As mentioned in my post from Saturday, we have reached the point in the season where some sample sizes are becoming significant in that we can infer trends in change of approach/skill for certain statistics. Since the Cardinals’ starting pitchers each have 150 total batters faced (TFB), it’s an opportune time to check in on their rates for strikeouts, ground balls, and line drives.

Keep in mind that I generated this graph on Saturday morning, so McClellan and Lohse’s most recent performances were not included. Of course, you can check out their respective pages at FanGraphs for updated statistical profiles. As with the batters, each pitchers’ career rates are shaded more lightly in the below bar graph.

McClellan’s results have been fantastic thus far (3.99/4.17 FIP/xFIP in 43.2 IP), but there is some cause for pessimism found in his greatly diminished strikeout rate. While it’s normal for a pitcher’s strikeout rate to drop when converting from the bullpen to the rotation, it shouldn’t by this much. I’m sure it’ll improve, but it’ll be interesting to see by how much (ZIPS’ updated projection has him at 6.25 K/9 for the rest of the season). There’s also some luck found in his HR/FB (7.3%) and strand (79.2%) rates.

Part of Lohse’s success in 2011 has been generating 6% more ground balls. That and rarely issuing walks has helped lead him to a refreshing 3.20/3.73 FIP/xFIP. Maybe he’ll justify some of that contract after all.

Looks par for the course for Mr. Westbrook despite his ugly results thus far (6.14 ERA in 36.2 IP). If he can harness a little bit more control (4.66 BB/9) and continue to generate ground balls, he should be fine. It should be noted that his 4.35/4.10 FIP/xFIP are relatively close to his career norms (4.17/4.00 FIP/xFIP).

Jaime Garcia’s batted ball profile looks similar to last season, and that’s a good thing. What’s so impressive though is that he’s added a strikeout per nine innings pitched to an already respectable career number (7.37 K/9) while reducing his walk rate by an equally impressive amount. That results in a stellar 4.00 K/BB ratio (this number doesn’t stabilize until 500 TBF) to start his sophomore campaign and it’s helped him craft a team leading 2.36/2.60 FIP/xFIP. Other fans around baseball are starting to take notice.

It’s easy to see why Carpenter hasn’t been the perennial ace that we’ve been accustomed to since he’s been allowing more line drives and fewer ground balls. With that said, he’s also been the victim of a high HR/FB rate (16.2%) as evidenced by his 4.26/3.43 FIP/xFIP.

Between McClellan’s competence, Lohse’s resurgence, and Garcia’s emergence, you can see how the Cardinals have somehow managed to withstand Wainwright’s season ending injury. We’ll see if I can write the same sentence in another month or two.

I’ll keep revisiting these thresholds as a majority of players meet them throughout the season. Next up for pitchers: Fly ball and GB/FB rates stabilize at 200 TBF.

Enhanced by Zemanta

At FanGraphs, Steve Slowinski called attention to the fact that many players are now reaching significant thresholds in playing time. Thanks to research from Pizza Cutter (all links can be found here), we know that players’ skill sets tend to stabilize after a certain number of plate appearances or batters faced. Steve explained what is meant by this well:

When I say “stabilize”, I don’t mean that these rates won’t change at all over the remaining course of the season. Instead, all it means is that once a player approaches these sample sizes, you can consider that there’s something more than just random variation going on: there’s some underlying change in a player’s approach/skill level/process/etc. in play as well. Matt Garza isn’t guaranteed to finish the year with a 12 K/9 rate because his strikeout rate has “stabilized”, but at the same time, I wouldn’t be surprised if his final strikeout rate is higher than what it’s been in the past.

So this isn’t to say that players will continue performing at current levels, but we can infer an upwards/downwards trend in approach or skill, even at these relatively small sample sizes.

If you followed the link to the FanGraphs article, you know where I’m headed. Most batters have now accumulated enough plate appearances to tell us something about their swing rate (50 PA) and contact rate (100 PA), while most pitchers have now faced enough hitters to lend insight into their strikeout rate (150 BF) and groundball rate (150 BF). While Steve’s article takes a look at all players across MLB, I thought it’d be useful to narrow the scope to our team.

Below is a graph for the Cards’ position players who have amassed enough plate appearances to be included. You can see the MLB average at the top of the bar graph. Numbers for 2011 are represented by the darker tones while career numbers are found in the bars shaded more lightly. Since Daniel Descalso only had 38 PA before 2011, I only included numbers from this season.

Overall, most guys sit pretty close to their career norms… but let’s take a closer look at a few:

Pre-DL David Freese’s .356 AVG looks pretty suspicious when you compare his contact percentage (71.7%) against the average major-leaguer (80.8%). While he’s making less contact, he’s actually swinging at a higher percentage of pitches than he has in the past. You’d be correct if you guessed that his BABIP (.460) was out of control. Obviously, this cannot be sustained. But Freese is making hard contact when he does put the ball in play as more than 30% have resulted in line drives. That’s not a predictive number but it is a descriptive one.

While I look forward to Post-DL David Freese returning to the team in a couple of months (fingers crossed), I remain perplexed by his precipitous drop off in power. In 750+ PA combined between AA and AAA, his isolated slugging percentage never dipped below .200. Nevertheless, it sits at .115 ISO in 398 PA with the big club. In fact, his career “success” to date includes a .396 BABIP.

StatCorner calculates players’ wOBA adjusting for their park and batted ball profile (they use acronym wOBAr). Freese loses 15-pts from his 2010 wOBA and 27-pts from his 2011 wOBA. I’m not saying that his potential contributions are overstated, but his numbers to date don’t exactly reflect his minor league performance (where he displayed decent power), and it’s scary to depend on a supposed power-threat to continue getting on base when he’s mostly relying on hitting singles and striking out 24.9% of the time. If the coaching staff considers this problematic for Colby Rasmus (25.8% career K rate), then why isn’t the same true of David Freese?

While Matt Holliday has decreased his swing percentage by 5.3%, Lance Berkman has increased his offerings by 4.7%. It’s hard to get too worried about either of these players. Holliday has already been worth 2.5 fWAR and there isn’t anything crazy about his batted ball profile (.481 wOBAr – StatCorner’s adjustment for park and batted balls).

Berkman’s case is a little more curious. While he has certainly benefit from a HR/FB rate ~20% higher than league average (only 10% higher than his personal norm), his wOBA (.511) is somehow lower than it should be based on StatCorner’s adjustments (.520 wOBAr)! Playing in Busch Stadium has robbed him of a little offensive value. Despite being a negative in the field – but not as horrendous as expected – Berkman has already been worth 2.1 fWAR. In other words, he’s pretty much justified the $8 million contract he signed this winter. The health of his knees will remain a concern until the final out of 2011, but kudos to TLR for continuing to schedule regular days off despite Berkman’s convincing offensive rennaisance.

What’s next on the horizon? For hitters, strikeout rate, line drive rate, and pitches per plate appearance “stabilize” at 150 TPA; walk rate, groundball rate, GB/FB rate “stabilize” at 200 TPA. Expect me to visit those as the season progresses.

I’ll try to make some observations about our starting pitchers in the next few days since a few of their rates have also “stabilized”: GB%, LD, and strikeout rate.

Enhanced by Zemanta

EDIT:  I should probably spell his name right…. oh well.  Apologies


Before Westbrook’s last start I put together the following chart.

The chart shows that Westbrooke is throwing inside more against both LH hitters and RH hitters in 2011 when compared to 2010.  We don’t know if this is on purpose or not, but will be something to pay attention to.  Later post coming on the success in these locations….


Ryan Franklin

Image via Wikipedia

Ryan Franklin is sure to hear more boos from fans. After getting ahead of the hitter (0-2) in yesterday’s game, he left a splitter over the middle of the plate that Miguel Cairo laced to left field to give the Reds a 2-run lead in the 8th inning. Of course, those weren’t his runners on base. And he did get the next four outs without allowing any more runs, even after accounting for an error by Theriot at SS.

In Derrick Goold’s P-D article about Ryan Franklin being officially demoted from the closer’s role, this quote caught my attention:

“I know that I’m going to go out there to the bullpen and whenever the phone rings and they say, ‘Franklin get up,’ I’ll get up,” the veteran said. “That’s what I know. … Stuff is fine. Everything in my arsenal is still there. I haven’t lost anything. However they want to put me out there, it doesn’t matter. I’m theirs. However they want to treat it I’m on board.”

Of course, with pitch f/x (available in a variety of places now including Brooks Baseball, Joe Lefkowitz’s site, and Texas Leaguers – the numbers in this post, however, can be found at FanGraphs), we can evaluate such claims to see if there’s been any change in a pitcher’s selection, velocity, or movement.

*Please note that none of the numbers in the tables and graphs below include numbers from yesterday’s game against the Reds (4/23).*

Franklin’s Pitch Selection
Year Fastball Knuckle-Curve Slider Cutter Changeup Sinker 2-Seamer
2007 48.7% 14.0% 20.6% 2.3% 7.6% 6.6% 0.2%
2008 50.6% 9.3% 22.7% 2.5% 5.5% 7.7% 0.4%
2009 39.1% 18.0% 5.0% 23.3% 12.1% 0.0% 1.6%
2010 37.6% 28.8% 1.0% 10.0% 8.1% 5.5% 8.5%
2011 25.7% 11.5% 0.0% 19.5% 7.1% 18.6% 17.7%

Franklin offers a good example of why pitch f/x analysis can be tricky. The system appears to have a difficult time differentiating between his pitches, especially fastballs, which have four different classifications: generic, cutter, sinker, 2-seamer. My guess is that the generic all-encompassing pitch is the one where the pitch f/x system dumps in fastballs that it doesn’t know how to categorize. As you can see below in the next table, it isn’t noticeably faster than Franklin’s 2-seamer, so I don’t think we can assume it’s a 4-seamer. Whatever it is, he’s definitely thrown fewer of them, but given the spike in sinker/2-seamer usage (which, for all practical purposes, are the same pitch), it could be that some of the generic fastballs are now being categorized as sinkers/2-seamers. I guess the bottom line is that we can’t make any confident assertions given the volatile nature of his pitch type classification.

It is clear, however, that Franklin has been relying on his off-speed pitches less than usual this season. Over the course of his career, Franklin has chosen breaking pitches (knuckle-curve, slider, or changeup) 36.9% of the time, but in 2011, he’s only done so 18.6% of the time. Of course, this begs the question, “Why?”

I can think of two answers: (1) As usual in April analysis, one reason could be small sample size. The guy has only thrown 113 pitches this season; that’s 11.5% of the amount of pitches he threw in 2010. Slice anyone’s season up into tenths and I’m betting that you will find some statistical quirks. But do that with a reliever, whose entire season of performance is in itself a small sample size, and you’re going to find some silly numbers. (2) Or perhaps he’s struggling with command of his pitches – he does have a higher walk ratio than normal (4.05 BB/9) – and has been forced to work from behind in the count. More on this later.

Franklin is 37-years-old. It wouldn’t be a surprise if this was simply the decline phase of his career. And for players of his age, drop-off in performances can happen precipitously. But check out his velocity below:

Franklin’s Velocity
Year Fastball Knuckle-Curve Slider Cutter Changeup Sinker 2-Seamer
2007 92.1 77.9 86.7 89.9 83.8 84.8 92.6
2008 91.8 77.0 85.7 89.6 84.9 85.9 89.4
2009 91.2 77.2 85.3 89.5 84.5 - 90.9
2010 91.2 78.3 80.3 89.2 84.7 83.8 90.8
2011 91.0 77.7 - 89.3 82.3 82.5 90.9

Nothing too crazy there. Maybe there’s something evident in the movement of his pitches:

Franklin’s Movement (Horizontal/Vertical)
Year Fastball Knuckle-Curve Slider Cutter Changeup Sinker 2-Seamer
2007 -4.4/8.5 6.3/-4.9 1.2/3.5 1.3/7.3 -5.4/1.8 -5.0/2.1 -14.0/8.4
2008 -5.3/8.2 5.8/-5.8 0.8/2.9 0.2/7.0 -5.9/2.1 -6.7/1.0 -1.2/3.9
2009 -5.1/7.8 5.8/-6.2 1.8/2.9 -0.7/6.8 -5.7/1.4 –/– -8.9/6.8
2010 -5.8/8.9 4.2/-5.0 2.8/-0.8 -1.0/7.4 -6.7/1.4 -4.9/2.8 -9.0/7.0
2011 -6.2/8.7 3.0/-1.5 –/– -1.8/7.8 -5.8/-0.2 -2.7/2.3 -8.7/6.8
Career -5.3/8.3 5.1/-5.4 1.0/3.0 -0.7/7.0 -6.0/1.6 -5.5/1.8 -8.7/6.8

The generic fastball seems to be within normal range; same for the cutter and 2-seamer. The sinker has lost a couple of inches in horizontal movement. But we’re not even sure if that pitch is worthy of its own distinction since sinkers are a type of 2-seamer. The knuckle-curve does appear to be noticeably flatter than previous seasons, but he’s only thrown 11 or 12 of them in 2011, so it’s way too early to attribute any significance to that observation. And, with that, he’s thrown even fewer changeups, so we really can’t make anything of the 1.8 inch difference in vertical movement on that pitch.

Onto the plate discipline numbers. How are pitchers behaving against Franklin in 2011? What have they looked like against him throughout his career? And what do batters look like, on average, against any pitcher?

Franklin’s career follows that of the average 2011 pitcher pretty closely when it comes to the plate discipline of batters faced, but there are a couple of exceptions. Over his career, Franklin has thrown more pitches in the strike zone and, particularly, more first pitch strikes than the average pitcher. However, he’s thrown fewer first pitch strikes thus far in 2011. Franklin has thrown first pitch strikes 64.7% of the time over his career (that number is probably higher in his time as a Cardinal) , but only 53.1% of the time in 2011. For someone accustomed to working ahead in the count, it’s easy to imagine how he’d be tempted to serve up some fat pitches trying to reclaim the count.

But that’s not the only interesting aspect of this graph. Batters are swinging at more pitches outside of the zone and fewer pitches inside the zone, but maintaining absurd levels of contact, especially on pitches outside of the strike zone. In general, a pitcher’s ability to induce swings at balls outside of the strike zone is a positive development since those are the type of pitches you’d expect to generate weak contact. And you know what’s crazy? Franklin’s batted balls in play are surprisingly normal, if not improved from career trends. His 12.5% LD rate is down seven percent, 46.9% GB rate is up seven percent, and 40.6% FB rate is down four-tenths of a percent. Nevertheless, Franklin will enter his next game with an ugly 10.30 FIP.

How is this possible given all that’s been detailed in this post? His velocity is in tact. There appears to be similar movement on his pitches. He’s getting hitters to swing at a higher percentage of pitches outside of the strike zone and fewer pitches inside of it. He’s allowing fewer line drives and generating more ground balls. What gives?

Check out his xFIP (5.69) for a clue. Normalize Franklin’s HR/FB rate, which currently sits at an unbelievable 30.8%, and his numbers come back down to earth a little bit. No, that’s still not what you want from a guy who’s supposed to be handling the higher leverage innings out of the bullpen, but I’m betting you can find other stretches over his career when he’s seemed just as helpless on the mound. April is a time for minuscule performances to be placed under a microscope and greatly exaggerated, whether it be Ryan Theriot’s early success batting leadoff or Ryan Franklin’s failure to secure the last out of games. Time will tell which trends are sustainable. Franklin said himself, “I haven’t lost anything.” As much as I wanted to roll my eyes at his insistence, maybe he was right. I certainly couldn’t find anything in the numbers to suggest otherwise.

Enhanced by Zemanta

Sorry we went all AWOL for a few days. These game analytics are a little more fun when they’re done individually.

At any rate, follow the jump for shotgun analysis of Carpenter’s return to form, Westbrook’s ongoing struggles, Lohse’s dominance, closer controversy, and Pujols-ian slumps.

Also, the Cardinals are above .500; and I’m pretty sure that’s the first time this year.

Continue reading »

Without needing to rehash the events of today’s game it is fairly clear that something needs (and likely will) be done about the closers role for the Cards.  So that seems to pose two separate questions

  1. What will Tony and Dunc do?
  2. What should they do?

I’ll take them in that order.  It seems that the scuttlebutt points toward Miguel Batista’s aged carcass taking over the closer’s role.  As I type this it is far from official as the decision makers still have to meet on it.  That said I think Batista’s veteren-ness and experience will win out over some of the younger guys stuff and potential.

The more interesting question is what should they do?  All of the likely successors have some issues, some major, some minor.  The young right handers all have some platoon issues with lefties, Batista is simply not good at pitching, and the lefties aren’t very good against righties.  With that said, if I were king I would scrap the traditional vision of a closer (as would most sabermetricians in this setting) and platoon my pitchers accordingly in the appropriate leverage situations.  I would use Boggs in the highest leverage situation that will be predominantly RHBs, Motte the second highest, and Sanchez the third.  I would use Miller in the highest leverage situation that sees predominantly LHBs.  Sure this would take some finagling as I’m not extremely confident with any of the above listed against opposite handed hitters, but since Tony is likely to carry 97 pitchers matching up late should not be a problem.

A final bonus question is what should happen to Franklin?  Despite not being able to miss bats in years past he has still had a FIP below 4 in 3 of the last 4 seasons.  There is some value in keeping a guy like that around in your pen, just not in high leverage situations.  I would personally like to see him take the Batista role and have Batista sent packing.  Unfortunately if Franklin gets Batista’s role it is because Batista and he have just flipped, not because Batista got voted off of the island.

Game 15

Link to page

The Good: The offensive surge continued as Allen Craig provided the fireworks in this one. His 3-run homer in the 5th inning improved the team’s win expectancy by 25.5%. It did my heart good to see TLR resist the temptation to play Berkman’s streaking bat despite his pronounced split (.423 wOBA as LHB; .338 wOBA as RHB). Of course, even his “bad” split is about league average… though it’s been much worse in recent seasons. At any rate, Craig needs at-bats, and this is an appropriate way for him to get them.

The Bad: Again, there just really isn’t a lot of blame to pass around in this one. The worst win probability belonged to Tyler Greene at -.057 WPA, but even he contributed a hit and 2 SB (though he easily could have been picked off on the second one had Macdougal’s pick-off not been thrown to CF instead of 2B).

Gerald Laird’s non-tag at home plate on Andre Ethier was definitely the most baffling play of the game (see it for yourself and enjoy the narration of Vin Scully). Pujols had time to double-pump before making a throw home that still beat Ethier by several steps, but Laird headed up the  line instead of blocking the plate completely whiffed on the tag in the process. It didn’t cost the Cardinals last night… but it a closer game, that could have really hurt.

The Impressive: Eduardo Sanchez struck out the side in the bottom of the ninth to mop up the win. I don’t know if you’re counting (I am), but he’s now struck out 8 of the 11 batters he’s faced. He’s impossibly lowered his FIP to -2.27(!). With this kind of dominance, Franklin’s room for error will be even slimmer in the fans’ eyes. Don’t get me wrong, I don’t think TLR is anywhere close to supplanting Franklin in the closer role, but I’m not naive enough to believe that fans won’t start clamoring for this decision given Sanchez’s hot hand. Of course, we’re talking about three innings here. Despite the cause for hyperventilation, he’ll come down to earth at some point.

One more thing: In an unbelievable turn of events, the Cardinals now lead all of MLB in offensive production (.361 wOBA); the Cincinnati Reds come in 2nd (.357 wOBA). See the FanGraphs leader board.

Enhanced by Zemanta

Bernie beat me to the analytical punch with the info in his article today in the PD (on line version here) and Albert went off last night with two home runs both of which were on the inner half of the plate.  That said I did have a few interesting pieces of info to share that I found in my investigation.  Breaking the plate into 4 horizontal slices and plotting the percentage of pitches in each zone shows that Pujols has been attacked more inside so far this season (SMALL SAMPLE SIZE) than last

The chart is from the catcher’s perspective, so -1 is slightly off the inside corner.  It’s not anything conclusive because of the small sample size, but it is something to keep an eye on the rest of the year to see if the trend continues.

To look into possible causes, I turned to AP’s performance against high velocity to see if there were any indications of a slowing bat.  The following table has the run value per 100 swings against pitches 95 mph and greater by year

2008 2009 2010
rv100 1.81 2.61 -5.15
sample size 98 125 103

I wonder if other teams spotted something similar and are seeing if they can attack more hard in.  We’ll have to wait and see if 2010 was an anomaly or something more.

PHOENIX, AZ - APRIL 12:  Lance Berkman #12 of ...

Image by Getty Images via @daylife

The Cardinals now rank second in the league in team wOBA (.356). This is not a joke. Since I wrote that piece about their slow start offensively merely one week ago, they started scoring runs at a torrid pace. Their BABIP has jumped 61-pts (.324)! They have hit fewer ground balls (48.8%), more line drives (20.2%); their fly ball rate hasn’t changed much (31.0%), but they’ve cleared the fence at a more realistic rate (12.9%). While they’ve started walking a little less (7.9 BB%), they have the third lowest strikeout rate in the league (16.7 K%). Altogether, the Cardinals have transformed from an unlucky team to a surprisingly fortunate one in the span of seven days. Fans, let’s keep our cool. Much like we shouldn’t have panicked when the offense was seemingly MIA after one week, we shouldn’t be quick to anoint them league leaders either. Perspective, friends; you can either have it now or the long season will eventually force it upon you.

With that said, it has been a lot of fun watching Lance Berkman hit 6 home runs in the span of 5 games.

Let’s take a look at the past few games…

Continue reading »

Over at Fangraphs, with the help of Dan Szymborski, they’ve finally rolled out their Rest of Season ZiPS Update, along with a final season update of the end-of-season numbers.  I’ve compiled all of the Cardinals hitters’ numbers in the spreadsheet below (y’know, with a protractor in my mother’s basement) with some notes below:

Click to embiggen

For those who don’t know how these work, Dan takes the players’ numbers-to-date, which are “in the books,” and regresses them to his true talent projections to get the “rest of season” numbers.  Then he adds those rest of season numbers to the players’ numbers-to-date to get the final statistics.  For example, he thinks that Colby has outperformed his true talent up until now, so Colby’s rest of season numbers are .268/.349/.450 instead of the .375/.464/583 he’s hit so far.  However, since those 56 PA are already in the books, they get added onto 533 PA of .268/.349/.450 ball to get his final line of .278/.360/.463. Also note that the projections assume that each player plays full time.

Notes, in bulleted format:

  • Pujols’ end of season wOBA would be the lowest of his career
  • I would love Berkman’s final line, though I would probably take the over on his OBP. Anecdotally, he looks like classic Berkman so far this season
  • Give me a firm over on Theriot’s numbers. I think the walk rate so far this season is for real, and would guess that .350 would be the 50% mark for his OBP
  • Yadier Molina’s line looks almost EXACTLY like his 2010 line. 2010: .262/.329/.342. 2011: ..263/.329/.348
  • Holliday’s projection baffles me.  I understand that ZiPS projects an aging curve, but he projects a .317/.388/.543 career hitter to hit .297/.378/.494 for the rest of the season. Looking at the numbers, it’s almost entirely attributed to a regression in BABIP (.329 vs a career .344), so I understand why the projection sees Holliday this way. Still, for a player who hits the ball as hard as consistently as Holliday does, I could see him being a true talent .335-.340 BABIP guy.
  • I love Colby Rasmus. He’s a Fire Burning in the Outfield. That is all.

I’ll be traveling for the next few days, but I’ll try and get to the pitchers when I get a chance.

© 2011 Gas House Graphs Suffusion theme by Sayontan Sinha