Understanding Pitch F/X: Release Point Chart

It’s a common problem in pitching to see pitchers experience difficulty due to improper arms slot.  Either they never had a very good one or something is happening earlier in the motion to throw their arm slot off.  Poor arm slot is generally indicitive of a poor mechanical process elsewhere.  Poor mechanics lead to rushing which not only throws a pitch off, it can hurt a pitcher as well.

For today’s little tutorial we are going to look at a single game from Justin Verlanders 2011 MVP/Cy Young/Better Than You Could Hope To Be season.  Our data set will come from his May 7th start against the Toronto Blue Jays.  (hint:  It was a really good game.)

 

So first off, lets decipher what we are looking at.  In the charts key we see indicated JV’s pitch types.  On the X and Y axis we see distances from the catchers perspective (in feet).  The box in the center of the graph denotes a generic strikezone.  Strike zones by definition can vary from batter to batter so the one shown is more or less just an average and should only be used to create a perspective, not a truth.

There is something about how this data is gathered that is very important to note.  It is not gathered at actual release.  It is gathered at the 50ft. from home plate.  This is important to note because looking at actual release points (aka video) can tell us so much more than any pitch f/x chart ever could.  However, given the absence of solid video sometimes the release point chart serves as an integral part of  my work.

What we want to see when we look at these charts is consistency.  In determining consistency, you must first determine someplace to work around.  There is generally going to be at least one good sized clump on the chart.  The center of that clump is going to be where you determine the consistency of the release from.  Deviation from that main point is the determing factor here.  The more pitches that deviate and how far they deviate from the main point, the worse the release is.

Looking at JV, I’d pick (-1.75, 6.25) as his general release point.  All his pitches are being picked up within approximately a foot of that point.  Given the nature of the data collection, magnus effect (pitch movement), and any other immeasurable factor I’d say that this a very good release.  It’s very deceptive to come around with so many different pitches, but have them all fly at you from the exact same spot.

For general purposes I’d have to say that falling within that 1ft deviation is very good.  The rating gets worse the larger that number gets and by the time you hit 2ft, you should probably be hitting up the minor leagues for some mechanics work in my opinion.

So when I see a bad release point, I know that there is a larger mechanical issue at stake here.  It’s at this point that I hope to find some video so I can play amateur pitching coach and tell the Big Leaguer what exactly he’s doing wrong.  All from the comfort of my couch of course.

Understanding Pitch F/X: Pitch Virtualization Charts

It is normal when evaluating a pitchers mechanics to take a video, slow it down, freeze it, and dissect the nuances of the motion.  Personally, I’ll look for positioning of the limbs, rotation of the hip, stride length, arm movement, etc.  It’s basically the most effective way to truly evaluate a pitcher.  Unfortunately, I don’t always have video to use.  I either don’t have access to it or I just have poor video to work with.  However, there is still a way for me to make strong guesses into how a pitcher is acting.

Pitch F/X are a very useful tool and are essentially the best tool for seeing just how well a pitcher is actually pitching.  They describe everything from how the ball locates and how fast it is thrown to how the ball spins and moves through the air.  These are things that are sometimes difficult to pick up in a video and are things that anybody can use to reinforce an opinion in attempt to establish new fact.

However, they can be tricky because the charts do not work how our brain wants them too.  There are things in the charts that must be understood before reading them and hopefully reinforcing ourselves.

So to help teach myself as well as my faithful readers, we’ll follow the best of the 2011 MLB season:  Justin Verlander.  To begin the tutorials we are going to look at the Pitch Virtualization charts at Texas Leaguers.

 

 

First to describe the basic information on the charts.

The box on the right side are the pitch types.  It shows JV featuring five different pitches.

• FF:  Four Seam Fastball.  This is a standard straight moving fastball that generally features greater velocity.
• FT:  Two Seam Fastball.  A very popular pitch because it mixes a fastball’s velocity with some movement.
• SL:  Slider.  A pitch that breaks laterally and down with a moderate velocity.
• CU:  Curveball.  A pitch that dives downward as it approaches the plate.
• CH:  Changeup.  Essentially a slow fastball.

I know many of my readers didn’t need me to tell you how these pitches act, but it’ll prove a useful resource when we actually start looking at the data.  In need of less explanation are the X and Y axis on the chart.  The X describes horizontal distance in feet (from release to glove) and the Y denotes height in feet (from release to glove).

So based off all that information everybody here has a bit of a grasp as to what those charts mean and what they describe.  In a nut shell they show how a pitch moves from release to the catchers glove and when the stages of movement happen over a distance.  One important piece of information to remember though are the charts titles.  Each contains a very key word:  Virtualization.  Virtualizations aren’t always representative of what actually happens.  In these charts they tend to represent what should happen.

So where does the virtualization come from?  The answer is data.  At the top of  Texas Leaguers Pitch F/X pages you can find the averages for their collected data.  Vertical and horizontal positioning, spin rates, spin angles, velocities, and pitch usage are all available averages for you to look at.  From this information a virtualization can be created showing us how the pitch should act.

Now maybe you’re asking a new question:  What good does it do us if it isn’t showing exactly what’s happening?  Simple answer is comparison.  What should happen is useful data when compared to what is actually happening.  But, I’m not gonna write a ten page paper at 11:30 at night (save that for my college professors).  You’ll just have to learn more in one of the upcoming segments where I will further describe some of the information that those data sets infer.

What’s happened to Jose? (pt. 1)

Jose Valverde started off his season for the Detroit Tigers putting up great numbers.  He pitched 34 innings up to June 28th (approximately the all star break) and put up a .53 ERA with 32 strike outs.  Not bad at all.  Since than he has a 6.58 ERA in 26 innings pitched, but has still managed 29 strikeouts in that time.  But still it’s been ugly at times.  The question is what’s wrong?  Since I like the pitching aspect of this game, I guess I’ll be the one to explore.

First thing is to check out his release in the first half and compare it to the second half.

First Half:

Second Half:

Now I see that his release point is a bit higher in the second half of the season.  This is the first thing I shall note.  Something more important than release point is the actual location of the pitch.  No one cares about release if the result is good.

First Half Locations:

Second Half Locations:

And this is why we look at the location chart, because at first glance theres not much difference.  But that is why you must than pay attention to the pitch types as well.  Even though the locations are technically in the strike zone, I see sliders hanging and cutters not cutting.  Something else that is noteworthy is that the difference here is not that great.  But what makes me believe that once again release is his problem is simple percentages.  First half of the season Valverde got his outs about 30% of the time via ground out.  Second half:  16%.  Big difference which helps reinforce my point.

Next I want to find out what’s going on here behind the mechanics, but I’ll save that for Pt. 2.

Comparing Perfect Games: Who’s More Perfect?

What a year in pitching, as of right now it’s June 17th and there have already been three perfect games.  Officially only two but we all know that there have been three.

So the question begs to be asked, who pitched the most perfect game?

Determining what embodies a really good pitching performance is hard because it’s a matter of perspective.  Obviously going as many innings as possible while giving up a limited amount of runs is the ultimate goal, but since we’re dealing with three perfect games here, our criteria has to be a bit more critical.

So we set up our criteria for a perfect game and our grading system.  First off I will be giving each player (Dallas Braden, Roy Halladay, and Armando Galarraga) points based upon how they do.  First place in each category gets 5 points, second gets 3, and third gets one measly point.

The categories:  I will be judging off of grouder/fly ball ratio, ball/strike ratio, overall pitch count, strike outs, and release point.

So now I analyze.

Ground Out/Fly Out Ratio

Armando wins this one hands down.  Of the 28 outs he recorded in his perfect game, 14 were ground outs.  By my own personal judgment, grounders are more efficient outs to get.  Second place goes to Halladay who got 29% of his batters faced out via the ground ball.  So third place?  That goes to Mr. Braden, with a paltry 26% of batter getting out via the ground out.

Ball/Strike Ratio

Armando hit the strike zone 76% of the time.  Roy Halladay hit the K-Zone 62% of the time.  Braden hit the strike zone 70% of the time.

Armando, once again you win.

Pitch Count

Now Armando won this category hands down.  88 pitches through 28 total batters faced.  Halladay?  115 pitches.  Braden?  109.  It’s not even close.  Apparently when you attack the strike zone and get efficient outs, your pitch count stays wayyyyy down.  Now that’s impressive.

Strikeouts

This is more of a measure of “nastiness.”  The strikeout is probably the most embarrassing way to get called out during a game.  Great pitchers make batters look silly.  Halladay wins this category.  He struck out 40% of batters.  That’s an insane number.  Coming into second place is Dallas Braden with 22% of batters struck out.  Galarraga brings up the rear with 10% struck out.

Sorry Mando, you can’t win ‘em all.

Release Point

Release point is an indicator of consistency in mechanics.  It also doubles as a “nastiness” indicator because it indicates deceptiveness in the mechanics.  While Mando and Halladay were pretty close, I have to give this category to Halladay because there are no real extreme outliers.  Dallas, once again you get third.

Tallying the Score

Armando Galarraga:  19 points

Roy Halladay:  15 points

Dallas Braden:  11 points

So Armando, I award you the most perfect of perfect games.  Seeing as I’m a poor college student, the notoriety of winning a competition on my blog is all you get.

All information used in this post was gathered from Fangraphs.com and/or TexasLeagures.com

Keeping Tabs on Rick…

I did a little story after Rick Porcello’s last start discussing the mechanical issues that have been causing his issues on the mound.  The obvious answer was his arm slot was highly inconsistent.  This lead to less sink on his pitches.  Sinkerballers like sink on their pitches.  Trust me.

So I figured, lets see if there been any changes?  Looks like there have been some positive ones.

To save me some time I’m just going to link you guys to the PitchF/X charts, courtesy of TexasLeaguers.com.

My main focus is the release point chart.  Compared to the 2 starts I previously analyzed, this is a vast improvement.  It’s still not consistent enough to be truly effective but improvement is what we’re looking for.  Ideally you like to see those dots line up as much as possible.  While the up down position is probably most important, the horizontal position is more telling of how the batter is seeing the ball out of the pitchers hand.  The wide spacing means that it’s a bit easier to pick up on Ricks pitches still.

Now I want to look at the Spin Movement plus Gravity chart and the Pitch Location chart.

In the first chart we see location based upon spin movement of the ball plus the added effect of gravity.  I love this chart to be honest.  What we see in the chart is that Rick’s 2-seam fastball is staying down and inside on right handed batters as well as his change-up.  That’s a hard pitch to hit.  When we look at the actual pitch locations in the next chart, we are confirmed in seeing that a significant proportion of Ricks 2-seamers and change-ups were hitting that low inside corner on right handed batters.

Like I said, that’s a hard pitch to hit.  But if your release is giving a tip a split second before the bat reaches the plate, Major League batters know how to adjust.  But Rick’s improving, and based on this data I’m encouraged that more improvements will come.  I’m not thinking AAA any more for Rick, but maybe shutting him down for a start so that he can focus on his mechanics in bullpen sessions might do him some good.

Analyzing PitchF/X: Rick Porcello

It’s been awhile since I’ve made a post here, but there has been good reason for that and it’s a reason I won’t delve into too much.  But I thought analyzing through PitchF/X might be a good way for me to break back in.

Now up til now my primary medium for analysis has been video and still shots.  It’s got it’s advantages such as seeing the whole motion, which gives me the opportunity to spot what’s wrong with a pitcher.  Often though, the problem with a pitcher comes down too a very simple thing, arm slot.  I started a series on arm slot a month ago (I’ll finish it eventually) and basically arm slot accounts for a lot.  Things like location, velocity, and movement can be affected, but location and movement are the big two.

But arm slot is pretty hard to pick up through video.  First you need good video that you can analyze frame by frame.  Then you need to watch every single pitch and look for even the slightest of changes in the slot that the pitchers arm is coming out of.  So saying off of video alone that his arm slot is inconsistent is really more of a guessing game.  But there is a system in place that makes this much much easier.  It’s called PitchF/X and basically it takes information from things like the radar gun and translates that into useful information such as release point, movement, and velocity.

So lets begin the analysis.

It’s no secret that Rick Porcello of the Detroit Tigers has been struggling thus far.  Reasons probably being mechanical because even without really analyzing you can see that the movement just isn’t there.  So first we look at his arm slot/release point.

Now the graph above covers a sample from 6-1-2010 through 6-9-2010 which captures a couple of starts.    Now each dot shows us each pitch and the different colors and shapes tell us what kind of pitch.  What we’re seeing here is that there is essentially a large discrepancy at Porcello’s release point.  The lowest pitch is being picked up at around 5ft off the ground while the highest is being picked up around 6.5ft off the ground.  That’s a big difference.  However, it’s important to remember that these charts are not made off data coming right at the moment of release.  The pitch is picked up a few feet afterward.  These charts merely serve as strong indicators, but a discrepancy this large tells us that there is definitely a problem here.  So we see the mechanical flaw, but what is the issue that it causes?

Now this is a unique chart because it takes gravity into consideration by virtue of an equation figured out by Trip Somers at TexasLeaguers.com (great place for pitchf/x charts btw).  By taking into account gravity, we see a truer representation of the movement that the pitches have and how they’ll end up when they cross home plate.  Looking at the chart, figuring out how many inches above the plate Porcello’s fastball’s (FF and FT) are ending up, I can see the problem.  The fastballs are staying in the happy zone for alot of hitters.  For a sinker baller, Rick isn’t really getting much sink.

Now normally I’d jump into finding out what might be causing the slot issues but after tentatively watching Rick pitch, I can say that he should think about his arm slot first.  If he makes that his priority any other issue should be resolved as a result of trying to fix the main problem.  It’s less fun analysis wise, but hey, I don’t feel like finding video right now.