Buy Low/Sell High July 5th (Bedard/Atkins)

Buy Low: Erik Bedard

Key Stats:

Pitcher:	BABIP	K/BB	PTO%	qERA	ERA Dif	P/IP	FIP
Hitter:	BABIP	P/PA	BB/SO	GB/FB	HR/FB
2008	0.275	1.950	28.2%	4.56	-0.893	16.93	4.28
2007	0.287	3.88	33.54%	2.68	0.477	16.19	3.09
2006	0.314	2.48	28.23%	3.96	-0.197	16.80	3.59

Erik has probably disappointed many fantasy owners who were expecting a CY-Young season. Fortunately for you (if you don’t own him) he’s likely to be able to be bought for less than usual. The only thing that I find slightly concerning is his loss of control. He might have started to turn the corner and if so, he could be really help your team.

Bedard, Erik SP SEA for Bay, Jason LF PIT

Bedard, Erik SP SEA for Glaus, Troy 3B STL

Bedard, Erik SP SEA for Granderson, Curtis CF DET

Sell High: Garrett Atkins

Key Stats: 

Pitcher:	BABIP	K/BB	PTO%	qERA	ERA Dif	P/IP	FIP
Hitter:	BABIP	P/PA	BB/SO	GB/FB	HR/FB
2008	0.315	3.54	0.36	1.129	12.90%
2007	0.288	3.90	0.70	0.971	14.62%
2006	0.309	3.82	1.04	1.037	15.34%

Garrett was a top prospect a couple of years ago but he has since run into down years and lofty expectations. This year he is doing fairly well with 50 RBIs and 12 HR but still not in the upper tier of 3B. He’s a candidate to get traded to most likely an underperforming team and also I don’t like some of the things I see above. He’s much less selective in his plate discipline and he’s also hitting less HR/FB yet more GB at the same time. I think this is a little lucky to have 12 HR right now so I’m going to recommend sell high, but only for the right price.

Atkins, Garrett 3B COL and Tulowitzki, Troy SS COL for Cabrera, Miguel 3B DET

Atkins, Garrett 3B COL and Harden, Rich SP OAK for Howard, Ryan 1B PHI

Atkins, Garrett 3B COL and Jackson, Conor 1B ARI for Fielder, Prince 1B MIL

Jeff Francis Breakdown – What went wrong?

First my methodology. I have built my own Pitch FX database with a lot of help from Mike Fast, Dan Turkenkopf, and one of my friends Steve Comeau. After an email question several weeks ago to Will Carroll, I focused on only home starts in order to eliminate as much noise and inaccuracy from different stadiums. Finally, I decided to focus on specific aspects; release point, velocity, pitch location/command, movement, and the resulting event. I use MySQL for my database and SPSS 16.0 for all graphs and tables.Movement

I looked at Jeff Francis' problems in early May and you can read both posts here and here. After reading yesterday's excellent Under the Knife (subscription required but well worth it) article by Will Carroll at Baseball Prospectus, I decided to revisit my analysis.

Release Point

I always begin my analysis of pitchers with a look at their release point. Release points can tell you a lot if you look carefully and analyze it. It can help tell you if there is an injury or if the pitcher may be tipping his pitches. It can also show you if the pitcher is changing his mechanics by looking at his arm slot. The only slight disadvantage to this system is that it doesn't measure the true actual release point, it measures a point in space 50 feet from the front of home plate. This has a slight effect on the "release point" of curveballs as they tend to show up a couple inches higher than all of the other pitches. Looking at Francis' this rings true.

We see with this that the curveball does indeed come out higher at the measuring point. But one of the problems I see is that he also releases his changeup a little higher than his fastball. It's as if each pitch has a separate release point, as much as a half foot in some cases. Now the question is which of the release points is the "wrong" one. My assumption is that it is the changeup as he changes his motion to let the ball slide off his fingers leading to more pronation. Inevitably, this will lead to some internal rotation at the shoulder and could cause some impingement in the shoulder. This is where some soft tissue, usually the rotator cuff tendons, are pinched between the shoulder and the humerus causing inflammation or minor tearing. For an excellent visual explanation, I recommend this link.

The other thing I notice is that he is wildly inconsistent in his release point across his starts. This could be due to an injury or a break in the kinetic chain needed to reliably replicate his motion. The interesting thing is that he wavers back and forth horizontally and vertically. Horizontally it ranges by close to 2 feet and vertically it ranges across 1 foot. This is a huge amount in terms of pitching mechanics and is concerning to me. The last 3 starts it appears that he begins to waver prominently to the right, "dropping" his arm slot each time. If there was inflammation or impingement, pain usually occurs during overhead activities which is just a wee bit common during pitching. The pitcher will unconsciously limit how high he will bring his arm up in order to save himself from pain. Francis himself said that he feels the pain in the shoulder during some warm ups later in games. If this is true, we would expect to see a lower arm slot/release point either from fatigue or from the pain the inflammation causes. This again holds true with Francis:

This chart is slightly different than most are used to with release points but I feel it illustrates my point the best. The lines are coming out from the center of all the release points in the inning. You can better estimate where the center is even from background innings by looking at the colored lines and following them to their center. What I see is that his last three innings do in fact show that the release point drops compared to the previous 4.

Pitch Location

Looking pitch location by innings is a little harder due to the amount of overlap however there is one major thing that I noticed. The number of pitches that were outside of the zone in innings 6 and 7 are extremely small while other innings show a great amount of them outside of the zone. This is something that I wasn't quite sure what to make of it except to guess that he has less control of exactly where the pitch will go. This does occur with muscular fatigue around the shoulder and forearm as the pitcher is less able to reliably control his pitch. Looking at the other chart on the right, you see that he threw as many strikes as balls for his fastballs as does his curveball. He has slightly better command of his changeup but not greatly so.

Movement

This is one of my favorite graphs when evaluating a pitchers' effectiveness to his pitches. His movement of the changeup and the fastball are earily similar. When this much overlap of the two pitches it means that the two pitches are moving exactly the same just at slightly different speeds, only 6MPH according to the top chart. The other thing that I notice is that he doesn't have great command of his pitches either with the exception of his curveball. With the fairly small diameter of group of pitches, it shows that when he throws his curveball it is consistent in its movement and he knows where it will go. Compare that to the diameter of the changeup and the fastball and you can see that he really doesn't know how well those pitches will move. This can be a result of extremely small differences in release point, the level of muscular fatigue, or the unconscious protection of the body from pain.

Velocity and numbers

pitch	Start Speed	End Speed	Movement In x (in.)	Movement In z (in.)	# Thrown	% Thrown
CH	79.67	73.23	8.60	3.35	203	26.7%
CU	70.78	65.15	-0.92	-4.01	98	12.9%
FA	87.42	79.86	10.94	5.81	449	59.1%
SL	85.1	78.5	4.6	3.7	8	1.1%
Total	83.10	76.11	8.70	3.87	760	100%

Evaluating the average speed during his starts is another method to evaluate the health of a pitcher, most importantly fastball velocity. By following the top line, there is a slight bump in the 4^th and 5^th starts but then his fastball velocity begins to tail off again. This coincides with his dropping of his arm slot noted previously, leading me to believe that the shoulder was beginning to bother him during those starts. Many of the symptoms associated with shoulder inflammation and impingement result in what we see above.

Summary

My analysis, while certainly not the most scientific, shows evidence of the problem that Francis describes and also what the medical personnel are telling us. Eventually I would love to be able to use this system to help Major League teams analyze their players health but I know for a fact that we are years away from being able to use the PFX system as a predictor or diagnostic test for injuries. One thing that I do know though is that we will only learn about the future by focusing on what we know and don't know about the past.

Buy Low/Sell High July 4th (Kobayashi/Cabrera)

Buy Low: Masahide Kobayashi

Key Stats:

Pitcher:	BABIP	K/BB	PTO%	qERA	ERA Dif	P/IP	FIP
Hitter:	BABIP	P/PA	BB/SO	GB/FB	HR/FB
2008	0.268	2.360	27.1%	4.28	-0.995	15.76	4.40
2007	#N/A	#N/A	#N/A	#N/A	#N/A	#N/A	#N/A
2006	#N/A	#N/A	#N/A	#N/A	#N/A	#N/A	#N/A

His numbers aren’t exactly great but with Borowski apparently losing the closers role, save opps are save opps at this stage of the game. He might have to split with Betancourt but remember that he struggled in that role earlier on in the year. He’ll probably be available in most leagues right now so go ahead and grab him now or put in a waiver claim now.

Kobayashi, Masahide RP CLE for Millar, Kevin 1B BAL

Kobayashi, Masahide RP CLE for Thompson, Daryl SP CIN

Kobayashi, Masahide RP CLE for Owens, Jerry CF CHW

Sell High: Daniel Cabrera

Key Stats: 

Pitcher:	BABIP	K/BB	PTO%	qERA	ERA Dif	P/IP	FIP
Hitter:	BABIP	P/PA	BB/SO	GB/FB	HR/FB
2008	0.273	1.400	26.4%	5.05	-0.708	15.61	4.95
2007	0.308	1.54	26.52%	4.83	0.720	17.44	4.77
2006	0.32	1.51	28.27%	4.90	-0.158	18.11	4.17

After his amazing run earlier in the season, he’s had a rough June, but I think June is still more indicative of his talent than the greatness. How many times does someone who’s been in the league for so many years all of a sudden develop control. Honestly very rarely. Everything above points to trouble in the future, so I would sell high on him with the pitch being that he strikes out a lot of guys.

Cabrera, Daniel SP BAL for Sheffield, Gary DH DET

Cabrera, Daniel SP BAL and Johjima, Kenji C SEA for Sherrill, George RP BAL

Cabrera, Daniel SP BAL for Drew, J.D. RF BOS

Pitch F/X (PFX) explanation

Pitch F/X was created by Sportvision, Inc. and uses cameras to track the pitch from about 50 feet away from home plate all the way to the very front of the plate. The cameras are usually set up on the first base side and above home plate to get a 3D picture of the pitch. As the cameras are tracking the pitch, a computer algorithm actually does the measuring of a ton of different things with the help of a human operator/technician. This operator sets the strike zone for each hitter in the PFX system and helps the computer "track" the pitches from velocity to release point to a pitch break to the pitch location.

Rather than copying someone elses description of each field that PFX uses, I will have to refer you to Mike Fast's excellent glossary of terms. The main ones I focus on are those involved with release point, pitch location, movement, velocity, and pitch type.

Hopefully this will help give a very basic understanding to how I get these charts and definitions in my articles.