A bit of talk in recent days and weeks around here has been centering around the defense of our respective teams. A happy coincidence, then, that The Hardball Times Baseball Annual 2008 focuses several analysis chapters to the subject of defense.
Perhaps the most interesting is Tom Tango’s look at shortstop defense over the past 15 years.
The article focuses on every Red Sox fan’s favorite whipping boy — Derek Jeter. I won’t go too much into the Jeter analysis, which I admit I did enjoy. It pretty clearly shows that Jeter isn’t just overrated on defense — he’s criminally overrated. Even arguments that he’s good at coming in on balls or leaping for line drives fall flat as Tango shows that Jeter makes fewer plays than almost every shortstop in the game when compared by the pitcher on the mound, batter at the plate, baserunner on first, or ballpark.
We’ll look a little at that, but what he doesn’t mention — and I was surprised to see — was how well this look at defense flatters Julio Lugo (Red Sox fans’ second-favorite whipping boy).
First, the method.
For example, based on pitcher on the mound, Tango found the play-made
percentage for Jeter while, say, Roger Clemens was on the mound, then
did the same for all the other shortstops behind Clemens when was on
the mound (John Valentin and Adam Everett were the two major ones). He
then did the same for every pitcher Jeter ever played with (and was on
the field with), and calculated all those pitchers’ non-Jeter totals.
So, in the end, you get a Jeter percentage and an "all others"
percentage with the exact same pitchers on the mound, or with the exact
same hitters at the plate, or with the exact same runner on first, or
in the same ballpark.
Finally, he does the same for every shortstop who started a significant
number of games since 1993. In each case, you get a SS percentage, an
"all-others" percentage and a difference between the two, which Tango
calculates per 4,000 chances (roughly a season’s worth). Whew! Got all
Needless to say, Jeter finishes in the bottom five in every category
but ballpark, which seems to indicate he’s not helped by Yankee
Stadium, as he’s "just" ninth from the bottom in that category, but
otherwise indicates that regardless of the type of ball hit to him, he
just makes — and has made — fewer plays overall than nearly everyone
else who has played shortstop over the past 15 years (ranging,
depending on the situation, from between 10 plays to 38 plays below the
I don’t think this comes as a surprise for many of us. What did
surprise me was that these numbers seem to show Julio Lugo in a much
more positive light than the Plus/Minus numbers that run on these same
principles. Lugo is below average in Plus/Minus and just about average
in other defensive metrics.
By pitcher, Rey Sanches and Adam Everett are the top two, 59 and 57
plays above average, respectively. Other names in the top 10 are not
surprising — Orlando Cabrera, Mark Grudzielanek, Cezar Izturis. But
Julio Lugo ranks eighth — 26 plays better per 4,000 chances than
average. He made plays on 12.8 percent of the balls hit into his zone
(as determined by UZR), while other shortstops behind the same pitchers
made 12.1 percent of the plays. That’s better than Alex Gonzalez
(12.3%, 12.1%, +11), Cal Ripken (12.2 %, 11.9%, +10) and Omar Vizquel
(12.4%, 12.4%, +2).
By batter, it’s the same story. Lugo of course made plays on 12.8
percent of the balls hit into his zone (obviously, that wouldn’t
change, as we’re discussing every single play possible). With the same
batters at the plate, other shortstops made 12.2 percent of the plays.
That’s good for 22 plays made above average — ninth among all
shortstops since 1993 (Everett and Sanchez rank 1-2 again).
By runner? That’s a bit different. Lugo only made plays 12 percent of
the time with a runner on first. With the same runners on first, other
shortstops made 12.3 percent of their plays. Lugo was four plays per
1,300 chances lower than average (the 1,300 chances reflects the lower
number of plays that occur with a runner on first). Why the difference?
I don’t know. Everett is still No. 1, Sanchez is No. 5, Jeter is still
By ballpark, Lugo returns to the top half of the rankings — other
shortstops made 12.4 percent of their plays while in the same ballparks
as Lugo (remember he’s at 12.8 percent). The 15 plays above average per
season is good for 12th. Sanchez and Everett are again 1-2, by the way.
Compare these results with PMR, which placed Lugo sixth from the bottom in 2007 (but sixth from the top in 2006 and similarly in 2005). RZR? Lugo ranked fifth in the AL (just behind Orlando Cabrera) in ’07. OOZ? Fourth. He had eight throwing errors in 2007, but that’s eighth out of 14 shortstops in the AL with at least 750 innings in the field and far better than his terrible 2006 or bottom-three performance in 2005.
It seems Lugo has been consistently underrated as a shortstop — he received no votes in Fielding Bible voting this year, while even Jeter received four — even though there seems to be a growing statistical consensus that he is in fact quite good, or at least above average. Defensive metrices remain incomplete and tend to vary widely, but these numbers are not unconvincing.
Lugo’s rankings in Tango’s study are consistently ahead of Nomar Garciaparra, Orlando
Cabrera (the only one who’s remotely close), Alex Gonzalez and Edgar
Renteria. Is it truly possible that in the revolving door of shortstops
the Sox have had since 2004 the defensive star may actually be the
one currently on the roster? It seems that may actually be the case.