No Mo, Thank Heavens
Why the Angels Will Be Better Without Mo Vaughn -- and Why Statistics Underestimate the Defensive Value of First Basemen
When Mo Vaughn went down for the season, the first thing I did was call my dad, an Angels season-ticket holder for 30-odd years.
"Did you hear the great news?!" I asked.
"Yea, man! We might have a chance next year!"
My Dad's seats have always been within an underhand toss of first base, so we have seen generations worth of clubfoots, from Bob Oliver to Cecil Fielder. None compares to big Mo. He is so excruciatingly slow, he is forced to play two feet from the first-base line in the late innings of close games (and I've still seen ground balls get by before he had a chance to fall down). The Angels have him play behind baserunners so he has a head start on balls hit in the hole; consequently, the team committed a whopping 16 percent of the American League's balks last year (10 total) against 4 pitcher pickoffs, and Mo still gets to fewer balls to his right than any other major league first baseman.
In the Tony Solaita days, we didn't have many fielding statistics to tell us how our naked eyes squared with the record (Rod Carew, for instance, was a deceptive defender, because he never looked like he was running hard, especially for foul pop-ups right in front of us, but he certainly wasn't slow). In the post-Bill James era, we have many more tools, each of which tells us that Mo Vaughn is indeed uniquely awful.
But I think the terrific work done by the sabermetricians of the world has underestimated and miscalculated the defensive significance of first basemen, precisely because they don't measure their most important task (and the thing Mo does worst of all): catching throws.
Hack-work
First, though, let's examine Mo by existing methods. To do so, let's contrast him with the average first baseman, and then with Gold-Glover John Olerud, who, helpfully, had the most equivalent offensive season to Vaughn last year. (According to James's Runs Created formula, which is my favorite method for combining hitting stats, Olerud created 6.16 per 27 outs, while Vaughn had 6.26. Baseball Prospectus' Equivalent Average, another combination stat that factors in ballpark effects, puts Olerud at .292, compared to Vaughn's .287.)
The evidence:
Vaughn led American League first basemen in errors, with 14. Only the Pirates' Kevin Young had more, with 17. The average major league starting first baseman, prorated to Vaughn's 1265 defensive innings, committed 9. Olerud had 5. Vaughn's .990 fielding percentage was worst in the league (Young kicked his way to .986); Olerud was first with .996.
Vaughn threw people out ("assists") less than any first basemen in baseball -- 69 times in 147 games, or once every 18.34 innings. So when you add up each underhanded toss to the pitcher covering first, or force-out at second, or successful relay to the plate, Mo was lucky to chalk up one every two games. The average starter managed one assist every 13.06 innings; Olerud was second in the AL with 10.29.
STATS Inc. has a statistic called "Zone Rating," a sort of numerical observation tool, whereby STATS reporters tally up the number of balls a defender got to, divided by the number he should have gotten to. Sure enough, Mo was the worst by that measurement, too, with .774. The average first baseman converted .845 percent of the balls in his zone into outs; Olerud was second in the AL with .872.
Baseball Prospectus has a stat called "Fielding Runs," which weighs heavily toward range, and also factors in the team's flyball/groundball ratio, how many balls are put into play, where they are hit, how many runners reach first base, etc. A neutral number is zero; good players save positive runs, bad ones yield negatives. Mo Vaughn, by this method, cost the Angels a wretched 18 runs last year, most in the AL (Kevin Young squandered 20 in the NL). Olerud saved 17, best in the league.
A 35-run difference, if true, would be enormous. When tacked on to the offensive stats, it boosts Olerud's Runs Created per 27 outs by a full run per game, from 6.16 to 7.21, which is Jim Thome/Rafael Palmeiro territory. Mo's output shrinks to 5.23, just a notch over his backup Scott Spiezio's RC/27 total of 4.86 (and well below the Angels' other first baseman, Wally Joyner, who posted a 6.44 last season in Atlanta).
Put another way, over 1265 innings, here's how Vaughn's error/assist ratio stacks up.
Vaughn: 14/69
Avg. 1B: 9/97
Olerud: 5/123
Ignoring all the peripheral effects measured by Baseball Prospectus, Mo made five more errors, and 38 fewer plays, than the average starting major league first baseman did over the same number of innings (nine errors and 54 fewer plays than Olerud, if he'd played 94 less innings). Even if we assume that every one of those missed plays had the value of a single, Mo's overall value goes way down. If we subtract those singles from his offensive line, Vaughn's average/on-base percentage/and slugging percentage totals plummet.
Real Mo: .272/.365/.498
Avg. Mo: .202/.307/.428
Ole. Mo: .169/.279/.396
Even without the defensive adjustment, Mo's offensive numbers stand in the middle of a pack that includes Olerud, David Segui, Fred McGriff, Paul Konerko, and Richie Sexson. Now that he no longer has the Green Monster to lob doubles against, Vaughn is nowhere close to the Giambi/Delgado class, let alone Thome/Palmeiro.
Well, you say, maybe the Angels just weren't a groundball staff. Wrong -- Troy Glaus led major league third basemen in range; on the other side of the infield, Adam Kennedy ranked seventh among all second basemen, and even the unspectacular shortstop combo of Benji Gil and Kevin Stocker was above average. Scott Schoeneweis is among the top five groundball pitchers in baseball.
And last year was no defensive fluke for Mo Vaughn -- he's been this bad a long time. Over his career, he has averaged 15 errors per 147 games (147 is a career high), versus only 81 assists (with a career high of 94). Olerud's career high in errors, by comparison, is 10, and he's had at least 104 assists in each of the past four years, for three different teams.
1Bs' unmeasured stat
As bad as all that sounds, I think Vaughn's defense was much worse. That's because none of the first-base defense stats I've found measures the position's basic function -- catching throws.
I've seen about a dozen Angel games a year from 1998 on, after an eight-year gap of living abroad. The biggest three changes to the game I've noticed are the explosion in offense, the increase in game length, and the official scorers' sudden reluctance to give out errors. Sitting in Mo's back pocket, I've watched him miss throw after throw that required him to do things like jump, move off the base and apply a tag, or bring in a short-hop.
If the play would have been close anyway, the scorer usually rules it a hit, but if Mo fails to catch one in the dirt when the runner is more than a step away, the error goes to the thrower, and shows up in the statistics.
And boy, does it ever. Troy Glaus, a marvelous fielder, led the world in third-base errors last year with 33 (the next closest, former catcher Phil Nevin, had 26). Adam Kennedy tied Ron Belliard for the major league lead in errors by second basemen, with 19. Stocker, Gil, and Gary DiSarcina combined for 30, which would have ranked third among all shortstops.
How many of those 82 errors were on throws to Mo Vaughn? I don't know. (If anyone has that kind of data, please e-mail me.) I do know the Angels committed 64 throwing errors in 2000, third most in the American League. Assuming, generously, that all 13 errors committed by their four main outfielders were errant throws, that leaves 51 in the infield.
Two-thirds of the Angels' putouts were at first base, and given the youth and talent at the other positions, I would guess that a much higher percentage than that were throws to Mo Vaughn.
Infield throwing errors are the great ignored fielding statistic for first basemen. If they typically make eight or nine putouts per game, shouldn't we try to measure their effectiveness at catching baseballs? What's the difference between a short-armed, stone-handed 6'1", 268-pound Mo Vaughn, and a long-armed, soft-handed 6'5", 220-pound John Olerud? Look at the infield errors.
With Olerud playing first base, the 1999 New York Mets set major league records for fewest infield errors and unearned runs. When he left, so did the D: Third baseman Robin Ventura committed 9 errors in 160 games in 1999, 17 in 137 last year. Second baseman Edgardo Alfonzo deteriorated from 5/158 to 10/146. Cuban phenom Rey Ordoñez made more errors (6) in 44 games last year than he did in 154 games (4) the year before.
Meanwhile, Olerud's new team suddenly improved their infield defense greatly. Seattle's 36 team throwing errors was fewest in the American League (the average was 51). Alex Rodriguez, the only real constant between 1999 and 2000, committed 14 errors in 129 games pre-Olerud, 10 in 148 throwing to The Helmet.
Nomar Garciaparra made a similar improvement once he stopped throwing to Mo Vaughn, going from 25 errors in 143 games in 1998, to 17/134 in 1999, and 18/136 last year.
It seems to me that first-base defense is the kind of thing the hardcore sabermetrics nuts and Project Scoresheet volunteers could begin to measure a lot better without too much trouble. Who digs out the most throws, and with the highest percentage? Who makes that jump/swipe tag play the most? How many bang-bang plays with off-kilter but catchable throws are scored base hits?
Without more precise data, I'd guess that Mo cost the Angels another 25 outs last year -- 13 scored errors, 12 scored hits -- with his below-average ability to catch baseballs thrown to him. That's arbitrary, but I would argue it's conservative.
So what's another 25 singles off his already-adjusted offensive line?
Real Mo: .272/.365/.498
1st Adj. Mo: .202/.307/.428
2nd Adj. Mo: .161/.272/.388
This may look too ghastly to be true; then again when the Red Sox let Mo go in 1999 they won two more games without him, while the Angels he was supposed to "lead" were 15 games worse, in no small part due to a clubhouse civil war triggered by Vaughn's refusal to participate in an on-field brawl.
Meanwhile, Seattle improved by 12 games with Olerud (and without Ken Griffey), while the Mets he left won three fewer. Olerud's teams have been to the playoffs five times, and won the World Series twice; Mo has lost in the first round twice, (going 0 for 14 in the Divisional Series in 1995 before busting out in 1998).
Bigger than the team
The statistics, even as crude as they are, show that Mo Vaughn costs his team dozens and dozens of outs on defense every year. He's been the worst defensive first baseman in baseball for a decade, so why isn't he DHing?
The answer may be the best reason of all why Mo Vaughn missing the season is such good news: because Mo insists on playing first base, team be damned. Vaughn is the only player who Manager Mike Scioscia, whom I greatly admire, cannot look in the eye and tell what to do. Scioscia knows Vaughn can't play first better than former second baseman Scott Spiezio, who DHed much of last year, but he and general manager Bill Stoneman inherited Mo, Mo's mythology, and Mo's $80 million contract.
For the last two years Scioscia has told Vaughn to lose weight in the offseason, and for the last two years Mo's showed up unapologetically fat. On a team personified by the no-nonsense hustle of Darin Erstad and humble professionalism of Tim Salmon, Vaughn's third-person schtick sticks out like a dislocated thumb, and competes (unsuccessfully) for the soul of the team.
And, even though 36 homers and 117 RBIs do speak for themselves, it is demoralizing to watch a guy who can't make it to second base when he hits it off the wall, can't hit it to the left half of the infield when the third baseman is playing shallow centerfield on the most exaggerated shift in the big leagues, and can't jump more than about five inches to catch a throw from the third baseman.
The Angels have a young pitching staff with pretty good potential. The 50-75 extra outs will come in handy, Glaus' reduced errors should make him a lock on a Gold Glove, and even Wally Joyner can hit better than .161. Break a leg, Mo.
Originally published at Sportsjones.