K-BB-HR Rod: Taking a closer look at his stuff

Discussing Francisco Rodriguez stuff, 2008-2010 numbers, more stuff, leverage and even more stuff.

Introduction

Much has already been said about overvaluing "savez" or overpaying for closers. Just a quick reminder of Francisco Rodriguez contract status, per Cots as always:

$2M signing bonus, 09:$8.5M, 10:$11.5M, 11:$11.5M, 12:$17.5M option ($3.5M buyout)

2012 option becomes guaranteed with: 55 games finished in 2011, and 100 games finished in 2010-11, and doctors declare Rodriguez healthy after 2011;

Rodriguez receives $3.5M termination buyout if 2012 option does not become guaranteed limited

no-trade clause allowing Rodriguez to block deals to 10 clubs

There is no chance he will be worth the contract (vest gah! /slitwrists) or that the Mets will trade him. So, leaving this aside for a second, let's take a look what we are "stuck with".

Considerations

1) Even the best closers will blow saves, give up walk-off homeruns or fail in the big spot. No matter much money as you throw in the position, your best reliever will give you the best chance to win, but won't guarantee it. It's frustrating when it happens to our team (and it's fun when happens to others), but it is part of the game and it doesn't mean the player is "awful".

2) Fangraphs WAR for closers is a different kind of animal (based on FIP). 

The average Leverage Index of a closer is about 1.8, meaning that each plate appearance is about 80 percent more important than an average PA. We give the closer credit for half of that, based on the principle of chaining. Because relief pitchers are mostly fungible, and can move from one role to another if needed, replacing a closer is not the same process as replacing a starter or a position player. If a team’s closer gets hurt, they do not then call up a replacement level reliever from the minors and use him to close out games.

BtB had a post explaining the chaining and why half of the leverage is credit to the closer (Bullpen Chaining and Reliever WAR). Statcorner WAR doesn't account for leverage and it's based on tRA.

3) Declining fastball velocities is a sign of a declining pitcher, but we have to check how it is really affecting the pitcher effectiveness (results, peripherals, pitch selection and control for example). Though it doesn't mean everyone is like Tim Lincecum.

Basic Statistics and Nicknames

Sample size for pitching statistics to become reliable: 

Pitching Statistics:

150 BF – K/PA, grounder rate, line drive rate

200 BF – flyball rate, GB/FB

500 BF – K/BB, pop up rate

550 BF – BB/PA

Basically, only after half a season K%, GB% and LD% will stabilize, and after roughly 2/3, FB% and GB/FB. We would need 2 seasons worth of data for IF% and BB% to be reliable, so let's compare the 2007-08 and 2009-10 periods. So, the 2010 data will be taken cautiously, but as a part of a multi-year trend.

Looking at his components, there was little change in his K% and BB%. Calling him BB-Rod isn't exactly new, his BB% has remained steady, above the 9.5% average for RP. And you can still call him K-Rod, since he still strikes out considerably more than the 19% average for RP.

His GB% has been declining since last year and he has been inducing more pop-ups (IF%) and also allowing more HR per ball in the air (HR/BIA). There is not enough sample to stabilize his IF% yet, and it's unlikely he'll keep this rate, but it has been trending up (part of that you can thank Ike Davis, and again, and again, ops this one was Ollie's). All this might suggest a change before the 2009 season, and this make him more risky too, given these extreme outcomes (Theories: pitch location, pitch selection, pitch stuff).

Since 2008, he has posted tRA+ of 122, 112 and 130 or non-leveraged WAR of 1.0, .6 and .3 (calculated base on tRA). For the 2010 data, his current BABIP and LOB% are going to regress anytime soon.

Note: all his Zone%, 1st pitch strike%, zone swing%, zone miss% and outside swing% are hovering around MLB average. What really impress is his outside miss%, above the 27.6% average.

Repertoire

He currently throws four-seam fastballs (FA), curveballs (CU), change-ups (CH) and sliders (SL). There is a discrepancy at the fangraphs page. If you look at his main page, you are lead to believe he completely abandoned his slider in 2009. However, if you click on the "pitch f/x" tab, you can see his correct pitch usage. It might be that the data is taken from different sources or there was a reclassification of pitch types, but the pitch f/x tab data is confirmed by the Texas Leaguers pitch f/x.

Pitch Selection, Swing and Whiff (2008-2010), aka "STUFF"

All data from Texas Leaguers, except for wPitch/C (runs above average per 100 pitches) and wPitch (runs above average) from Fangraphs. The Strike/Swing/Whiff MLB avg from 2008-2010 in the table are very similar, there is just a slightly different classification for fastballs in 2008.

In 2008, he threw more curveballs than change-ups. In 2009, he changed his mix, throwing more fastballs and change-ups and in 2010 he is using his change-up and curveballs at the same rate. Looking at the table, it's possible to see that his change-up has always been very good: he can control it for strikes and batters will swing at above average rate, and the whiff rate is more than double the average. One thing that changed during the years was his curveball. In 2008, he had average control, swing and whiff rates, but in 2009 he couldn't throw it for strikes and the batters were just laying it off. In 2010 he regained some of its control, but it is still below average, resulting in negative wPitch/C.  As for his fastball, it previous years it had been close to average value-wise (wPitch/C). This year, he improved his control a bit and, coupled with the fact that batters swing less than average against it, the pitch value has increased substantially. 

R/L Splits

Since change-ups are very effective against opposing handed batters, this might explain why his xFIP in the period was better against lefties than righties as well as a better GB/FB ratio (Fangraphs split data). His IF% (infield flyouts) for 2009 and 2010 are similar, but slightly higher against righties.

As for his pitch selection against LHB, he used a 45%/25%/25% (fastball, change-up, curveball) in 2008 and 50%/28%/15% in 2009 and 2010. The increase usage of the fastball and reduced curveballs might explain the difference in his batted ball profile, generating more pop-ups and also more flyballs/HRs (The "Johan Santana effect"). Another thing that could increase the pop-ups and reduce groundballs is pitching high and inside, but there is not enough evidence to suggest this.

Against RHB, 57%/5%/30% in 2008 to 60%/16%/13% in 2009.  He started to throw more change-ups against  them, with +20% Whiff rates. This can be related to his lack of curveball command, as he can control it better, he might use it more against RHB, like he is doing this year (50%/15%/25%).

Pitch Speed, Movement, aka "More Stuff"

Even though his fastball velocity is down this year, last year he also started with lower velocity and raised as the season progressed. He is also keeping the same "distance" to his change-up velocity. He raised his curveball vertical down movement and spin rate and reduced the vertical and horizontal movement and spin rate of his change-up. One interesting thing is that he is modifying the spin rate of his change-up this year, as you can see below.

He doesn't use much his slider and in the graphs they look all over the place (they don't cluster in one spot), and you can see now why some pitch f/x (and maybe his main Fangraphs page) could lump it with other pitches. It looks like his change-up with opposite horizontal movement.

Comparing the vertical and horizontal movement of his pitches to the Run Values by the Baseball Analysts article, we can see how his stuff compare to others stuff.

Fastball

Change-up

Curveball

Surprisingly, his change-up doesn't have extreme movement, but it's very effective. It's probably because of the difference of speed to his fastball, control and some deception. Those dots were based on the average movement, but it varies a lot. It's possible to see that a poorly placed fastball with -5 horizontal movement and +9 vertical movement can be trouble.

Conclusion

Based on the recent increasing flyball tendencies he has shown so far, his decreasing GB% and increasing HR rates are worrisome, partially compensated by the increase in pop ups. His different pitch selection can be partially explained by the effectiveness of his change-up and the lack of command of his curveball, specially last year. He has shown more control of his curveball (and fastball) in 2010, and that pitch can be used to get more righties out. His decreasing fastball velocity hasn't affected him much and this year he is controlling it better (SSS).

Following Fangraphs WAR, one way to increase his value would be to use him in high leverage situation, but that is obstructed by the mainstream-managing-closer-thinking that they are only going to pitch in the 9th inning, on a save situation. It's not a problem with just Jerry Manuel. Anyway, this has already been discussed in AA. Another thing would be if the Mets were projected to be in the playoffs. In short series of high leverage games and some more innings, the closer value would rise. A mid-to-low 3 tRA is still very good (not anywhere close to $ 17.5 good), but still good enough for a team playing the postseason (hey, maybe next year! j/k). 

Reference

Texas Leaguers Pitch F/x

Brookes Baseball Pitch F/x

Joe Lefkowitz Pitch F/x (error)

Fastball and Changeup Run Value by Movement - Baseball Analysts 

Francisco Rodriguez Fangraphs Page - Pitch F/x

Francisco Rodriguez Statcorner Page

Francisco Rodriguez Baseball Reference Page

Pitch Identification Tutorial - Baseball Analysts 

When Samples Become Reliable - Fangraphs (The original work can be found here) / Saberlibrary / DRaysBay

WAR and Relievers - Fangraphs

Bullpen Chaining and Reliever WAR - BtB

Changeup Has Become Great 'Equalizer'