I was poking around the pitch type leader boards and noticed an interesting fact: Trevor Hoffman had the best fastball of all qualified relievers in 2009. His 85mph pitch beat out the likes of Phil Hughes‘s and David Aardsma’s, which were almost 10 mph faster.

When he re-signed with the Brewers I noted that he has, over the course of his career, been able to maintain a HR/FB well below the 10% expected value. Playing in Petco for a big part of his career no doubt helped, but beyond that it seems like Hoffman has the ability to depress his HR/FB, which runs counter to some of the prevailing ideas about a pitcher’s ability to control his balls in play.

These two facts, the fact that he had the best fastball of 2009 and his historic ability to depress, inspired me to look at Hoffman’s pitchf/x numbers in a little more depth. Let me say that this analysis just scratches the surface of what makes Hoffman so great. His changeup is devastating and I am sure plays a big role in his HR prevention and probably makes his fastball better. In addition, his fastball has a lot of “rise,” which plays a big role in his high IFF%. But I am going to focus on HR prevention with his fastball, and specifically HR prevention against LHBs.

A lot of attributes determine if a pitch is going to be hit for a HR, but one of the most important is its location. Obviously the height of the pitch plays a big role, but here I am going to look at the horizontal location of the pitch. Here is how HR/FB varies for LHBs against all pitchers (not just Hoffman).

So pitches middle-in are hit for HRs the most often. Now let’s look at where Hoffman locates his fastball. Gold is Hoffman and gray the average four-seam fastball to a LHB.

Hoffman’s distribution is much narrower than average. He has been very good at locating his pitch in the same horizontal area with little spread. This should not be surprising: the fastball is just 85 mph, so for him to be successful, he needs that pinpoint command. And he puts the pitch about as far away as he can and still be in the strike zone. That is where LHBs have the least power.

I usually just display the 2-foot strike zone that John Walsh described, but in this graph I add the dotted line for the specific strike zone to LHBs. The zone is called differently to LHBs, with the outside and inside edge shifted away. That makes Hoffman’s pitch locations even better. He pitches more to the extreme outside where umpires often call pitches 14 or so inches from the center of the plate strikes against LHBs.

As I said, there is a lot going on and this just scratches the surface, but Hoffman’s ability to locate his fastball very well on the outside quarter of the plate to LHBs, I think, plays a huge role in his abnormally low HR/FB rate.

Congratulations to the 2009 World Series Champion New York Yankees. They were clearly the best team in the league during the regular season and dispatched the Twins, Angles and Phillies with minimal drama to take the World Series. In the process, many of Joe Girardi’s decisions have been questioned here and elsewhere, but one that has gotten near universal support (outside of Minnesota, Los Angeles and Philadelphia) was his heavy use of Mariano Rivera to get six outs an appearance and in non-save situations. Rivera’s performance justified that decision, as he gave up just one run on 10 hits over 16 innings.

It is easy to concoct a narrative of Rivera as Superman against whom there is no chance a run will be scored. MGL neatly dealt with this narrative.

For that matter, I don’t buy into the Rivera post-season mystique either.  He is a great reliever, regular or post-season.  That is why he has had such phenomenal success in the post-season – because he is a great pitcher!  However, there is no such thing as “magic” for any player.  There is nothing a player can do about the “Lady Luck” regardless of how good they are.  Mariano does not throw every pitch exactly where he wants to and he does not strike every batter out.  Sooner or later he is going to implode as any pitcher can on any given day, even when he has his best “stuff” (he seems to have the same stuff every outing).  One bad pitch, one bad call by an umpire, one batted ball that does not get caught or falls into the right spot, one batter that happens up square up a pitch, etc.

I like this viewpoint. Rivera is a great pitcher, so we should not be surprised when he has an amazing run of 16 innings. But that does not mean that he can will himself to pitch 16 one-run innings whenever he chooses. So in this post, when I look back at his amazing performance, I don’t want it to sound like I think this was inevitable and there was no way he could have given up any runs. Rather, I am just looking back and seeing how it happened.

Using the same method I used in the Lee post, I am going to look at those innings for a pitch location point of view. Pitches are color coded, those swung at full color and taken faded, strikes encircled, outs with triangles and hits exed. Full color pitches with no markings were fouled off.

As I have said before, the amazing thing about Rivera is how he can location his cutter on either edge and have few end up in the heart. Against LHBs he went all cutter and mostly pitched inside. The graph is a little busy, but you can see the few times he did go outside or hit the fat of the plate he got a good number of called strikes (faded for taken and encircled for strikes), so it looks like batters were looking inside. On those inside pitches he got tons of fouls and outs, but fewer swinging strikes than I would expect.

Again you see the bimodal distribution of pitches either along the inside edge or outside edge. Against RHBs he mixes in his fastball. Batters swung at it more often than his cutter and made contact at a good rate, but they were fouls or outs. With the cutter he got lots of whiffs up-and-in, called strikes down-and-in, and got more swings and contact, again mostly outs or foul balls, away.

Anyway that you look at it, another sixteen incredible innings in the career of the best relief pitcher ever.

Tonight Pedro Martinez takes the mound again for the Phllies. Dave C noted before the series that Martinez is still a solid pitcher, but very prone to giving up the long ball. That is just what we saw in Game 2: he got 8 Ks to just 2 BB over 6 innings, but gave up two solo home runs. That is what we have seen with Pedro over the past couple years: very good at limiting walks, solid strikeouts, but horrid ground ball numbers leading to lots of HRs.

To look at this tendency of solid K/BB ratio and bad GB% I am going to take a look at a couple of his pitches. Here is the breakdown of what he throws to RHBs and LHBs.

+-----------+-------+-------+
|           |  vRHB |  vLHB |
+-----------+-------+-------+
| Four-Seam |  0.43 |  0.28 | 
| Two-Seam  |  0.15 |  0.26 | 
| Cutter    |  0.13 |  0.10 | 
| Curve     |  0.16 |  0.12 |
| Changeup  |  0.13 |  0.24 |
+-----------+-------+-------+

Against RHBs he favors his four-seam fastball, which is an extreme flyball pitch, getting just 27% ground balls. Where he locates it in the zone is a big reason. Here is the height and horizontal locations for the pitch compared to the average four-seam.


It seems that in exchange for this height, which results in all those fly balls, he is able to hit the outside edge of the zone better than most pitchers and is able to keep the pitch out of the heart of the plate

Against LHBs he mixes in his two-seam fastball more often, and this pitch shows a very interesting pattern. Here I show the horizontal location of the two pitches to LHBs for Martinez and average. The four-seam is orange and two-seam blue and Pedro’s pitches are in the full color and average in the faded.. Since there are four different lines the graph is a little cluttered.

With his four-seam fastball, just like we saw against RHBs, he is better than average at locating the pitch on the outside half of the plate and avoiding the inside half. But with his two-seam, it looks like most of the time he goes for the outside, but then some times the inside. This is an interesting bimodal distribution I have only seen before with Mariano Rivera. It looks like Martinez will at times go inside against LHBs with his two-seam fastball. This ability to keep a pitch out of the heart of the zone while hitting both edges of the plate with the same pitch type seems to me to be quite rare.

Last night’s Phillies win closed the book on Cliff Lee’s amazing playoff performance, unless he comes back on Thursday out of the pen on two days rest. A couple weeks ago, Marc looked at the prospects the Phillies gave up to get Lee, but after Lee’s dominating playoff performance these guys are afterthoughts for Phillies fans.

Lee has pitched in five games over three series and the Phillies have won each of those five games.

+-------+-----+----+----+----+----+----+
|       |  IP |  H |  R | HR | BB |  K |
+-------+-----+----+----+----+----+----+
|   NYA |   7 |  7 |  5 |  0 |  3 |  3 |
| @ NYA |   9 |  6 |  1 |  0 |  0 | 10 |
|   LAN |   8 |  3 |  0 |  0 |  0 | 10 |
| @ COL | 7.1 |  5 |  3 |  0 |  3 |  5 |
|   COL |   9 |  6 |  1 |  0 |  0 |  5 |
| Total |40.1 | 27 | 10 |  0 |  6 | 33 | 
+-------+-----+----+----+----+----+----+

He pitched at least 7 innings in each game (and completed two of them). Overall, it works out to 40 innings of 1.56 ERA, 5 K/BB, 0 HR postseason baseball.

Dave C. noted after Lee’s first game against the Yankees that the Fox broadcast team was drooling over his curve, but his changeup was more impressive. Here I give a visual representation of this fact. I plotted the location of these two pitches over the course of the playoffs. As I often do, the full color dots represent pitches swung at and faded ones those taken. I encircle strikes (taken strikes are encircled faded dots, whiffs encircled full color), put triangles around balls in play that were outs, and Xs on top of hits. Solid color pitches with nothing around them were fouled off.

You can see that Lee is actually fairly comfortable mixing in his changeups to LHBs and got a nice number of called strikes on the outside edge of the plate. Overall, lefties only got one hit against his change and curve over the course of the playoffs.

Against RHBs his change is a huge part of his repertoire and he pounds it away and mostly up. Batters swung at it a lot, but didn’t have much luck mostly making outs or fouling it off. He got a fair number of called strikes on his change right down the middle of the plate.

The Phillies got just what they wanted in Lee with a mid-season trade. In the regular season he helped them keep their place comfortably atop the NL East, and now in the playoffs he has pitched in five of their wins (including their only two World Series wins), giving them 40 impressive, and bullpen-preserving, innings.

One interesting match-up in the upcoming World Series is all of the great left handed Phillies hitters against a Yankees’ rotation that features two solid left handed pitchers, CC Sabathia and Andy Pettitte. The Phillies will face a lefty starter in at least half of the World Series games no matter if the Yankees go with a three or four man rotation or how many games the series goes.

The biggest issue for the Phillies is Ryan Howard, who is a very bad hitter against LHPs.

On the splits page, you can see that the biggest difference is in his ISO and K rate. On the other hand, in recent years, the difference in his walk rate against LHP and RHP has decreased. Looking at a pitch-by-pitch basis he takes more called strikes against LHPs than RHPs (13% of pitches versus 11% of pitches). His whiff rate increases from 28% against RHPs to 38% (Mark Reynolds territory) against LHPs. Here is how his whiff rate varies by horizontal pitch location:

Interestingly, he whiffs less on inside pitches from LHPs than RHPs. Next, let’s check out his slugging on balls in play:

Here you see a big difference. Against RHPs he maintains high power across most of the plate (it peaks in the middle of the plate, not surprisingly). But against LHPs his power is relegated to the middle-away, and it falls off sharply inside.

Putting the graphs together explains his overall weakness against LHPs. On inside pitches, he doesn’t whiff that often, but has little power. On the outer portion of the plate, he can hit for some power, but he whiffs at a huge rate.

If you project Howard as roughly a .450 wOBA hitter versus RHBs and roughly a .300 wOBA hitter against LHPs that works out to a 0.13 run difference per at-bat. That is over half a run every four at-bats, an enormous difference and significant cause for concern for the Phillies in the games they face Sabathia and Pettitte.

Andy Pettitte pitched a solid game last night, giving up just one run over 6.1 innings, and helping to send the Yankees to the World Series. Pettitte had a good, but not great, year. Actually, his peripherals were quite poor compared to his recent past. His walk rate was at its highest since 2000:

And his GB and FB numbers were the worst they have been in years:

On the other hand, his K/9 rate was in line with his recent past.

On a per pitch basis, the culprit for these changes seems to me to be his four-seam fastball (he throws a four-seam fastball, a two-seam fastball, a cut fastball, a changeup and curve). It went from being in the zone 53% of the time (pre-2009 pitchf/x era) to just 50% of the time this year. And the ground ball rate on it went from an amazing 49% (very good for a four-seam fastball) to just 36% this year.

Pettitte throws all three of his fastballs to both LHBs and RHBs (about 75% of his pitches to both RHBs and LHBs are fastballs), and the variation in their movement and location is a major weapon for him. I used the FanGraphs pitchf/x section color scheme to display their locations below: green for four-seam fastballs, dark blue for two-seam and light blue for cutters.


The two-seam is thrown to his arm-side (away to RHBs and in to LHBs) and the cut fastball down and to his glove-side (in to RHBs and away to LHBs), while his four-seam thrown in a larger area within the zone, but generally up and to his arm-side. This ability to throw all three fastballs to both RHBs and LHBs is a solid skill. His cutter is his best pitch by our valuations. It gets lots of out of zone swings and whiffs.

Even with the poor BB and GB peripherals his bottom line did not change much, as his numbers were buoyed by a BABIP and HR/FB both lower than his career averages. As a result, he provided solid value to the Yankees this year, 15 millions dollars worth. This past off-season saw him sign a 5.5 million dollar deal with which he was not entirely happy, but with bonuses it was worth $10 million. It will be interesting to see what Pettitte does this off-season. Even if his BABIP and HR/FB regress to career levels and his BB/9 and GB stay as poor as they were this year, he is still a good enough pitcher to provide a solid 10ish million dollars worth of value in 2010.

Adrian Beltre’s contract was often lumped in with other poor free agent signings because his offensive numbers in his four years with Seattle never came close to his 2004 career year with the Dodgers. But it was a good contract. Beltre provided 67.2 million dollars of value, a small premium over his 64 million dollar contract.

Detractors looking at his offensive numbers ignored his Gold Glove-caliber defense at third. He averages better than 10 runs saved with his glove over the average third basemen during a season. Couple that with his league average bat and you have a very good player.

The thing about that league average bat is that it would play much better away from Safeco, which kills RHBs. Beltre came to Seattle in 2005; check out his home/road splits since.

This year he was injured and had a down year, but in the three previous he had solid away wOBAs. In neutral parks you expect a hitter to have a better home wOBA, so in a neutral home park he should outperform those already good away wOBAs.

To examine this further, I looked at his slugging rate on balls in the air (all non-grounders), play by angle, for 2005 to 2009, separated for home (Safeco) and away. The lines are estimates based on the data with standard error indicated.

You can see Beltre’s power is to left, typical pull power for a RHB, and that he gets more power in away parks through much of left and left-center field. The differences look slight, but many flies and lines to left over the course of a season makes up for the big difference. Interestingly, he also gets more power in away parks to extreme right field, where I had thought Safeco was a little more generous.

Beltre is an excellent defensive third basemen, and in many other ballparks he should be an above average offensive player. He has the chance to be a solid free agent signing once again.

Yesterday, Dave C. noted Vicente Padilla’s extreme platoon splits and how he matches up poorly against a lefty-heavy Philly team. I wanted to expand on this with a pitchf/x look at the issue.

As Dave noted, Padilla throws a ton of fastballs. He is fourth among starters behind sinker ball specialists Aaron Cook, Mike Pelfrey and Rick Porcello, in fastball percentage at 75%. Fastballs, on a pitch-by-pitch basis, show a large platoon split. So a pitcher with lots of fastballs should show an more extreme split. Curveballs and changeups generally show little platoon split. Padilla throws a splitter/changeup and a curveball, but he does so rarely and they are poor (changeup/spiltter) or average (curveball) pitches.

Padilla throws both a two-seam fastball and four-seam fastball. To RHBs, he throws the four-seam 35% of the time and two-seam 45% of the time, so to RHBs he throws fastballs 80% of the time. To lefties, he goes 55% four-seam and 16% two-seam. This makes sense: of fastballs, two-seamers have a larger platoon split, so it would be better to throw them more to RHBs and limit them against LHBs.

I plotted the location of all of this fastballs (green four-seam and blue two-seam) to LHBs and RHBs this year. The figures are messy, but should give a general idea of the location he goes with the two pitches.

To righties, you can see a kind of diagonal orientation of his two-seam fastballs running from up-and-in through the center of the zone to down-and-away. This is good: within the zone, pitches do best up-and-in and down-and-away. His two-seam to RHBs is thrown regularly to these locations. With his four-seam he mostly stays away.

To LHBs, the orientation of his two-seam would be opposite of ideal, where hitters do better. Because of this, he does not throw it as often and there is no general trend to where it ends up. He is left throwing his four-seam fastball much more, concentrating it on the outside half.

The differences are seen in the value of the pitches. Both of the fastballs are good to RHBs (0.4 runs above average for his four-seam and 0.7 for his two-seam), but against LHBs things fall apart (0.8 runs below average for the the four-seam and a horrid 2.5 runs below average for the two-seam).

As Dave noted, this is a really questionable call against a team with some good lefties.

Jimmy Rollins is having a pretty bad year at the plate, with a wOBA of .316. Before this year, going back to 2004, he has had a wOBA of at least .341, and had four years above .350. That was very good offensive performance for a shortstop. Not a big walker, the offensive value came from contact skills, a good BABIP and solid power. Well, this year the second of those three fell out from under him. His BABIP dropped to 0.253 taking his batting average and on-base percentage (under .300) with it.

For a speedy switch hitter, often batting from the first base-side batting box, a BABIP of 0.253 is surprisingly low. In fact, it is the third lowest in the game, down with guys like Carlos Pena, Aubrey Huff, David Ortiz and Bengie Molina. His IFFB% is 13.7%, thirteen in the league, and that probably plays some role, but there has to be more to it. To look deeper I went to Baseball Reference which breaks up BABIP by hit type. Here they are for Rollins’ career, for Rollins in 2009 and the NL 2009 average.

BABIP
+--------------+---------+---------+---------+
|              |  Career |    2009 | Average |
+--------------+---------+---------+---------+
| Ground Balls |   0.225 |   0.213 |   0.235 |
| Fly Balls    |   0.095 |   0.094 |   0.142 |
| Line Drives  |   0.725 |   0.629 |   0.718 | 
+--------------+---------+---------+---------+

It looks like the big issue is with his line drives. They are falling for hits way less often than in his career or league average. He has 118 line drives this year. If he had his career average BABIP on those he would pick up about 10 extra hits, which would bring his BA up to .263 and his OBP .304. Better, but still off his career numbers.

To look at this a little closer I plotted the heavily smoothed frequency distributions of the angles in play of his line drives for 2009 versus 2005 to 2008 (the years the data are available from GameDay).

It does look quite different. He is hitting many more to right field; in addition, these are turning into outs much more frequently.

BABIP on Line Drives
+----------+---------+---------+
|          |  Career |    2009 |
+----------+---------+---------+
| Angle < 0|   0.707 |   0.650 |
| Angle > 0|   0.784 |   0.620 |
+----------+---------+---------+

Usually the line drives he hits to right field fall in for hits at a very high rate. This year he is actually hitting more to right field, but they have a much lower BABIP.

Even breaking the data down this far it is still hard to say how much of this is bad luck and how much of it is something wrong with Rollins. As Jack talked about earlier today, this is a place where the HITf/x data would really come in handy.

In Game Two of the upcoming ALCS, Jered Weaver will face the Yankees in Yankee Stadium. Facing the Yankees anywhere is a tough task this year — they lead the league in HRs and BBs, but are in the bottom five in strikeouts. Usually players, and teams, with lots of walks and home runs also strike out a lot. Not the Yankees, they have all the good without the bad. At least part of the reason for all those HRs is that the new Yankee Stadium seems to be quite HR-prone, particularly for left-handed batters.

So Weaver is up against a tough opponent. The added issue for Weaver is that he is one of the most extreme fly ball pitchers in the game. Over half of his balls in play are fly balls, second among starters to Ted Lilly. The big culprit is his four-seam fastball, off of which 62% of balls in play are fly balls. Those fly balls are especially dangerous against a team with the Yankees’ power in a park like the new Yankee Stadium.

When I looked at Lilly I saw that his four-seam fastballs are higher up in the zone than the average four-seam fastball, probably explaining his fly ball rate. Surprisingly, this is not the case for Weaver’s.

What is striking about Weaver’s fastball is the amount of “rise” it gets. That is, the number of inches less it drops than expected from gravity — due to its backspin — as it travels to the plate.

It looks like even though his four-seam fastballs end up at about the same height as average four-seam fastballs, the added “rise” causes them to be hit in the air more often. An added benefit of this is that he has a high IFFB%, but the HRs will come, too.

Jered Weaver is a very good pitcher who gives the Angels a solid chance to win in Game Two, but his fly ball tendencies will make him prone to HRs against a team with lots of power in a HR-park. The Angels really didn’t have much choice, though; Scott Kazmir gives up tons of fly balls, too, so switching Kazmir to Game Two and Weaver to Game Three in Anaheim, a larger park, would make only a slight difference.