“We learn more by looking for the answer to a question and not finding it than we do from learning the answer itself.”
I love a good question almost as much as a good answer. While an answer satisfies the curiosity to be sure, it is frequently the end of inquiry. A question on the other hand spurs the mind and leads us down avenues not originally hinted at. This week I’ll take a time out to address three recent questions, two that have answers and one that remains (at least to me) a bit of an enigma.
Directional Balls in Play
The first question comes from reader Evan Rodwell who asks:
Reading Marc Normandin’s profile of Juan Pierre, the directional ball-in-play data you provided caught my eye. I’ve been wondering about the difference between Ichiro’s good years and bad years, and while there is a clear difference in his GB and FB rates, anecdotally I think he pulls the ball more in his weaker seasons (as Pierre does). I thought I might be able to find that out for sure.
So, I guess I’m asking if that directional ball-in-play data is publicly available.
A year or so ago I developed a Windows application, cleverly titled BIPChart, that can be installed on a PC and used to display just this kind of information. In the last several months Marc had asked for it to be updated for use in his player profiles, as Evan noted.
I’m pleased to report that version 1.7 of the software is now available for download here. It requires the .NET Framework which is already present on most machines but can be downloaded separately from Microsoft if needed.
Once installed you can simply click the icon and begin searching for data on your favorite player using the simple interface shown below.
You’ll notice that the application defaults to viewing data for all left-handed plate appearances in the majors for the years 2003 through 2006. It then breaks down the plate appearances that resulted in batted balls by percentage of grounders, fly balls, popups, and line drives in the diamond on the left. On the four diamonds to the right the various batted ball types are further subdivided directionally into three vectors by associating each batted ball with the position of the fielder. As you can imagine, the left slice includes third base, shortstop, and left field. The center slice includes catcher, pitcher, and centerfield and the right side includes first base, second base, and right field.
To search for an individual hitter, simply begin typing the last name of the hitter in the Player drop down and you’ll see data for each of the 2003 through 2006 seasons (if available) and by side of the plate that the hitter swung from. Switch hitters get two entries for each season.
To get to Evan’s question we can use the tool to compare Ichiro Suzuki‘s groundball vectors for each of the past four seasons as follows.
In Ichiro’s 2004 season in which he collected 262 hits and hit .372, he did indeed pull the ball on the ground less often. That year he hit the ball to the left or center of the field 52.5% of the time as opposed to anywhere from 48.3% (2006) to 49.7% (2005) thereby confirming Evan’s suspicion. Even more revealing. as Evan also noted, is that in pulling the ball less often he was also hitting significantly more ground balls overall. In 2004 63.9% of his batted balls were on the ground whereas in other seasons the percentages range from 51.8% (2006) to 54.4% (2005).
New in version 1.7 is similar data for pitchers. By clicking on the Pitchers radio button you can view the data for all pitchers from 2003 through 2006 broken down by the handedness of the batter faced. For example, you can use the interface to compare how and where left-handed batters hit the ball off of Carlos Zambrano (on the ground to the right side) versus right-handers (more fly balls to the right side) and how that has evolved over the past several years.
I hope you enjoy the tool and find it useful for answering, and perhaps even spawning a few questions.
Outfield Arms Redux
Last week on the new Unfiltered blog here at Baseball Prospectus I posted two items related to John Dewan’s rating of the best and worst outfields in terms of suppressing runner advancement and throwing runners out.
As discussed in the second post, several readers pointed out that the top eight teams were all from the AL and the bottom eight from the NL. This cries out for a systemic difference between the leagues since it is highly unlikely that skill differences alone could explain such a top heavy AL performance and bottom heavy NL one. After positing a few ideas in the post related to risk taking, Run Expectancy, and the general level of play, I let the question lie content that I at least did not have the answer. Readers took up the challenge however and began e-mailing theories.
Shortly after the second post reader Joon Pahk chimed in with this:
Runners are going to be much more aggressive taking the extra base with two outs, which will dramatically change both the advancement percentage and kill rate. I don’t know if this would explain the difference between the NL and AL, but it seems to be an important factor to consider.
Indeed, Joon is correct in that advancement rates vary considerably by the number of outs, hence the reason I use outs as one of the primary axes on which to compute the various baserunning metrics I developed this summer.
Where this might affect the outfield ratings is if the NL saw a higher percentage of its hits occur with two outs, then those runners would more often be in motion and therefore be both more likely to advance, and at the same time less likely to be thrown out as supported by the data.
But beautiful theories are sometimes killed by ugly facts. It turns out that in 2006 non-interleague games, 39% of the singles and doubles with runners on base in the AL occurred with two outs while in the NL it was 38.3%. In other words, even in the lower run scoring environment, NL runners were not enjoying a greater percentage of advancement hits with two outs thereby enabling them to advance more safely.
Perhaps, though, Joon is on the right track. Reader J.P. McIntyre pointed out that it could simply be the case that in the NL runners are put in motion more often regardless of outs (via the hit and run for example) leading to more success in advancement as well as fewer double plays grounded into. Our play by play data set includes this information but unfortunately a quick query revealed that runners moving with the pitch on singles and doubles were no more common in the NL than in the AL.
Although we don’t yet have a solution, there is one other tantalizing tidbit offered by reader Bill Johnson. In looking at the percentages published by Dewan he noticed that in the NL those numbers correlated fairly strongly in a negative direction (r = -0.36) with AEqRA but in the AL the correlation is equally strong but in the positive direction (r = .44). While the latter result at least makes sense logically (teams that allow fewer runs may also be better at preventing runners from advancing), the former seems to defy logic. Perhaps that’s another question that thoughtful readers like you can chew on.
Stealing the Show
As regular readers will recall, we spent a good portion of the summer discussing baserunning in this space. With the season in the books several readers have inquired as to when the baserunning metrics will be published for the 2006 season.
While I’m not ready to reveal the entire enchilada, here are the leaders and trailers in EqSBR, the measure that calculates the run impact from stolen base attempts and pick offs.
Top 20 EqSBR for 2006 Name Team SB PO CS EqSBR Ichiro Suzuki SEA 44 1 2 6.03 Carl Crawford TBA 57 1 9 4.63 Dave Roberts SDN 49 3 6 4.00 Jimmy Rollins PHI 36 0 4 3.99 Corey Patterson BAL 45 0 9 3.25 Chris Duffy PIT 26 2 1 3.24 Orlando Cabrera ANA 27 0 3 2.95 Brian Roberts BAL 36 2 7 2.53 Eric Byrnes ARI 24 2 3 2.15 Derek Jeter NYA 33 0 5 1.90 Brandon Phillips CIN 25 2 2 1.74 Josh Barfield SDN 21 0 5 1.65 Nate McLouth PIT 10 0 1 1.63 Kenny Lofton LAN 31 1 5 1.54 Mark Teahen KCA 10 1 0 1.46 Hanley Ramirez FLO 51 0 15 1.45 Carlos Beltran NYN 17 0 3 1.40 Juan Pierre CHN 58 1 20 1.40 Melvin Mora BAL 11 0 1 1.25 Coco Crisp BOS 22 0 4 1.24
Ichiro easily outpaced the rest of the baseball while Juan Pierre surprisingly came out better than break even despite being caught stealing 20 times and picked-off once.
Bottom 20 EqSBR for 2006 Name Team SB PO CS EqSBR Jamey Carroll COL 10 3 12 -4.97 Bill Hall MIL 7 1 9 -4.19 Alfonso Soriano WAS 41 2 17 -3.88 Jose Bautista PIT 2 3 4 -3.67 Jeff Francoeur ATL 1 1 6 -3.01 Ryan Zimmerman WAS 10 1 8 -2.90 Craig Counsell ARI 15 2 8 -2.86 Alfredo Amezaga FLO 19 1 12 -2.82 Ronny Cedeno CHN 7 0 8 -2.81 Mike Lamb HOU 2 1 4 -2.80 Scott Podsednik CHA 39 3 19 -2.74 Corey Hart MIL 5 1 8 -2.71 Miguel Cabrera FLO 6 0 6 -2.70 Reggie Abercrombie FLO 6 2 5 -2.69 Magglio Ordonez DET 1 0 4 -2.67 Juan Encarnacion SLN 5 0 5 -2.62 Dan Uggla FLO 6 1 6 -2.60 Russell Martin LAN 10 1 5 -2.56 Adam Kennedy ANA 14 0 10 -2.54 Juan Rivera ANA 0 1 4 -2.38
Apparently Clint Hurdle’s plan of putting Jamey Carroll in motion, not to mention Carroll’s tendency to get picked off, was literally a non-starter. And after coming in at the top of the heap in 2005 (4.92 EqSBR), this time Alfonso Soriano
And now of course the team totals.
Team EqSBR for 2006 Team SB PO CS EqSBR SDN 123 5 31 2.76 BAL 116 3 32 1.99 PHI 91 2 25 1.16 NYN 139 8 35 0.61 NYA 131 5 35 -0.05 OAK 60 2 20 -0.14 BOS 51 1 23 -3.15 TEX 51 1 24 -3.17 ARI 72 4 30 -3.41 CLE 55 4 23 -3.77 PIT 67 7 23 -4.00 SFN 55 4 25 -4.17 CIN 114 9 33 -4.96 SEA 103 9 37 -5.09 CHN 116 6 49 -5.60 TBA 132 5 52 -5.66 LAN 125 3 49 -6.99 MIN 98 8 42 -7.27 ATL 51 2 35 -8.12 TOR 63 2 33 -8.22 HOU 79 6 36 -8.33 SLN 55 1 32 -8.42 MIL 66 6 37 -8.47 ANA 135 4 57 -8.83 KCA 61 6 34 -8.95 CHA 91 5 48 -9.29 COL 82 8 50 -12.96 WAS 119 4 62 -15.29 DET 56 11 40 -15.49 FLO 105 9 58 -15.98
Keep Them Coming
In Ring Lardner’s 1920 novel The Young Immigrants the youthful narrator and his father enjoy this exchange.
“Are you lost, daddy?” I asked tenderly.
“Shut up,” he explained.
Never let it be said that this author takes your feedback and questions in a like manner. So keep them coming and thanks to all the readers whose questions, observations, and constructive criticism (and to whom I apologize for sometimes failing to acknowledge) serve to make this column and this site better.
Thank you for reading
This is a free article. If you enjoyed it, consider subscribing to Baseball Prospectus. Subscriptions support ongoing public baseball research and analysis in an increasingly proprietary environment.Subscribe now