a pitch 50 feet from home plate. The graph below is a comparison of Verlander’s last start of 2007 when he was regularly hitting 97-98mph and his start from April 12th when he was sitting substantially lower maxing out at 93.9mph. I chose these two games because I wanted data from different seasons and from the same ballpark. Both of these starts were in the Cell.

Yes it was a very cold day on April 12th which could have had an effect, but the velocity readings didn’t differ significantly in his other starts.

This data shows Verlander releasing the ball higher in 08 than 07 and more towards the third base side. Now this is far from conclusive, but I do find it interesting. It’s possible that the system was recalibrated from last year to this year. Or maybe the mound is higher. Or maybe Verlander is positioned more towards the third base side of the rubber.

Further investigation will be forthcoming, but this is what I could pull together quickly. We’ll see if it has any effect on velocity. The Tigers don’t seem too concerned with his fastball and they are more concerned with the command of it. And it sounds like the intent of the arm angle change was to address the tilt on the breaking pitch (we can check this with pitch f/x also). But my armchair speculation is that both Verlander’s control problems and lack of break on the curve ball could have been the result of over throwing as he tried to hit those mid 90′s radar gun readings.

From time to time this year (as time permits), I’ll delve in to MLB.com’s pitch f/x data to analyze a starters outing. Tonight we look at Jeremy Bonderman’s start against the Kansas City Royals on April 3rd.

Pitch Mix

This season MLB.com started classifying pitches. This seems pretty convenient, but from what I’ve seen so far the classifications don’t quite match. In the case of Jeremy Bonderman we know he throws both a 2 seam (sinker) and 4 seam fastball, a slider, and an occasional change. The data had Bonderman throwing a splitter, which looks to be a misclassification of his slider. Because of this, I did my own pitch classifications using K-means clustering and some judgment.

The table below shows his pitch mix and average velocity for the 87 pitches tracked by the system today.

	    n     mph
2seam       39   92.0
4seam       25   92.6
change       4   83.8
slider      19   85.6   

It’s not uncommon for Bonderman to rely heavily on his slider, to the tune of a third or more of his pitches typically. Today it was a mere 21%. And while categorizing the various fastballs can be hit and miss, Bondo’s slider is pretty distinctive so I have to believe it isn’t a classification issue.

The plot below shows the horizontal and vertical movement. Technically it’s the difference between the actual pitch and a similarly thrown pitch without spin. Pitches can’t technically rise, so the positive vertical movement is an indication of a resistance to sinking. Don’t worry too much about the physics, but the graph is simply to show that the clustering of the pitches by movement is fairly accurate – especially with regard to the slider.

While there has been much talk of more change up use, it appeared to be pretty limited today and consistent with his use of it in the past.

First Inning

Yes, yes, Bonderman struggles in the first inning. But not today. One of the theories is that he throws too many fastballs. He didn’t deviate today but had pretty good results. His pitch sequence in the first was:

2seam-4seam-2seam-4seam-slider-2seam-slider-2seam

But it worked because he was throwing it for strikes.

Location and Results

We can also see how Bonderman pitched to both right handers and left handers. From the graph below he was pretty effective staying away from left handers and staying down in the zone against right handers. Also, when pitching to righties he picked up a number of outs on pitches on the inner half of the plate. Bonderman’s pitches tail in towards right handers meaning hitters were having to fight off those inside pitches.

The other thing to note here is look at how selective the Royals were. There were very few swings at pitches outside of the strike zone – or at least far outside the strike zone. I wonder if there was a causal relationship between the chasing of pitches and the relatively low number of sliders.

Wearing down

Jeremy Bonderman was cruising early in the game, but in looking at the data it becomes clear that he ran out of steam. That shouldn’t be at all surprising given how early it is in the year. The plot below is the speed of his 2 seam fastball as the game progressed.

In fact, his last 5 pitches were pretty much identical. They were all 89 mph fastballs with very similar movement, and they showed an increasing tailing action. The same action that we saw when Bonderman was fighting his elbow late in the year. I’m not suggesting he’s hurting, just that he was fighting to find anything that last inning.

Scouting

Now for those of you who saw the game, does the data mesh with reality? I was at work today and was only able to listen. Does the data make sense to those of you who saw the game?

Jim Leyland has come out on several occasions and said that Jeremy Bonderman is one of the keys to any success the Tigers might enjoy in 2008. Bonderman’s second half swoon, which I attribute largely to his elbow pain that he finally fessed up to, clouded what was starting out to be a phenomenal season. An ERA of 8.50 over his last 9 starts, combined with the arm troubles meant that Bonderman finished with the highest ERA and lowest innings total since his rookie season. Like with Dontrelle Willis, we’ll delve into the pitch f/x data and see what we can find out about the veteran 25 year old pitcher.

What does he throw

Jeremy Bonderman throws 2 flavors of fastballs, a nasty slider, and the very occasional change-up. I used K-means clustering on the horizontal and vertical movement parameters, and the velocity to classify the pitches. The relative frequencies and mean velocity are listed below:

And the following graph shows the movement of Bonderman’s pitches from the catcher’s perspective in inches.

Both fastballs tail away from left handed hitters, with the 2 seamer having more lateral movement. The slider drops about 2 feet and dives away from the right handed hitter. I’m actually a little surprised that we don’t see more horizontal movement on the pitch. The change-up is hardly thrown, especially in relation to the number of stories that come out each year about it’s development.

Part of the problem with the change-up is that he doesn’t get enough downward movement on the pitch, because it ends up more like a batting practice fastball. Mike Fast when working with pitch f/x data likes to calculate the spin rate. Looking through Fast’s past work where he’s done these calculations, the spin rate for a change-up is in the range of 1500, give or take a few hundred depending on the pitcher. In the case of Bonderman his mean spin rate is 2155. As a point of comparison the spin rate on Bonderman’s fastballs are 2333 for the 2 seam and 2421 for the 4 seamer. The more backspin, the less sink so Bonderman’s change-up isn’t really behaving like a change-up.

Where does he throw it?

The scatterplots below show the pitch location, by pitch, from the catcher’s perspective. Each pitch has a graph for the pitches thrown to left handers and right handers. Reference lines are drawn such that the resulting box corresponds with a typical strike zone. So 0 on the horizontal axis represents the center of home plate and 0 on the vertical axis represents, well, the ground.

The slider is the bread and butter pitch where Bonderman gets the most swings and misses, many of which take place well outside of the strike zone. He also gets a fair number of called strikes right on the third base side black as the ball swings back – or at least fools the umpire enough into thinking that it is.

the 4 seam fastball is the one that Bonderman seems to have the most trouble controlling. Either by design or by accident, it seems to be up more than his other pitches.

The 2 seam fastball is typically bearing in on the right handers and tailing away from the left handers. From watching Jeremy the last several years, his ability to harness that tailing action will often dictate the good starts from the bad starts.


Bonderman hardly throw the change-up to right handers, and hardly throws it period with the pitch only about 5% of his arsenal. If you’re interested int he full size graph feel free to click the thumbnail.

If you prefer a tabular view, that is included as well here:

Left Handed Batters

Right Handed Batters

Those second half woes

Delving into this analysis I thought I’d be able to turn up something really cool regarding Bonderman’s second half struggles. Maybe a consistent failure to locate a pitch, or his slider flattening out and not biting. There had to be some graphical evidence of whatever caused Bonderman to go from striking out 4 batters for every free pass he issued to walking a batter every other inning and allowing 8 homers in his final 48 innings of the season. Alas I have nothing.

I fail to see any discernible changes in pitch location when looking before and after his 7/13 start against the Mariners.

In terms of movement, graphically it showed nothing so I went with the hard measurements below. Pfxx is horizontal movement, adjpfx_vert is vertical movement, and adj_start_speed is velocity.

Results for Pitch = 4seam 

Variable         Pre-Post    Mean  StDev
pfxx             1st Half  -7.528  1.610
                 2nd Half  -7.310  1.939

adjpfx_vert      1st Half  -7.596  1.504
                 2nd Half  -7.866  1.721

adj_start_speed  1st Half  93.220  1.143
                 2nd Half  92.865  1.460

Results for Pitch = Change 

Variable         Pre-Post     Mean  StDev
pfxx             1st Half   -9.109  1.255
                 2nd Half   -8.937  2.075

adjpfx_vert      1st Half  -14.172  2.403
                 2nd Half  -13.016  2.154

adj_start_speed  1st Half   84.264  1.691
                 2nd Half   85.464  1.637

Results for Pitch = Slider 

Variable         Pre-Post     Mean   StDev
pfxx             1st Half    1.104   1.317
                 2nd Half   0.2258  1.5330

adjpfx_vert      1st Half  -18.720   1.673
                 2nd Half  -20.017   2.537

adj_start_speed  1st Half   84.598   1.456
                 2nd Half   83.409   1.967

The differences are minimal, and given the newness of the system and the various stadium deployments I’d consider the bulk of this as random variation. The only possible change would be the slider which had more drop and less horizontal movement in the second half, perhaps an indication it isn’t quite as tight?

Also notice that there weren’t appreciable changes in velocity, which is good news from a shoulder health perspective.

Outlook

Once again we head into a season wondering if this will be the year for Bonderman. Last year I really thought that “this was it” because of the way he threw for the first 100 plus innings. People like to point to the lack of a legitimate 3rd pitch as a real limiter for Bonderman. His success early on last year shows that either the change-up was legitimate, or he doesn’t in fact need the 3rd pitch. What he does need is health.

Bonderman will only be 25 this season, but has over 900 major league innings on his right arm. Twice now he has suffered tenderness in his elbow late in the season. Will this be a recurring problem? Only time will tell. But until Bonderman can put together a full season people will continue to question 1. his stuff, 2. his ability, 3. his mental toughness, etc. He does appear to be pretty ripped coming into camp this year and he’s sporting a new tattoo, so he’s got that going for him.

One final note, I know you’ve read several pages now with no mention of Bonderman’s first inning struggles. You know the struggles. A quarter of the runs Bonderman has allowed in his career have come in the first inning. The thing is, when I wrote my program to parse the data and build my pitch database I neglected to grab the inning. The reason for this: I’m an idiot. Changing the code is easy enough to have things re-parsed. But everything after that point becomes time consuming, so it ain’t happening right now. Sorry.

On Friday Lynn Henning wrote a detailed look at Dontrelle Willis with a heavy emphasis on scouting. I found the article fascinating from the stand point of getting a better understanding of Willis’s repertoire as well as the thought processes that went along with approving the deal for the lefty. He was after all coming off a pretty rough year. I also viewed it as a chance to dust off that pitch f/x database I’ve had sitting dormant and explore whether or not the reports meshed with what the system had reported.

A little background

As a precursor, pitch f/x is a system deployed during the 2006 playoffs, and subsequently throughout MLB over the course of the 2007 season. It captures the velocity and trajectory of a pitch using several cameras and a lot of math. I wrote several articles using data from the system during the course of last season. Since the last time I delved into the data, there has been a ton of analysis done. Joe Sheehan, Dan Fox, John Walsh and others have cranked out some awesome stuff using this. But Mike Fast has really taken to the data and not only has bookmarked all the research to date, but has conducted a considerable amount on his own. I mention Fast so much because I borrowed heavily from his methods and I am very grateful.

The system wasn’t in place at all stadiums for the full season. The bulk of the data presented below was captured from seven starts, all of which came from July forward.

What does he throw?

From Henning:

His repertoire is as unique as his delivery. While his sinking, two-seam fastball is conventional, Willis throws a breaking ball that is not quite an overhand curve, and not a slider. It is closer to a “10-to-4″ curveball than a classic, diving “12-to-6″ breaking pitch.

His best weapon in the estimation of Wiley is Willis’ version of a change-up. It’s an off-speed fastball that cruises from 84-87 mph with tremendous sink.

Henning also notes that Willis’s 4 seamer tops out at 93 to 94.

Using a process called K-means clustering I went into Dontrelle’s pitch f/x data and grouped pitches by similarity of velocity and movement. The results are displayed graphically below:

The graph is from the catcher’s perspective and represents the movement that a hitter would see on Willis’s pitches including gravity. So Willis’s curve-ish slider dives towards the feet of right handed batters.

The 2-seam fastball has a little more sink than the straighter 4-seamer with both tailing into the left handed batter. The change-up does appear to tumble as described, with more sink than the 2-seamer but less lateral movement.

A tabular view including the average velocity is below:

This seems to reason check as well. An average heater in the low 90′s with the sinker a couple ticks slower. The change-up is on the lower end of what the scout explained, but that could be my issue. The number of breaking pitches seems pretty low, and the difference between the change and slider is pretty small mathematically. There’s a decent chance that I’ve misclassified some of the sliders as change-ups, dropping the overall velocity of the pitch.

Where does he throw it?

Dontrelle had problems with location last year.

He must be able to spot his four-seam fastball, which tops out at 93-94 mph. In particular, he must be able to throw it high and tight to right-handed hitters.

When Willis misses — and he missed a lot last year when that same fastball swung back across the plate — the ball can end up on a ride into the seats.

“His location was not nearly as good last year,” Russell said. “He was a little off as far as being able to pitch inside (to right-handed hitters). Last year, he was more middle-out, much more than he had normally been.

We can also take a look at where Willis was throwing to last year, and how that differentiated for left and right handed batters. Below is the graph for the 4-seam fastball. The pane on the left is for left handed hitters, and the pane on the right is right handed hitters. The center box in each pane is an approximation of the strike zone.

There certainly doesn’t appear to be a trend of jamming the right handed hitters with the 4 seamer, and when it was on the inner half of the plate it was put in play -with considerable success. Right handed batters against the 4-seam fastball hit .404 when putting the ball in play. Below is a table with the outcomes of each pitch to right handers.

Right Handed Batters

And now for the left handers, of which there weren’t many that faced Dontrelle.

Left Handed Batters

For the sake of completeness, here are the pitch location graphs for Willis’s other 3 pitches as well.

Willis keeps the 2-seamer away from both lefties and righties.

But he comes inside with the change-up to right handers. Again, some of those maybe sliders. And speaking of the slider, Willis doesn’t throw it in the strike zone a whole lot, but he manages to get a decent number of strikes via swings and misses

Staying ahead

By missing spots, the thought was that Willis was pitching from behind and having to go more fastball heavy.

He pitched behind in the count too many times and got into fastball counts.

Turning our attention to pitch count we see that Willis was throwing more pitches behind in the count than in a pitchers count.

And graphically…

When behind in the count, Willis goes to one of his fastballs nearly 80% of the time. It keeps him from throwing the breaking pitch, which is quite effective but hard to control. In fact he only seems to go to the pitch with any consistency when he has two strikes. Also worth noting is the reliance on the fastball to start off an at-bat as well. The 85% fastball rate on first pitches is only surpassed when the count is 3-1.

As for whether pitching behind is new for Willis, it is. At least to a certain extent. In 2004 and 2005 Willis was staying ahead of the hitters. In 2006 it tipped the other way and last year was the widest gap.

Pitches thrown when Ahead/Even/Behind in the Count

Conclusions

Everything works out. The scouts validate the data and the data validates the scout – or at least the two are reasonable consistent. Any issues at the end of the year didn’t appear to be velocity related – at least as far the pitch f/x data reveals. This is consistent with the scouting reports. The issues seem to be centered squarely on control. I don’t know that his pitches were any more hittable than in past years, just that he ended up in less desirable situations because he wasn’t able to locate like he had in the past.

Whether this was due to a change in mechanics, a subconscious reaction to a poor defense, or the result of some arm pain remains to be seen. The challenge for Chuck Hernandez and Willis is to regain that control. It will be interesting to see how quickly Hernandez can get comfortable diagnosing and working with Dontrelle’s unorthodox delivery. Or perhaps there will be something to the change of scenery and playing in front of a full house. Still, given the 3 years the Tigers committed to Willis I have to believe they have a handle on the situation and are confident it can be corrected.