Greinke's pitches

Zack Greinke has undoubtedly been the talk of the baseball world following Sunday's Royals/Brewers trade that sent Greinke to Milwaukee for four Brewers prospects.  Since 2008, Greinke has established himself as one of the best pitchers in the game; his 19.6 wins above replacement (Fangraphs' implementation) is fourth best in the majors over that time behind Roy Halladay, Cliff Lee, and Tim Lincecum.  Since Greinke is the "hot topic," I'll give a brief look to the pitches that Greinke throws and what results they got in 2010.

Greinke throws a four-seam fastball, a two-seam fastball, a hard slider, a changeup, and a few different types of curveballs (more on this later).  He fed hitters with a near-even mix of four-seamers and two-seamers last year, and threw all three of his off-speed pitches about 13% of the time.  The charts below show the spin deflection from the catchers' perspective for the five pitches and the velocity distribution for each pitch.  (click to enlarge) 

As the velocity distribution shows, Greinke covers a lot of different of speeds.  His fastball can reach the upper 90s, and he'll drop his curveball down to around 60.  He even threw a 67 mph eephus changeup this year in an at-bat against the Braves' Kris Medlen.  The curve, in particular, is very interesting --- as Harry Pavlidis showed us last year, the curve looks like three distinct pitches.  Splitting the curveballs by game into three different speed buckets, you get a "slow" curve at ~70 mph, a "medium" curve at ~75 mph, and a "fast" curve that borders on slider territory and averages ~80 mph:

	% of total CU	Velocity	pfx_x	pfx_z
Slow	.244	69.28	4.88	-2.42
Regular	.433	74.53	4.38	-1.83
Fast	.323	80.35	4.45	-1.55

(pfx_x is horizontal spin deflection with higher numbers indicating a pitch that moves away from a right-handed batter.  pfx_z is vertical spin deflection in which 0 is a theoretical spinless pitch.)

I'm not sure if it's more effective to think about the curve as three distinct pitches or as one pitch that gets a lot of speed variation.  (Thoughts from those who get to watch Greinke more than I do: your thoughts would be appreciated.)  In the tables below are some pitch results for Greinke's pitches, with all the curves lumped together.

	#	Pitch%	Swing Rate	Whiff Rate	Zone Rate	Chase Rate	Watch Rate	RV/100	xRV/100
FF	1093	.318	.434	.181	.514	.275	.416	-0.46	-0.92
FT	975	.284	.454	.081	.441	.257	.295	-0.87	0.69
SL	491	.143	.536	.376	.322	.426	.234	-2.32	-2.24
CH	454	.132	.476	.208	.374	.345	.306	0.96	0.54
CU	406	.118	.345	.193	.478	.217	.515	2.06	-0.47
	3430		.450	.191	.442	.436	.262	-0.37	-0.40

Swing rate - swings divided by total pitches

Whiff rate - all swinging strikes divided by total swings

Zone rate - pitches in the strikezone divided by total pitches

Chase rate - swings at pitches out of the zone divided by total pitches out of the zone

Watch rate - pitches not swung at in the strikezone divided by total pitches in the strikezone

RV/100 - linear weight run values per 100 pitches, a stat that John Walsh (among others) popularized in February 2008.  These values (kindly provided to me by Dan Brooks) do not perfectly reflect the 2010 run environment.

xRV/100 - expected linear weight run values per 100 pitches with league-average values based on batted-ball types substituted for actual outcomes

	GB Rate	FB Rate	LD Rate	PU Rate
FF	.307	.331	.209	.153
FT	.478	.265	.217	.039
SL	.500	.280	.159	.061
CH	.600	.171	.219	.010
CU	.429	.397	.127	.048
	.456	.279	.199	.066

Judging from these numbers, it would look like his two best pitches are his slider and his four-seamer.  The two-seam fastball is thrown to contact and doesn't get enough ground balls to be an extremely effective pitch, and the change and curve don't look that special (though that is a sweet groundball/flyball ratio for the change).  

The next chart shows Greinke's pitch selection in each count.  It's clear that the slider is his go-to pitch with two strikes. 

Greinke had a good season last year, though it was nowhere as good as his 2009 - while he maintained his impeccable control, his ERA rose by two runs and his strikeout rate fell.  His whiff rate of .191 last year was unspectacular compared to his 2009 mark of .234 (which was aided greatly by his well-documented slider).  I'd expect his numbers to be better in 2011 with his move to the NL Central.   

Harden returns to Oakland, but where's his stuff?

On Wednesday, the Athletics agreed to a deal with Rich Harden worth about $1.5 million.  Harden began his career with the A's in 2003 and remained in Oakland until a 2008 trade sent him to the Cubs.  He signed with Texas in 2010 and turned in a sub-replacement level performance that included two trips to the DL; his strikeouts plummeted and his walks climbed to a career high in 2010  Clearly, the Athletics will be hoping he can return to his 2008-2009 form, in which averaged 11 strikeouts per 9 innings over the 289 innings he pitched during those two years.  

Harden's calling-card has always been his ability to get the strikeout; his control has never been very good (career 4.16 BB/9), and over the course of his career he's been below average at getting ground balls (39.7% overall with no season over 40% since 2006).  He's also not very durable, as he hasn't thrown 150 innings or reached 30 starts in any season since 2004.  In 2010, his K/9 rate dropped from 11, where it was in 2008 and 2009, to 7.3.  The two aspects of his game that were clearly affected the most were his velocity and, maybe not unsurprisingly, his previously superb ability to miss bats.  The charts in this post will only focus on his four-seamer and changeup, which pretty much make up his entire arsenal.  I counted 5 two-seamers in 2008, just 2 in 2009, and 24 in 2010 (1025 four-seamers in 2010, which makes for a ratio of about 43:1).  I feel pretty good about the fastball classifications, but there may be a few more two-seamers from 2010 that were lumped in with the four-seamers.  The charts below shows the velocity distribution by year for Harden's four-seamer (top) and change (bottom).

If the chart for the fastball looks kind of funky, that's because it is.  Harden throws his fastball at a lot of different speeds - at the beginning of a game, he'll often be in the mid-high 80s and work his way up to the mid 90s.  You probably also noticed that Harden's fastball was significantly slower in 2010 than it had been previously.  Pitching half of his games in Texas, where the gun was slow, has something to do with that, but it doesn't constitute a difference that big.  

Now, as we know, limited velocity means nothing if you can still find a way to fool batters.  The big problem for Harden last year was that for the first time in his career, his whiff rate (.197) dropped below the league average.  As I mentioned before, without an ability to prevent walks or home runs, the strikeouts and whiffs become extremely important, and in this category Harden went from elite in 2009 to mediocre in 2010.  To illustrate his decline, here are the cumulative whiff rates for his four-seamer and changeup for the three seasons dating back to 2008.  For context, consider a league average four-seam fastball whiff rate of around 16% and a league average changeup whiff rate of around 30%.   

As you can see, both of his pitches surrendered a lot more contact in 2010 than they did in 2008 and 2009 - the changeup was ridiculous for the two years, with a whiff rate near 50%, and it plummeted to just 30% this year.  The fastball was also considerably above average and dipped significantly this year.  

We all know what Harden is capable of, and it wasn't that long ago that he was a dominant strikeout demon.  But can he regain his pre-2010 form?  The A's certainly hope so.  

Two interesting relievers with new homes

Amidst the flurry of deals that have gone down over the past week (and those that will continue throughout the Winter Meetings), I'd like to focus on two that brought versatile relief pitchers to new leagues.  On Friday, the Brewers traded Carlos Villanueva to the Blue Jays for a player to be named; on Thursday, the Angels signed former Met Hisanori Takahashi to a two year, $8 million contract.  Although Villanueva was strictly a reliever in 2010, he made 27 starts in addition to consistent relief work for the Brewers from 2006-2009.  Takahashi, an MLB rookie in 2010, filled pretty much every role over the course of the year.  He was an innings-eating middle reliever at the beginning of the season before being moved in the rotation in May.  He returned to the bullpen in August and took over as the Mets' closer in September.  I've chosen to examine both Villanueva and Takahashi in this post because they have quite a few similarities beyond the fact that they both switched leagues this week.  While it appears that both pitchers are going to be spending the majority of their time in their respective bullpens next year, they both have significant experience as starters and don't profile as "typical relievers" based on their stuff and pitch selection.  I'll look at Villanueva first, then Takahashi.  

I detected four pitches from Villanueva - a straight four-seam fastball, a straight change that gets a few different looks to it (maybe another type of changeup in there as well?), a slider, and a curveball.  (Note that there may be a few two-seam fastballs lumped with the four-seamers.)  The graphs (click to enlarge) show the spin deflection from the catcher's perspective for each pitch, and the charts show some results for each pitch.  

                      

	#	Pitch%	Swing Rate	Whiff Rate	Zone Rate	Chase Rate	Watch Rate	RV/100	xRV/100
FF	270	.292	.244	.152	.404	.118	.569	2.26	1.19
CH	249	.269	.510	.323	.446	.362	.306	-0.15	-0.23
SL	216	.234	.588	.441	.375	.467	.210	-2.92	-2.29
CU	181	.196	.331	.267	.436	.225	.532	-2.36	0.06
	921		.413	.324	.416	.288	.413	-0.44	-0.16

  

	GB Rate	FB Rate	LD Rate	PU Rate
FF	.167	.333	.433	.067
CH	.442	.209	.302	.047
SL	.306	.472	.167	.056
CU	.542	.167	.208	.083
	.361	.301	.278	.060

Villanueva's overall whiff rate of .324 ranked 7th in the majors among pitchers with at least 200 opposing plate appearances.  The slider is elite, and the change doesn't look bad, either.  There were a ton of liners, particularly off of the fastball, which is not good news.  What's particularly interesting is that, as you can by the pitch percentages, Villanueva throws the four pitches with more or less the same frequency.  Another way to show this is by looking at what Villanueva throws in each count:

vs LHB	First	Ahead	Behind	Full
FF	.341	.205	.319	.115
CH	.236	.186	.526	.308
SL	.000	.327	.009	.423
CU	.319	.268	.121	.471

      

vs RHB	First	Ahead	Behind	Full
FF	.411	.253	.276	.115
CH	.124	.142	.483	.346
SL	.174	.416	.147	.500
CU	.294	.189	.095	.038

I was surprised by the ridiculous amount of changeups in hitter counts, both to lefties and to righties.  He's definitely in the "crafty" category and could best be described as a pitcher with unusual patterns.  Let's move on to Takahashi.

	#	Pitch%	Swing Rate	Whiff Rate	Zone Rate	Chase Rate	Watch Rate	RV/100	xRV/100
FF	761	.370	.406	.136	.472	.226	.393	0.20	-0.51
FT	222	.108	.414	.174	.527	.200	.393	0.65	0.89
CH	673	.327	.578	.393	.312	.492	.233	-1.72	-1.24
SL	239	.111	.437	.280	.410	.311	.383	-1.60	-0.14
CU	134	.065	.321	.233	.366	.259	.571	0.40	0.07
FC	17	.008	.824	.214	.353	.727	.000	19.39	4.65
	2037		.465	.266	.410	.343	.360	-0.34	-0.40

	GB Rate	FB Rate	LD Rate	PU Rate
FF	.370	.303	.218	.109
FT	.356	.311	.289	.044
CH	.458	.282	.168	.092
SL	.333	.333	.182	.152
CU	.313	.375	.250	.063
FC	.167	.500	.333	.000
	.391	.306	.209	.094

The curveball is slow and he'll sometimes drop it into the low 60s.  The cutter was only used a handful of times, so disregard the rate stats for it; also, the four-seam/two-seam split is shady.  Like Villanueva, Takahashi didn't throw a whole lot of fastballs last year.  The change looks like a really good pitch, and Takahashi used it with great frequency.  Like Villanueva, Tak has a reputation as a pitcher without patterns; let's look at the count-based data (encompassing Takhashi's entire season, both as a starter and as a reliever):

vs LHB	First	Ahead	Behind	Full
FF	.290	.379	.564	.471
FT	.217	.074	.111	.147
CH	.098	.132	.009	.118
SL	.201	.358	.299	.265
CU	.304	.058	.017	.000
FC	.000	.000	.000	.000

vs RHB	First	Ahead	Behind	Full
FF	.262	.317	.342	.374
FT	.335	.074	.144	.131
CH	.190	.460	.400	.486
SL	.292	.048	.033	.009
CU	.423	.077	.031	.000
FC	.059	.022	.003	.000

So, I guess my takeaway would be that neither of these guys throw very hard, they don't appear to have very effective fastballs, and they have to rely a lot on their offspeed stuff to succeed.  Though they'll be pitching in relief in 2011, they both have starters' repertoires and probably would be able to easily slot into their team's rotation if needed.