Gas House Graphs

A Little on Wainwright

It’s no mystery that Waino has been struggling a bit in the second half.  A quick glance at his splits page from Fangraphs shows just how stark the difference has been.  Bernie outlined some possible causes in a column earlier in the week, highlighting among other things a more vulnerable sinker (go read the entire column for the rest – it’s a worthy read).  I wanted to walk through Bernie’s bullets on the sinker and fastball in a little more detail with some visuals.

First, let’s look at location starting with “hard stuff” before July.  All charts are from the catcher’s perspective.

and then since July


The general trend is the same, pound low and away to right handed hitters; however the numbers are slightly lower since July which is a slight indication that location may be off just a bit in recent starts. What about the results of those pitches; first looking at how often hitters are swinging



It appears that in the more recent starts (second chart) that the hitters are not chasing that pitch low and away as frequently.  The impact of that is heightened when you look at the next set of charts – whiffs per swing



As you can see from the first chart, Waino was getting a high rate of swings and misses on the hard stuff off the plate low and away.  The change in the second chart is less whiff rate, and more indicative of the lack of swings that we already saw (although the whiff rate drop in the “knee high” but outside box is interesting).

Not sure that any of this explains the performance drop (as Bernie noted there’s some BABIP wrapped up in all of this as well), but my 30 second summary is the location of the sinker and four seamer is slightly off and when it is well located hitters are doing a better job of laying off.  In turn, Waino’s getting fewer swings and misses than previously in the year.



In case you haven’t noticed, Allen Craig’s ground ball rate has jumped rather drastically (55% so far this year, compared to 45% last year according to Fangraphs).  While we’re still operating in relative small sample size land, it’s still a rather alarming jump, so much so that I wanted to get some historical context.  With that goal I grabbed all of the seasons of greater than 250 plate appearances since 2007( rather arbitrary cut-offs, but good enough for illustration purposes), and looked for seasons with at least a 10% increase in ground ball rate.  I found only 26 such instances in the sample, of which 4 of them are this season.  The following table has the full list

Name Season1 Season2 GB1 GB2 wOBA1 wOBA2 PA1 PA2 s2-s1 wOBA
Allen Craig 2013 2014 45 55 0.363 0.305 563 307 -0.058
Andruw Jones 2009 2010 34.4 44.4 0.337 0.362 331 328 0.025
Carlos Guillen 2009 2010 36.3 49.8 0.333 0.327 322 275 -0.006
Carlos Pena 2009 2010 29 44.9 0.377 0.325 570 582 -0.052
Carlos Pena 2012 2013 37.2 49.7 0.309 0.298 600 328 -0.011
Chipper Jones 2010 2011 38.1 49.3 0.356 0.348 381 512 -0.008
Cody Ross 2009 2010 33 45.6 0.342 0.322 604 569 -0.02
David Eckstein 2007 2008 40.7 51.4 0.33 0.314 484 376 -0.016
David Murphy 2010 2011 44.2 54.2 0.354 0.32 471 440 -0.034
Domonic Brown 2013 2014 42.4 54.1 0.351 0.263 540 276 -0.088
Eric Hinske 2010 2011 33 45.5 0.342 0.315 320 264 -0.027
Felix Pie 2009 2010 40.8 51.1 0.331 0.314 281 308 -0.017
Hanley Ramirez 2009 2010 38.6 51 0.407 0.369 652 619 -0.038
Howie Kendrick 2013 2014 51.3 66 0.336 0.321 513 314 -0.015
Ian Kinsler 2009 2010 30.1 40.1 0.351 0.355 640 460 0.004
John Mayberry 2011 2012 41.8 51.8 0.368 0.303 296 479 -0.065
Julio Lugo 2007 2008 46.4 59.5 0.286 0.315 630 307 0.029
Kevin Kouzmanoff 2010 2011 42.9 53.9 0.296 0.287 586 257 -0.009
Luis Castillo 2009 2010 58.6 70.4 0.334 0.284 580 299 -0.05
Manny Ramirez 2009 2010 32.7 43 0.398 0.383 431 320 -0.015
Marlon Byrd 2009 2010 40.5 52.2 0.346 0.343 599 630 -0.003
Michael Morse 2011 2012 44 55.3 0.39 0.34 575 430 -0.05
Placido Polanco 2011 2012 42.4 55.6 0.303 0.279 523 328 -0.024
Reed Johnson 2011 2012 43.4 56.7 0.355 0.322 266 288 -0.033
Robinson Cano 2013 2014 44.3 54.8 0.384 0.366 681 302 -0.018
Ryan Howard 2007 2008 31.5 41.5 0.397 0.368 648 700 -0.029

As you can see only 3 players in the sample actually had a higher wOBA in the second season than the first.  With that in mind let’s be the pessimist for a moment and have his current ground ball rate hypothetically continue through the season.  How did these players that had drastic increases in ground ball rate hit in season 3?  the following table summarizes the results for those players that had at least 250 PAs in year 3.

Name GB1 GB2 GB3 wOBA1 wOBA2 wOBA3 PA1 PA2 PA3 S2-S1 WOBA s3-s2WOBA s3-s1 wOBA
Ryan Howard 31.5 41.5 36.2 0.397 0.368 0.392 648 700 703 -0.029 0.024 -0.005
David Murphy 44.2 54.2 43.4 0.354 0.32 0.369 471 440 521 -0.034 0.049 0.015
Ian Kinsler 30.1 40.1 35.3 0.351 0.355 0.364 640 460 723 0.004 0.009 0.013
Chipper Jones 38.1 49.3 46 0.356 0.348 0.36 381 512 448 -0.008 0.012 0.004
Carlos Pena 29 44.9 37.3 0.377 0.325 0.356 570 582 606 -0.052 0.031 -0.021
Julio Lugo 46.4 59.5 40.5 0.286 0.315 0.335 630 307 293 0.029 0.02 0.049
Cody Ross 33 45.6 34 0.342 0.322 0.321 604 569 461 -0.02 -0.001 -0.021
Hanley Ramirez 38.6 51 50.9 0.407 0.369 0.317 652 619 385 -0.038 -0.052 -0.09
Marlon Byrd 40.5 52.2 50.1 0.346 0.343 0.317 599 630 482 -0.003 -0.026 -0.029
John Mayberry 41.8 51.8 42.8 0.368 0.303 0.298 296 479 384 -0.065 -0.005 -0.07
David Eckstein 40.7 51.4 46.1 0.33 0.314 0.296 484 376 568 -0.016 -0.018 -0.034
Michael Morse 44 55.3 44.7 0.39 0.34 0.286 575 430 337 -0.05 -0.054 -0.104
Placido Polanco 42.4 55.6 48.3 0.303 0.279 0.279 523 328 416 -0.024 0 -0.024

This news is at least somewhat promising.  It’s not unheard of for a player to have a season’s jump in ground ball rate and “bounce back” the following year.  In fact most of the changes from year 1 to year 3 would fall under “standard aging”.

1 comment

During the broadcast on Fox Sports Midwest the other night (I think it was Friday night, but I reserve the right to be wrong) the crew showed a graphic about the Cardinals great gains in Defensive Runs Saved (DRS) this year compared to last.  Currently the Cards have a healthy lead (>10 runs) on the second place Rockies in that category according to Fangraphs.  Ultimate Zone Rating (UZR) a similarly designed metric has them at seventh in MLB.  Independent of your metric of choice, both show substantial improvements over last year’s squad which was ranked in the bottom 10 of MLB of both DRS and UZR.  Clearly that improvement is great, but what does it mean in the world of small samples (as this year’s numbers are)?  What type of defensive performance should we expect going forward?  With that question in mind a brought out some defensive projection generation code that I’ve  been putting together over the last few months.

[click to continue…]


Thursday Shorts

In honor of Bernie’s Bytes (which are a must read for those that don’t) I thought I’d start throwing out Steve’s Shorts. The idea is they’ll be reasonably quick takes on things with a sabermetric bent.  There won’t be a ton of research that goes into these, but will likely be a graph or two based on readily available data that I think highlights a key point.  Without further discussion let’s just jump right in

  • It’s nice to have this defense on the field



but it’s hard when it comes with


I was downright shocked on the down the middle square. Just cleaning that up would make a ton of difference. I’m wondering how much of this


might make a difference. It appears that anything slow is being pulled more than it had early in the season. Just an observation, haven’t checked performance on those particular balls in play.

  • It’ll be nice to see what Carlos Martinez can do with his start.  Steamer has him projected at a ~3.60 FIP as a reliever right now.  Avert your eyes from the Jake Westbrook projection if you click through the link.
  • Projections aside, I’ll go a little old school on you and say that I respected Jake for taking his beating and saving innings in the bullpen last night.  I haven’t thought too long or hard about whether it was the right move to bring him in, but once he was in, leaving him in (barring risk of injury) was the right move.


Looking at Approaches

Clearly the offense has been struggling over recent weeks (at least as I type this they have put up two runs in the first against the Pirates, so we have that going for us).  With that in mind I picked a couple of players, Allen Craig and Carlos Beltran, to look at in some detail to see if there was anything we could learn.  Before I do that though I wanted to offer a positive graph.  As you may have read we had some contests going on last week, and mine was to offer up your best graph.  Our winner was Ben Chambers who provided the following graph for Allen Craig

Craig zone

with the following commentary:

It’s the hot/cold zones for Allen Craig from for 2013. It’s amazing to me because I there isn’t a single place inside the strike zone that is below .308. That’s pretty amazing. Most of the time that he gets out, it’s outside of the strike zone. If someone wants to go up there and throw strikes, but he can hit pretty much anything inside the zone, what is the pitcher going to do?

Clearly that guys is still in there, so can we find out if he’s been doing anything differently over the last month or so? First to set the baseline I’ll give you swing rate by zone through June


Then just the month of July


While there isn’t a whole lot that jumps out at me, the biggest thing I see is the decrease in swings on belt high pitches.  I’d especially point out the middle in pitch that he has only swung at 50% of the time in July compared to 70% or so on the season.  I’d also highlight how much he has gone out of the zone low and in (70%) compared to 48% in other months.  I’d also note that he has a 0.167 SLG on balls in play in that low and in spot .


Moving on to Beltran.  I’ve isolated it to only at bats versus RHP.  First the pre July numbers


Then the month of July


Beltran has taken many more swings on pitches off the plate away in July than in the previous months. He’s had some success out there on the belt high pitches (not a ton, but some), but the others he’s not had the kind of success to justify the swing rate out there.

All in all it seems like there are some approach differences that could be leading to the performance drop off. Hopefully things will turn around shortly.


Your Turn

Hi folks, unfortunately work has gotten in the way of content generation for me.  That said you may have seen some contests coming across the other blogs here at the Conclave, and I’ve got one for you too.


What: You have the chance to win a free copy of St. Louis Cardinals Baseball’s Greatest Games: 2011 World Series, Game 6 DVD/Bluray Combo

How: Send me (steve.sommer05 at gmail dot com) your best Cardinals related graph/table.  It can be informational (like those I’ve tried to post here) or humorous if you want to go that route.  I’ll select the best one (where best is judged based on both content and look) and the kind folks at A&E will send you your copy of the DVD.  Also, as an added bonus, I’ll post your graph here at the Conclave, so you’ll get some exposure in the blogoshpere if that’s important to you.

When:  Send me your entries by July 30th

Other info: Aside from MLB Bloopers and Prime 9: MLB Heroics, available programming includes The Best of the Home Run Derby and “Prime 9: All-Star Moments;” Official World Series Films dating back to 1947, including the 1969 and 1986 films; the first season of “This Week In Baseball,” which originally aired in 1977; a documentary offering a fresh perspective on Jackie Robinson’s life and career; recent productions including a comprehensive film chronicling every era of World Series play and documentaries created to celebrate notable anniversaries for the Mets, Astros and Red Sox; bloopers titles highlighting the funniest MLB moments; and many other titles.   Any of these films can now be downloaded from the iTunes store ( Prices range from $1.99 for individual episodes of “Prime 9” and “This Week in Baseball” to $19.99 for the Official 2012 World Series Film in HD.



Second Base Musings

Yesterday over at StlToday and in the PD Joe Strauss wrote about the upgrade Matt Carpenter provides compared to Cards second basemen of years past.  Being as how this is the Gas House Graphs section of the Conclave I figured what better way to depict that upgrade than with a graph.  I grabbed the production (offense and defense) the Cards have received from the second base position over the last ten seasons and put them in a graph


Clearly we can see that there’s no competition on the offensive side as the second base position has produced nearly 5 times the wRAA of the second best season in just over half a season (and that includes numbers from Descalso which likely brings the total down).  And while there have been better defensive seasons, it’s not as if Carp has been problematic there (some of the negative in the chart is Descalso as well).  While I wouldn’t expect Carp to necessarily be a plus defender at second over the long haul, even something that’s slightly below average is fine when you can get the kind of offense he’s been giving.

All in all nothing said in Strauss’ article nor shown in the graph above is news to anyone that has been watching the Cardinals for any amount of time.  That said, the order of magnitude of the difference is rather staggering.  I’m sure as the season continues I’ll have more to say/show on M. Carp (more of the how rather than the what), but for now, merely take in the graph above.

1 comment

Happy 4th of July!  I hope everyone is enjoying it with friends and family.


In lieu of an actual post today I’m going to simply present a graph without much comment courtesy of


The out of zone swing percentage is a little worrisome as it also coincides with a month in which the Koz wOBA’d .220 (which for those not following at home is REALLY bad).

1 comment

I’ve spent a decent amount of time over the past few weeks/months thinking of ways to better visualize pitch f/x data especially in the context of an at bat.  It just so happens that in my day job I spend an equal amount of time analyzing transportation networks where a common data visualization is via network graphs.  I thought there may be some cross applicability there, so I started to create some graphs just to see what they might look like and what they might tell us.  I decided to start with Shelby Miller for a couple or reasons 1) His arsenal is reasonably limited 2) It feels like he has lengthy at bats against him frequently 3) He’s really good!

So with that in mind I present the following Shelby Miller network graph (I’ll explain in detail after the graph itself)


You’re probably going to have to click on the graph to enlarge it and see what’s actually going on.  Let me spend some time orienting you to the graph.  Each node (the blue-ish boxes) represents either a count or an at bat’s outcome where for this particular graph I binned the outputs into BB, K, Hit (H), and Out (O).  The color of each line represents the pitch type (legend in the upper left) and the width is in proportion to the number of pitches.  Starting at the top we have the first pitch of an at bat and can walk through the counts and look at pitch selection and how we progress through various counts, looking for thick lines along the way to see trends.


Now is probably a decent enough time to talk about the weaknesses of the above.  The primary weakness is that this ignores sequencing, so a 1-2 count is a 1-2 count no matter how you get there.  Clearly in practice this is a false assumption (although I’ll try and deal with that in a separate post later), but it you want to capture the whole at bat we need to simplify the combinations somewhat.


With that in mind we can probably still glean some things from the visual reasonably quickly.  1) Shelby pounds away with the fastball really no matter the count. 2) He doesn’t appear to waste a ton of pitches (look at the width of lines between 2 strike counts and Ks and compare to the width of lines between 2 strike counts and the next count if he would have thrown a ball).  3) He gives up a decent amount of 2 strike foul balls (the loops that return to the same count).

Now for basically the same graph, only with the type of hit broken out.

Shelby_Simple_Graph_ResultsHRs come from the fastball primarily, but come from a wide variety of counts (i.e. he wasn’t necessarily behind in the count serving it up per se).


All in all, I think there’s more to get at with these types of graphs (trying to look at sequencing, looking at a couple of pitchers to spot differences), but those will have to wait for another post.


1 comment

Over at Viva El Birdos Dan Moore talked about the Cardinal’s offense not relying on the home run nearly as much as it had in the past.  I found the subject interesting and thought I’d dig into it in my own way.  A chart style I’ve used in the past is to see how a particular player’s wOBA was broken out by its components (i.e. how much of the wOBA came from BB, HR, etc.).  For this particular post we’ll start by shifting that chart style over to the team level.  The wOBAs represented ignore SB/CS and HBP.


Along the y-axis we have composite wOBA with the various sections of the stacked bars representing the various outcomes.  From this chart it appears that yes, the Cards aren’t getting a ton of value out of the HR (although they’ve hit some since Dan penned his article) relative to their peers.  Maybe the following chart shows it a little easier

Team_wOBA_percentThis chart breaks the data out by percentage, so an attempt to put teams on a common baseline for comparison.  From here it is a little clearer that the HR (purple) contributes slightly less to the cardinals value than to other teams (and hey, how bout those Marlins?).  Taking the same two charts and putting it on the individual Cardinal player level and we come up with



Hello Carlos Beltran.  That’s a lot of value from the home run, and pretty minuscule value from the double and walk.  Other than Carlos there isn’t a whole lot that’s jumping out at me as unexpected.



Going the Other Way

One of the themes I hear fairly frequently is the Cardinals ability and desire to go up the middle and the other way. I thought it would be worthwhile to see how well reality matches perception. First the numbers for going to the opposite field (all data in the post is derived using splits data and is for 2013 only)

Team wOBA
Reds 0.360
Brewers 0.356
Red Sox 0.351
Indians 0.351
Rockies 0.336
Giants 0.335
Rangers 0.333
Angels 0.325
Tigers 0.324
Astros 0.321
Royals 0.317
Athletics 0.309
Diamondbacks 0.308
Braves 0.307
Cardinals 0.304
Orioles 0.300
Twins 0.298
Mariners 0.299
Mets 0.300
Blue Jays 0.288
Nationals 0.287
Phillies 0.289
Dodgers 0.287
Yankees 0.285
Pirates 0.282
Rays 0.274
White Sox 0.272
Padres 0.272
Cubs 0.267
Marlins 0.267

Clearly not much to write home about here. The Birds sit dead in the middle of the pack. From an individual perspective the folks that you would expect to be near the top of the Cardinals leader board are there(Freese, Molina,Carpenter,Craig) and the bottom has folks that aren’t too surprising either (Beltran and Kozma).

Moving on to up the middle

Team wOBA
Diamondbacks 0.382
Orioles 0.371
Braves 0.362
Cardinals 0.361
Padres 0.361
Red Sox 0.359
Rangers 0.355
Dodgers 0.345
Astros 0.344
Angels 0.344
White Sox 0.342
Rockies 0.342
Rays 0.339
Cubs 0.339
Tigers 0.338
Nationals 0.337
Pirates 0.335
Mariners 0.334
Phillies 0.333
Giants 0.332
Blue Jays 0.331
Reds 0.329
Indians 0.328
Brewers 0.328
Athletics 0.323
Yankees 0.322
Twins 0.317
Mets 0.305
Marlins 0.293
Royals 0.291

Here we see the Cards bubbling more towards the top of the list.  For this the team leaders were Beltran and Craig, with Carpenter and Holliday not far behind.

Finally I have a table that compares the wOBAs as a percent of the wOBA to the pull field, to see if teams are nearly as good at hitting the ball up the middle and the other way as they are at pulling it (where all teams wOBA is highest).

Team Middle wOBA Oppo wOBA
Brewers 83% 90%
Giants 89% 89%
Reds 81% 89%
Royals 79% 87%
Angels 88% 83%
Rangers 88% 83%
Marlins 89% 81%
Indians 74% 79%
Diamondbacks 97% 78%
Mariners 87% 77%
Astros 83% 77%
Tigers 79% 76%
Nationals 89% 76%
Dodgers 91% 76%
Athletics 79% 75%
Twins 80% 75%
Mets 75% 74%
Red Sox 75% 73%
Phillies 82% 71%
Yankees 80% 71%
White Sox 89% 71%
Orioles 87% 71%
Cardinals 84% 71%
Rockies 71% 70%
Braves 81% 69%
Padres 91% 68%
Blue Jays 78% 68%
Pirates 79% 66%
Cubs 83% 66%
Rays 77% 62%

Interestingly the table seems to indicate that the Cardinals are getting more of their value out of pulling the ball than a lot of other teams. It’s not to say that they aren’t good at going up the middle per se (the previous table shows they are) just that they are really good at pulling the ball so far this year. Clearly all of this is still in a reasonably small sample size.


WAR Birds

Reading through Derrick Goold’s chat on stltoday earlier this week the following answer jumped out at me

I see all stats as part of a larger picture. I don’t think there is one stat that tells me everything. Wins for starting pitchers have value because they make a starting pitcher accessible and we know by reading that he won the game that two things happened a) he pitched five innings and b) he left with his team in the lead. RBIs tell me about the hitter and the situations that he’s hitting in. WAR tells me a bit about how the player fits into the larger scheme of his team. Yes, it’s flawed. And one of its biggest flaws is its not accessible, it’s not readily available or tangible or easy to understand for most fans. That bothers me. But I try to explain how it works and how it fits into the larger puzzle of a team’s success and I hope that helps some readers use it in context. That’s key. Every start needs context.


There are definitely things in there I agree with (context is important) and some I disagree with (WAR not being accessible), but either way I thought it would be a good jumping off point to discuss a few things WAR related.  On to a few bullet points

  • As with any stat it’s important to first identify the question you are trying to answer and then finding the best stat to apply to that question.  For example, if you’re trying to identify the batter that is best at getting a baser hits independent of how many bases he may get in that at-bat the batting average is your stat.  If you’re trying to identify the best at getting on base independent of how far he gets then on base percentage is your stat.  If you’re trying to determine who is the best player then the WAR framework is your best option.  I think this is part of what Derrick was getting at.
  • Note the second to last sentence above.  WAR is a framework, a construct if you will, that has varying specific implementations (Fangraphs and Baseball Reference for example).  At its essence WAR is a framework that combines offense, defense (both how you perform relative to you position and how difficult your position is), and base running using the currency of baseball (runs and wins).  How can you really argue with that?  Sure you can gripe about the individual components, but you’re more than welcome to substitute your own (just make sure you do it across the board).  Don’t like UZR or DRS, then input your own defensive values somehow.
  • Once you pick your implementation you can consistently apply WAR across players within a season or even across eras.  This is one of the beauties of WAR, it adjust for the run environment of the era.  Sure you can tell  me that player X hit 20 HRs in year Y, but then I also need to know about the run scoring environment in year Y to know how good that was.  WAR handles that inherently.
  • Be careful with decimals.  There are error bars in WAR (especially the defensive numbers) so you probably don’t want to vehemently argue that a player with WAR of 2.5 has definitively been better than someone with a WAR of 2.3.  We should probably just round to the nearest integer or nearest 0.5 and call it a day.

With those thoughts on WAR in the bag, let’s look at how the Cards position players are doing so far this year.

Name Batting Base Running Fielding Replacement Positional WAR
Matt Carpenter 14.4 0.9 3.9 7.3 0.7 2.9
Yadier Molina 10.7 -1.7 1.1 6.5 4.1 2.2
Carlos Beltran 10.8 -0.3 -5.8 6.2 -2.2 1
Matt Holliday 2.6 -0.4 2 6.7 -2.4 0.9
Allen Craig 7.8 -3.5 -1.3 6.9 -3.8 0.7
David Freese 2.4 0.3 -3.3 4.6 0.6 0.5
Pete Kozma -7.6 -0.1 3.2 6.2 2.5 0.5
Shane Robinson -1.6 0.7 3.5 1.6 -0.2 0.4
Matt Adams 3.2 -1 -2 1.9 -0.9 0.1
Daniel Descalso 0 0.4 -3.8 3.3 0.5 0
Jon Jay -4.7 -0.5 -6 6.8 0.8 -0.4


Apologies for the table, I’ll get that cleaned up for future posts, but let’s just take a row and walk through it.  Kozma seems like a fine example.  Pete is decently below average offensively (-7.6 runs), about an average baserunner, and a plus fielder (when compared to other shortstops).  On top of that he gets 6.2 runs for basically being on the field and 2.5 runs for the difficulty of the position he plays.  Add that all up and he’s been worth ~0.5 WAR.  Seems to pass the smell test for whatever that’s worth.  What do you, the reader, think?  Which Cards WAR seem out of line with your gut?




Subscribe to The Conclave via Email

Enter your email address to subscribe to this blog and receive notifications of new posts by email.

Join 85 other subscribers