One of my more peculiar fascinations is with the first pitch of at bats. Specifically, I seem to pay an inordinate amount of attention to players and teams that swing at the first pitch more or less often than you would expect. With the pitch-by-pitch database that I happen to have, it is actually a trivial exercise to extract that information, so for tonight I decided to take a look at the first pitch on a team level for 2008.

Namely, I wanted to investigate the percentage of first pitches that a given team swung at. For additional context, I included that team’s walk rate under the theory that you would expect some negative correlation between how often a team swung at the first pitch and how often they drew a walk. The difference between the two is included as the final column with a positive difference expressing that the team drew more walks than expected.

Despite the promising sign of the top and bottom teams matching up perfectly, the dataset (limited as it is to just 30 data points) shows little correlation.

However, visual inspection reveals a possible pattern. It looks like a majority of the teams do follow a roughly a linear pattern between not offering at the first pitch and drawing walks, there might be two clumps perpendicular to the hypothetical trend line. There are two teams (Seattle and Pittsburgh) that took first pitches often but didn’t translate those into walks and a clump of six teams (Atlanta, Texas, Tampa, Chicago(N), St.Louis and Cincinnati) that drew walks at an above average pace despite offering at the first pitch more often than average.

Is there something (like a higher or lower percentage of free-swinging, high-slugging hitters) about the make up of the teams in those two groups that helped distinguish them from the other 22 teams, who show a correlation of 0.78 between the two axes? In part two, we’ll add in some data from past years to get a bigger sample size and if the pattern holds, see if we cannot tease out a possible explanation.

This is my first post at FanGraphs, and I would like to thank David Appelman for inviting me onboard. I have previously written for Seamheads.com and StatSpeak.net, and frequent “The Book” blog. If you’d like to know some more about my background, check out this article I wrote a few months ago.

Today I am going to start off by climbing up on my soapbox to address one of my pet peeves, the use of Line Drive rates as a predictor for Batting Average on Balls in Play (BABIP). The standard practice is to estimate BABIP by LD/Balls in Play + .12. It is claimed that LD rateas are more stable than BABIP from year to year, and that when the actual observed BABIP varies from the predicted by a large margin, this indicates a future regression to the mean.

I’m in the process of updating my park factors for 2008, along with adding in 1999, 1955 and 1953 that the folks at RetroSheet have included in their most recent release. I’ve added a couple more categories, foul flies and line drives. Now, I’ve never heard anyone mention park factors when using LD rates, but in fact they are quite large. I might guess that there could different opinions of what is a line drive from one ballpak to another, or maybe it’s the air or the hitting background. I limited my LD factors to 2003-2008, when the RetroSheet data has complete information on whether a ball is a line drive, ground ball, fly ball or popup on every batted ball, including hits. In Arlington, a batter is 18% more likely to have a batted ball coded as a LD, which may have helped Milton Bradley to have the 2nd highest LD rate in 2008 – while in Minneapolis, it’s 20% less likely. Four of the lowest six LD rates belong to Michael Bourn, Geoff Blum, Ty Wigginton and Hunter Pence, and Minute Maid Park has the second lowest LD park factor at 0.82. This is not saying that Houston batters hit fewer line drives – it’s that Houston and it opponents both have 18% fewer balls scored as liners in Houston than they do on the road.

PARK_ID PARK_NAME            First   Last    PAw     LDf	   
PHI12   Veterans Stadium     2003    2003    4768   1.23	   
ARL02   Ballpark Arlington   2003    2008   26850   1.18	   
TOK01   Tokyo Dome           2004    2008     283   1.13	   
CIN09   Great American       2003    2008   28827   1.11	   
DEN02   Coors Field          2003    2008   29158   1.10	   
STL10   Busch Stadium III    2006    2008   13967   1.09	   
KAN06   Kauffman Stadium     2003    2008   27530   1.09	   
WAS11   Nationals Park       2008    2008    4790   1.09	   
TOR02   Rogers Centre        2003    2008   27513   1.08	   
SFO03   Phone Co Park        2003    2008   29439   1.07	   
MON02   Stade Olympique      2003    2004    7684   1.07	   
STL09   Busch Stadium II     2003    2005   14280   1.06	   
STP01   Tropicana Field      2003    2008   27830   1.06	   
DET05   Comerica Park        2003    2008   28008   1.06	   
PHI13   Citizens Bank Park   2004    2008   24640   1.06	   
MIL06   Miller Park          2003    2008   29354   1.06	   
WAS10   RFK Stadium          2005    2007   14885   1.05	   
OAK01   Oakland Coliseum     2003    2008   26719   1.03	   
SEA03   Safeco Field         2003    2008   26683   1.01	   
CHI12   Comiskey Park II     2003    2008   28644   1.00	   
NYC16   Yankee Stadium       2003    2008   28722   1.00	   
MIA01   Dolphin Stadium      2003    2008   29849   1.00	   
CLE08   Jacobs Field         2003    2008   28136   0.99	   
BAL12   Camden Yards         2003    2008   29103   0.99	   
PIT08   P.N.C. Park          2003    2008   27652   0.98	   
PHO01   Bank One Ballpark    2003    2008   28810   0.98	   
SJU01   Hiram Bithorn        2003    2004    2598   0.98	   
SAN01   Jack Murphy          2003    2003    4943   0.98	   
LOS03   Dodger Stadium       2003    2008   29555   0.98	   
CHI11   Wrigley Field        2003    2008   28663   0.96	   
SAN02   PetCo Park           2004    2008   24432   0.95	   
NYC17   Shea Stadium         2003    2008   29299   0.92	   
BOS07   Fenway Park          2003    2008   28311   0.86	   
ATL02   Turner Field         2003    2008   29016   0.86	   
ANA01   Anaheim Stadium      2003    2008   26490   0.86	   
HOU03   Minute Maid Park     2003    2008   28271   0.82	   
MIN03   Metrodome            2003    2008   28048   0.80

Point Two – are line drives really more predictive? It’s said that if a player’s BABIP is not close to his LD+.12, that it’s becuse of luck, and this should be expected to correct itself next season. Expect the overachiever to come back to Earth.

For all the batters from 2003-2008, in non-bunt plate appearances, I added up the base hits, line drives, ground ball, fly balls and popups. I compared the predicted BABIP to the observed one in each season, which showed a root mean square (RMS) error of .045. Then I compared each years predicted value to the next years observed, and the RMS was .048 – slightly larger. For pitchers, the RMS was .039 in the same season, .039 in the next. I don’t see the evidence of future regression.

Complete line drive data is only available since 2003, and for a few seasns in the 1990s. In the seasons when it was not available, a “true talent level” of BABIP can be estimated by using a rolling weighted mean of past data, commonly referred to as Marcel. I used a seasonal weight of 0.7 – the most recent season is weighted at 1.00, the one before that at 0.70, two seasons back at 0.49, etc, each previous year 0.7 times the next. In this test, I did not use any regression to the league mean. The RMS of LD+.12 compared to the Marcel for the same season was .048 for batters, .046 for pitchers. The Marcel compared to the observed BABIP in the NEXT season was .041 for batters, .039 for pitchers. Historical BABIP data is better than the current season’s LD rate.

If LD data is available, so are GB, FB & PU. I tried a more complex model using .15*FB+.24*GB+.73*LD to estimate BABIP. This worked much abtter at reducing the mean errors, even surpassing historical BABIP. For batters, the yearly RMS came down from .048 to .036, for pitchers from .041 to .031.

Still, you can’t assume that every batter has the same rate of hits on their ground balls. Some batters hit more balls to the left side than the right, some run fast and some run slow. Instead of trying to profile each batter on each type of batted ball, I will continue to use Marcel to weight each batter’s historical BABIP in my projections.

On the other hand, DIPS theory states that a pitcher has little control over the outcome once a ball has been put into play. There is clearly an ability to be a flyball or groundball pitcher. Line drives are considered mistakes, and that may be evidenced ny looking at the six-year totals which show the lowest LD rates nelonging to Mariano Rivera, Fausto Carmona and Derek Lowe, while the highest belong to guys like John Van Benschoten, Edwin Jackson and Tony Armas Jr. Using the FB-FB-LD estimator on the six-year totals drops the pitchers RMS all the way down to .016.

Even so, some pitchers consistently defy the estimates. Roger Clemens, Brian Bannister, Chien-Ming Wang, Carlos Zambrano, Dan Haren, Brandon Webb, Chris Young and Greg Maddux all do at least .020 better than estimated. On the other end, Zach Duke, Sidney Ponson and Glendon Rusch all under perform by at least .020. Is it the ballpark? Is it their defense? The batters they faced? Or is it their own skill or lack of it?

Here’s my plan (I won’t have the answers next week) I want to compile park factors for each type of batted ball in each ballpark – what is the normalized rate of hits for flyballs to left in Dodger Stadium? Then do a WOWY analysis of fielders, showing the rate that each fielder allows more or fewer hits than expected on each groundball, flyball, linedrive and popup. Finally, each batter’s rates. Then go back and look at how many times each pitcher faced each batter, and with which fielders, and in which ballparks. Once those are controled, see how many hits, plus or minus, are left over for each pitcher.

For the last day or two we have begun to delve deep into the baseball economy and how it relates to fair market value, free agency, and context. This morning, our discussion centered around the necessity to add context in order to more accurately represent free agency analyses. This afternoon, Dave brought forth a great evaluative tool, showing average dollars/win figures for all 30 teams. The figures in the spreadsheet were arrived at after concluding that a team of replacement players would cost $12 mil and win 50 games; that the goal of each team would be to win 40 games above replacement; by subtracting the $12 mil from the estimated payroll; and finally by dividing that difference by the 40 wins above replacement.

The logical next step is to actually put this into action, combining an analysis of fair market value with the context of a specific team. With the news that the Brewers have decided to offer arbitration to Ben Sheets, he seems like the perfect person on which to test our new methods. To make the post a bit easier to read, I’ll break it up into different sections.

Brewers Payroll
Over the last three years, the Brewers have increased their payroll from $57 mil to $80 mil. If we factor in about a 5% markup, their 2009 payroll can be estimated at $84 mil. Therefore, they would have $72 mil ($84 mil payroll – $12 mil replacement payroll) to win 40 games (90 wins – 50 replacement wins). The quotient amounts to an average of $1.8 mil/win. Now, this is just an average, and does not imply that every signing the Brewers make has to be for $1.8 mil/win. However, every player of theirs that exceeds $1.8 mil/win will necessitate an adjustment in what they have left to spend.

Ben Sheets & Fair Market Value
Weighting the Bill James and Marcel projections for 2009 puts Sheets at 180 innings with a 3.60 FIP, very solid numbers. He might not be the guy who, in 2004, produced a 2.65 FIP, 2.70 ERA, and > 8.0 K/BB, but he still has the ability to dominate. A replacement level starter would log 150 innings with a 5.50 FIP while a replacement reliever would pitch the remaining 30 innings at a 4.50 FIP.

Ben Sheets:     180 IP, 3.60 FIP, 72 runs
Replacement SP: 150 IP, 5.50 FIP, 92 runs
Replacement RP:  30 IP, 4.50 FIP, 15 runs

All told, that would put Sheets at 72 runs and the replacement level at 107 runs. This +35 run advantage for Sheets puts him at +3.5 wins above replacement. Normally, we would add a quarter-win or half-win for the fact that the starter in question would pitch the innings and help save the bullpen. Sheets does not get this advantage from me, and given his inability to consistently make more than, say, 22 starts in a season, I’m actually going to dock him a quarter-win, bringing his value to a solid +3.25 wins above replacement.

The going rate right now is $5 mil/win, so if Ben were to sign a 1-yr deal, the appropriate fee would be $16.25 mil. Factor in a 10% discount for a multi-year deal, and his 3-yr deal would be worth $44 mil, an average annual value (AAV) of $14.7 mil. That AAV divided by his 3.25 WAR amounts to a fee of $4.52 mil/win, about two and a half times higher than the Brewers’ average of $1.8 mil/win.

The Contract & The Team
If we knew nothing else about the Brewers, other than that they were striving for 40 wins above replacement and had 73 mil left with which to accomplish the feat, then signing Sheets to this deal would mean they now had 37 wins to add with about $58 mil to spend. We do know about the Brewers, though, and thanks to the guys at BrewCrewBall, calculations about returning players and their salaries can be made.

According to their figures, the Brewers will have 16 players returning to the roster, at an approximate sum of $66,449,000. The link suggests the total is closer to $73 mil, but Salomon Torres retired, and I’m unsure what is going on with Craig Counsell, so they were both removed. Calculating the wins above replacement for all 16 players, the total comes to about 35 wins, at the $66,449,000. This means that the Brewers have $17,551,000 left to spend in order to accrue the five remaining wins necessary to reach 40 WAR.

If they signed Ben Sheets to this 3-yr deal at fair market value, which would be an AAV of $14.7 mil for three seasons, they would have 38 wins above replacement amongst 17-19 players, and about $3 mil left to spend. Because they are getting such great deals on Braun (an astrounding $149,000 per win), Hardy, Weeks, and Hart, they can use the money saved in those areas to afford the inevitable jump in arbitration for the likes of Prince as well as other potential free agent targets.

Based on the projections for those returning, the Brewers appear to be in a solid position for 2009, and with about $14-17 mil remaining, should be able to increase from 35 wins to 40 wins above replacement, bringing their projection to 90 wins. Signing Sheets would be nice, but doing so may necessitate the usage of prospects to fill out the roster, who, could add quality for the league minimum, but might not be very useful.

Moving Forward
The next step in these types of analyses is to evaluate the remaining money in terms of the spots that need to be filled. With their $17 mil remaining outside of those returning, it is quite different in signing starters as opposed to a LOOGY, long reliever, and backup infielder. So, Brewers faithful on the interwebs, what are the positions that are not returning next season? And, on top of that, would you prefer to sign Sheets to his fair market value and fill the remaining spots with farm players, or invest that $14-17 mil elsewhere to fill out the roster, all the while keeping the goal of adding at least 5 WAR to the team?

In case anybody out there has failed to notice, I have been particularly obsessed this month with WPA and interesting situations involving the metric. We have explored the ten best offensive plays of the season via shifts in win expectancy, the ten best pitching performances via single-game WPA, as well as instances when hitters and pitchers either exceed +1.0 WPA, or fall below -1.0, in a game. When discussing hitters who have been worth one or more wins in a single game, our focus was on Kurt Suzuki and Cody Ross, both of whom accomplished the rare feat in 2008.

There are, however, 35 other players who have done so at one point in their career, and it just felt natural to share these players and their great games. From 1990-2008, 19 players were so great in a single game that they actually contributed more than one win to their team. First, here are the players from the Y2K era:

6/20/08   Kurt Suzuki      1.002 WPA   4-5, 1B, 2 2B, HR, 5 RBI
6/7/08    Cody Ross        1.133 WPA   2-4, 1B, HR, BB, SB, 3 RBI
6/29/07   Mark Loretta     1.002 WPA   2-3, 1B, HR, BB, 2 RBI
9/7/05    Ryan Langerhans  1.115 WPA   3-4, 2 1B, 2B, BB, 3 RBI
6/11/04   Todd Helton      1.071 WPA   4-5, 1B, 2 2B, HR, 5 RBI
8/24/03   Brandon Inge     1.032 WPA   3-5, 2 1B, HR, SB, 3 RBI
8/21/00   Brian Daubach    1.273 WPA   3-5, 2 1B, HR, 4 RBI
5/10/00   Midre Cummings   1.023 WPA   1-2, HR, 3 RBI

Not exactly your standard list of all stars, eh? Sure, Cody Ross and Brandon Inge have power, and Mark Loretta has been solid offensively his whole career, but outside of Todd Helton this list is not all that impressive. And yet, these eight hitters put together arguably the top single-game performances of the whole decade. To top things off, Daubach’s game in late August of 2000 was over one-tenth of a win better than the next closest player. Did Brian Daubach really have the best game of the decade?

Now, let’s take a trip back to the 1990s, when Alanis Morisette and Hootie and the Blowfish hogged the radio, South Park began its first season, pogs were popular, and people knew who Jon Nunnally was:

4/8/99    Raul Mondesi     1.055 WPA   4-5, 2 1B, 2 HR, BB, 6 RBI
6/13/98   Travis Lee       1.036 WPA   3-5, 1B, 2 HR, 5 RBI
6/10/98   Dante Bichette   1.074 WPA   4-6, 1B, 2B, 3B, HR, 5 RBI
9/10/96   Steve Finley     1.063 WPA   4-5, 2 1B, 2B, HR, SB, 3 RBI
9/2/96    Mike Greenwell   1.029 WPA   4-5, 1B, 2B, 2 HR, 9 RBI
8/24/96   Fred McGriff     1.093 WPA   5-5, 2 1B, 2B, 2 HR, 4 RBI
6/11/94   James Mouton     1.005 WPA   2-4, 2 1B, BB, 2 RBI, 3 SB
8/12/91   Barry Bonds      1.103 WPA   2-4, 2 HR, BB, SB, 4 RBI
5/10/91   Roberto Alomar   1.042 WPA   3-4, 1B, 2 HR, 2 BB, 2 RBI
4/16/91   Dave Henderson   1.082 WPA   5-6, 2 1B, 2 2B, HR, 5 RBI
6/23/90   Dwight Evans     1.147 WPA   3-5, 1B, 2 HR, 3 RBI

Okay, so yes, Dante Bichette’s game was a cycle, let’s clear that up first. And, yes, Mike Greenwell knocked in nine runs on that fateful September 1996 day. Fred McGriff is the only player not to be retired. Also, Dwight Evans, whose game actually led all 1990-1999ers, appears to be the least interesting, at least relative to the stats posted next to the WPA. Which brings me to the next point: Evans’ WPA of 1.147 is 0.126 wins, behind Daubach’s 1.273. Did Brian Daubach really have the best game offensively from 1990-2008!?

Via WPA, which counts certain plate appearances as worth more than others, due to the clutchiness factor built in, it appears so: Brian Daubach had the best offensive game relative to shifts in win expectancy over the last nineteen seasons.

This past weekend, while out watching an absolutely dreadful Eagles game, I found myself explaining WPA and WPA/LI to some friends. They were curious about the site, but perhaps embarrassed to express their lack of knowledge with regards to certain areas. I explained that WPA is basically, as Studes calls it, the story stat: it tracks the positive and negative shifts in win expectancy over the course of a game, and accumulates these measurements for the entire season. The single-game part of the explanation piqued their interest moreso than the overall seasonal total.

“So, in theory, could someone be worth more then one win in a single game?” Bill asked.

Sure, I responded, though as we saw earlier this morning, the instances of such an event are so few and far between that it is pretty remarkable when someone can accomplish such a feat. My friend Ryan then chimed in:

“Could it go the other way, too? Like, could someone technically blow more than one game’s worth of games in a single game?”

I had never really thought about it like that, but I didn’t see why not, given that it is merely the opposite of the aforementioned scenario. I assumed that these instances would also be few and far between, but still existant. Luckily, when David sent me the information regarding players recording a WPA of 1.0+ in a single game, from 1974-2008, he also sent along pitchers who have recorded a WPA of -1.0 or lower in an individual outing.

There have been 26 games over the past 35 years during which a pitcher has cost his team more than one win in a single game; interestingly enough, there have only been 13 games in this same span wherein a pitcher’s WPA met or exceeded +1.0. It’s much easier to be awful.

Just a small number of these games have even taken place recently, as well, with just four occurring between 2000-2007. No pitcher exceeded a +1.0 WPA or fell below a -1.0 WPA in 2008. None at all. The most recent terrible outing took place on June 1, 2007, when Todd Jones of the Tigers blew a save on the road against the Indians. In his three appearances prior to the June 1st outing, Jones had given up five runs in 2.1 innings, without a strikeout, raising his ERA from 2.37 to 4.22 in the process. Clearly, with a WPA below -1.0, things did not improve against the Indians.

Jones would face 12 batters in his one inning of work of June 1, 2007, surrendering seven hits and two walks, en route to five runs and a -1.01 WPA. His ERA skyrocketed from 4.22 to 6.04.

The next most recent game on our list took place almost five years to the day before Jones’ fateful outing. On June 5, 2002, Hideki Irabu made the fifth worst outing since 1974. In his final season, then with the Rangers, Irabu entered in the ninth inning, attempting to preserve the win for Ismael Valdez—who had a stellar game. The Rangers led the Angels, 4-2, but Irabu wanted to make things interesting. After retiring Garret Anderson on three pitches, he gave up back-to-back home runs to Brad Fullmer and Tim Salmon, throwing just one pitch to each hitter.

The game was now tied at four, but the Rangers did score in the top of the tenth to go ahead by a run. Now, Irabu was in line for the blown save win. Adam Kennedy led off the bottom of the tenth with a double, and moved to third on a David Eckstein single. With two on and nobody out, Darin Erstad grounded out, scoring Kennedy in the process. Once again, the game was tied, but this time, the Angels still had a runner on base. Four pitches later, Troy Glaus knocked the ball into the stands to give the Halos a 7-5 victory, and Irabu a -1.21 WPA for his efforts.

The other two games this decade took place in May, and July 2000, respectively. The first saw Jason Isringhausen give up four runs on five hits in 1.1 innings, while the second involved Jeff Brantley getting tagged for three runs on four hits in a mere one-third of an inning. Both outings resulted in a -1.09 WPA. Tomorrow we will take a look at like games that took place prior to 2000, but now you have an interesting trivia question moving forward, in that the game in 2000 or later that cost his team the most in terms of win expectancy involved Hideki Irabu on June 5, 2002.

As many of you now know, last week we unveiled some tremendous new metrics. Available on individual player pages as well as the leaderboards, you now have access to plate discipline metrics for pitchers and pitch type statistics for hitters. The former includes information along the lines of how often a pitcher induced a swing out of the zone, in the zone, as well as his percentage of first-pitch strikes. The latter includes the percentages, and velocities, of pitches seen for hitters, as well as his percentage of first-pitch strikes seen.

I wrote a bit of an introduction to these new statistics last week, and David has written several glossary-type entries as well. This is the type of information that has piqued my interest for a long, long time, and it now adds another dimension to evaluations. For instance, did you know that Johan Santana posted an O-Swing % (percentage of pitches out of the zone that batters swung at) of 30.1 in 2005 and 2006, which decreased to 28.2% in 2007, and 26.8% this past season?

Using the new statistics, I decided to run some correlations to see if certain statistics held strong relationships to each other. First, here are the results for correlations run between the percentage of first-pitch strikes and six prominent evaluative statistics:

F-Strike %

K/9:    0.194
BB/9:  -0.719
WHIP:  -0.515
BABIP:  0.096
ERA:   -0.31
FIP:   -0.406

The results here are not that shocking, or at least they should not be. Getting ahead of the hitter is generally considered key for the pitcher. Doing so, in theory, should correlate quite strongly to any metric involving walks. As we can see, there is a very strong relationship between the percentage of first-pitch strikes and the walks per nine innings issued by pitchers. The relationship loses a bit of its strength when hits allowed are added to the equation in the form of WHIP, but the -0.719 correlation between F-Strike% and BB/9 is actually the strongest of any that I ran. Here are the results for O-Swing % and the same six evaluative metrics:

O-Swing %

K/9:     0.281
BB/9:   -0.493
WHIP:   -0.462
BABIP:   0.036
ERA:    -0.362
FIP:    -0.428

Here, the results are a bit different. Nothing is incredibly strong or on the same wavelength of strength as the FStrike-BB/9, but we have a few relationships of moderate strength. What exactly is O-Swing? It is the percentage of pitches that a pitcher threw out of the zone, that a hitter swung at. With this in mind, we might initially expect that pitchers with the highest percentages in this area would strike more batters out, walk less, and therefore be very effective in the ERA and FIP department. One thing to keep in mind, though, is the percentage of pitches that these pitchers throw in and out of the zone.

Jake Peavy and Barry Zito, for instance, were amongst the bottom in terms of percentage of pitches thrown in the zone, at around 47%. However, Peavy induced many more swings on these pitches than Zito, which is a big reason for the difference between the two, since their percentages of pitches in and out of the zone were virtually identical. When we have pitchers with different percentages in the mix, as is the case in the correlations using O-Swing, the results should not be as concrete. Overall, the strongest relationship here also involves BB/9, as the idea goes back to the Peavy/Zito example: Peavy gets swings and outs on pitches out of the zone, Zito does not. The higher the percentage is of swings out of the zone, the better the chance is that the BB/9 will be lower.

Lastly, Z-Swing%, which is still a bit curious. For instance, does a pitcher want a higher or lower percentage here? I would venture a guess that a lower percentage would be better, as the pitch is already in the zone and therefore very likely to be called a strike. A hitter failing to swing will take a called strike. It probably is not as important as FStrike or O-Swing, but here are the correlations:

Z-Swing %

K/9:   -0.067
BB/9:  -0.014
WHIP:  -0.037
BABIP: -0.150
ERA:   -0.027
FIP:    0.087

Well, I guess it really doesn’t matter for pitchers, as the percentage of swings induced on pitches in the strike zone does not share anything close to a strong relationship with any of the above six metrics. Interestingly enough, the highest correlation for Z-Swing involved BABIP, which was the lowest for F-Strike and O-Swing. The -0.150 isn’t significant by any means, though, so nothing should be taken away by that. At the very least, these results show what we would generally expect: the more first-pitch strikes, the lower the rate of walks or vice versa, and inducing swings out of the zone can result in better rate and run prevention stats.

With the advent of the calendar year data here on the site I have gotten a few questions regarding what constitutes “good” win probability statistics in the various time parameters. One question in particular piqued my interest: How do the context-neutral wins look across the position spectrum? The reader essentially wanted to know how, say, Brian McCann’s WPA/LI over the last two calendar years stacked up not just to all other offensive players but all other catchers. Not only would something like this help show which players’ context-neutral contributions were above- or below-average but it would allow a look at how the averages change from position to position.

Looking at the last two calendar years, with anyone amassing 450+ plate appearances (to use a qualifier but allow for mid-season callups), here are the positional averages with the top player(s) at each:

C: -0.86 WPA/LI, Russell Martin, 3.62 WPA/LI
1B: 2.17 WPA/LI, Albert Pujols, 10.77 WPA/LI and Lance Berkman, 10.65 WPA/LI
2B: -0.10 WPA/LI, Chase Utley, 8.77 WPA/LI
3B: 1.03 WPA/LI, Chipper Jones, 10.26 WPA/LI and Alex Rodriguez, 9.33 WPA/LI
SS: -0.65 WPA/LI, Hanley Ramirez, 6.75 WPA/LI
OF: 0.95 WPA/LI, Matt Holliday, 9.21 WPA/LI
SP: 1.10 WPA/LI, Roy Halladay, 6.90 WPA/LI and C.C. Sabathia, 6.06 WPA/LI

The outfielders on the leaderboards here are lumped together rather than separated by left, center, or right, so their results may shift a bit when properly divided. I also did not use relievers since there are all different types of them—closers, setup men, long relievers, etc—and I don’t much like comparing one set to another out of their element.

These overall results will change as the season goes on as well since calendar years implies a duration spanning yesterday to the same day two years ago; since we are currently in the middle of the season this is not a concrete look at the WPA/LI from concluded years, which I’ll get to sometime later this week. The scary aspect of these numbers is that, of those meeting my previously established cutoff point, any Catcher, Second Baseman, or Shortstop that has a 0.00 WPA/LI over the last two calendar years—meaning their efforts ended up cancelling each other out to the point of zero contribution—is above average. Essentially, someone at these positions contributing, on average, no context-neutral wins, is above average. For now, though, you can see that the averages supply, at the very least, the general ranges for where the benchmarks should be set.

A conundrum is loosely defined as anything that puzzles… so it makes perfect sense to use the term when describing the anomaly present in the ERA and FIPs of both Carlos Zambrano and Jeremy Bonderman. We’ve written about pitchers either outperforming their FIP or failing to live up to it plenty of times here, but, in probing the last three calendar years feature recently instituted on the leaders page, it appears that things tend to even out a bit. Except, of course, with regards to Zambrano and Bonderman.

Sixty starting pitchers qualified for inclusion over the last three calendar years and they produced the following averages:

ERA-FIP: 0.12
BABIP: .303
LOB: 71.8%
K/BB: 2.48
HR/9: 0.98

One standard deviation of the ERA-FIP is 0.28, meaning we can expect about 2/3 of the data to fall within the -0.16 to 0.40 range; additionally, 95% of the data can be expected to fall within the -0.44 to 0.68 range. Of the group of sixty pitchers, just two fell beyond the 95% confidence interval: Carlos Zambrano at -0.53 and Jeremy Bonderman at 0.83.

Now, one potential reason that someone like Zambrano consistently posts better ERAs than his FIP would suggest could deal with his BABIP: the average BABIP of this group in this span is .303 and Zambrano comes in at .273, a full thirty points lower. On the other end of the spectrum, Bonderman comes in at .325, over twenty points higher. In fact, when looking at the eighteen pitchers who fell beyond one standard deviation of the ERA-FIP mean, the nine higher than 0.40 ranged from .297-.332 in BABIP while those below -0.16 ranged between .269-.309.

I actually discovered whilst writing this post that a question regarding Zambrano outperforming his FIP was posed in the Inside the Book mailbag, to which MGL mentioned the possibility of him posting a lower than average BABIP after concluding that it is definitely possible for certain pitchers to post certain types of BABIPs. This is definitely the case. As MGL also noted in the mailbag, “FIP is a very good at eliminating the noise in BABIP, which allows us to get a better estimate of a pitcher’s run prevention skill, in the short run. In the long run, ERA, RA or ERC is MUCH better because it captures the differences in BABIP skill among pitchers, as well as the other things I mentioned above that contribute to a pitcher’s run prevention skill but are not addressed at all in FIP (like WP rate).”

So, one reason these two guys are constantly posting ERAs much better or worse than their FIP would suggest could be that they have posted above or below average BABIPs with enough regularity to show they have some type of control over it; in that regard, their ERA would be a better indicator of run prevention. Then again, they might not have control over their BABIP and this could all even out, but it would seem that this is a very likely cause at this juncture.

Much of my work this week has focused on the ‘Clutch’ statistic kept here, attempting to shed light or help the confusion surrounding its meaning and usage to dissipate. A great discussion took place in the comments section at my post ‘All About Clutch’ wherein it was suggested that the best hitters in the league will struggle to post high clutch scores because, essentially, they would be so high up the performance chart that there would be no higher ground to which their games could be raised. The inverse would then be true for poorer hitters; since their games were so low much more room exists for game-raising performance.

The major confusion stemmed from the fact that a player with a .333 BA in situations with a high leverage index could be less clutch than one with a .225 BA in the same situations. The way the clutch statistic works is that it measures a player against himself, comparing production to what that production would be in a context-neutral environment. Clearly, I would rather have the .333 guy up to bat in a crucial situation and, because of that, heads begin to spin when it is realized that the .225 guy could have a higher clutch score because in all others he hit .200; the .333 guy posted the same BA in all situations, therefore failing to raise his game.

With this in mind I decided to do a little digging in order to see if this generally holds true. I took the qualifying major league players from 2000-2007, first found the average WPA/LI, and then calculated the average clutch score for those with above average WPA/LI as well as the average clutch score for those with below average WPA/LI. Keep in mind that, in the results below, BA refers to the average clutch for below average WPA/LI with AA meaning the same for above average:

2000: 1.15 WPA/LI, -0.10 BA, 0.07 AA
2001: 1.39 WPA/LI, 0.05 BA, -0.10 AA
2002: 1.38 WPA/LI, -0.02 BA, -0.19 AA
2003: 1.15 WPA/LI, 0.03 BA, -0.32 AA
2004: 1.20 WPA/LI, -0.06 BA, -0.25 AA
2005: 1.15 WPA/LI, 0.01 BA, -0.27 AA
2006: 1.07 WPA/LI, 0.22 BA, -0.13 AA
2007: 0.98 WPA/LI, 0.03 BA, -0.14 AA

As you can see, other than in 2000 and 2007, the average clutch score for those with below average WPA/LI was much better than their above average colleagues. Not to say that their clutch scores were earth-shatteringly spectacular, but, rather just much higher and more indicative of game-raising performance. Deciding to go a little deeper, I looked at the top and bottom 10% in each year to see if the results differed:

2000: 0.06 BA, -0.25 AA
2001: 0.03 BA, -0.54 AA
2002: 0.05 BA, -0.87 AA
2003: 0.02 BA, -0.39 AA
2004: -0.20 BA, -0.11 AA
2005: -0.01 BA, -0.46 AA
2006: 0.16 BA, 0.21 AA
2007: 0.34 BA, -0.27 AA

Here we get very similar results; those in the bottom 10% of WPA/LI generally post much higher clutch scores than those at the top. 2004 and 2006 are the exceptions to this “rule” but even they do not differ too heavily; they actually come within ten points of each other whereas every other year is vastly different in the average clutch scores.

Based on these results it would seem that, yes, the players with below average performance are more likely to post higher clutch scores because they have more room to work with, so to speak. I would still rather take, with much confidence, those in the top 10% of WPA/LI in crucial situations, even though the clutch statistic, in its current state, will debit their performance for having nowhere to go really but down.

Now, to clarify the above paragraph, after some tests, there is no correlation between WPA/LI and Clutch, meaning that it is not a concrete rule that all good players will post lower clutch scores and vice versa. From these results, though, it does seem that those with a higher WPA/LI have more opportunity to post lower clutch scores.

The end of each season brings with it a few certainties: eight teams make the playoffs, one team wins the world series, and we are likely to argue or debate about which player’s performance merits the Most Valuable Player award. Some years house less debates than others but the award’s definition is so ambiguous that there are usually a few players that meet the loose “criteria.” By definition, the MVP award was spawned from the idea back in 1922 to honor the player “who is of greatest all-round service to his club and credit to the sport during each season; to recognize and reward uncommon skill and ability when exercised by a player for the best interests of his team, and to perpetuate his memory.”

Now, in 21st century language, this translates to the player who was most valuable to his team; the player who, if removed from his team, would hinder the success of the team the most; the player the team cannot live without. From a statistical standpoint this would seem to refer to which player contributed the most wins to his team. Luckily, we have a statistic for that here, known none other as WPA.

I decided to look at the win probability statistics for all years currently on Fangraphs (1974-2007) in order to see if the definition of MVP has held true, as well as see the average total and rank for a few of these statistics. The stats in question are WPA, WPA/LI, and Clutch. WPA/LI refers to context-neutral wins and so the different game states comprising plate appearances are not taken into account. Clutch, which I will discuss a bit more in-depth later tonight, measures a player’s performance in high leverage situations against his performance in all others.

Using just the National League for now, I recorded the WPA, WPA/LI, and Clutch, as well as the league ranks, for all MVPs from 1974-2007. The only exceptions were Chipper Jones in 1999, since we don’t currently have that year recorded, and Willie Stargell’s co-award in 1979; according to the league leaders page he didn’t even qualify that year. After calculating the average scores and ranks, here are the results:

WPA: 6.10, Rank: 3.88
WPA/LI: 6.11, Rank: 3.48
Clutch: -0.15, Rank: 19.69

A few things initially stand out. First, the average WPA and WPA/LI are virtually identical. Second, the average rank for MVPs in these categories is between 3rd and 4th. Lastly, the average clutch score is negative.

Of the 33 NL MVPs recorded, 14 finished #1 in WPA; 15 were #1 in WPA/LI; and nobody finished #1 in clutch. In fact, just 3 of the 33 finished in the top ten, the highest being Steve Garvey’s second place rank in 1974 (the other two were Kirk Gibson as #8 in 1988 and Bonds as #6 in 2004). So, despite the hoopla surrounding clutch ability prevalent in today’s mainstream media, it has not necessarily translated into MVP success.

Now, of the 17 players who won the award while posting negative clutch scores, 13 finished 1st-4th in WPA while finishing 1st or 2nd in WPA/LI. The only negative clutch scores that did not were the following players, with their WPA and WPA/LI ranks in parenthesis:

1987: Andre Dawson (19,11)
1991: Terry Pendleton (9,7)
2000: Jeff Kent (7,7)
2005: Albert Pujols (5,2)

Of those with positive clutch scores, 7 of 16 finished 5th or lower in WPA, 6 of 16 finished 5th or lower in WPA/LI, and just 3/16 were in the top ten in clutch.

The highest WPA in this span belongs to (guess who?) Barry Bonds, with a 12.63 in 2004. In fact, from 2001-2004, Bonds averaged 10.79 wins contributed. All four of those seasons ranked in the top four, with Ryan Howard’s 8.10 in 2006 being the only other above eight wins. The lowest two WPA scores came with Dawson’s 1987 season (2.84) and Jimmy Rollins last year with a 2.69. The highest WPA/LI totals were Barry Bonds 2001-2004 and fifth place happened to be Bonds in 1993. Again, the lowest belonged to Jimmy Rollins.

It appears that clutch has not factored into NL MVP voting since at least 1974 and that those with great all around numbers/win contributions have been more than capable of winning the award while seeing a decline in their performance during high leverage situations. I tried to see if anyone this year matched up with the average ranks but the results were not too strong. Lance Berkman is currently 1st in WPA, 1st in WPA/LI, and 15th in clutch, which was the closest. When we get closer to the end of the season it should be interesting to see which players come closest to these averages, if not exceeding them.