Earlier this month, the Dodgers signed Andruw Jones. His signing, and the renewed Pierre-bashing which arose, and the question of just how much Pierre had helped or hurt the Dodgers (and his previous employers) arose. I'm astonished that Jones was able to be signed for such a tiny total contract. Remember this time last year... the Bill James Handbook had projected him for 677 HR in his career, and there was open talk about him making $140MM for 7 years, etc. etc. But it's a “what have you done for me lately” market, to be sure. Only time will tell whether 2007 was a fluke, or the start of an ugly decline. For me, it seems as though the Dodgers got the best FA bargain I've seen among big-ticket signings in recent years.
First off, Andruw Jones “fits” the Dodgers roster well. They have some lefty bats already, although the days of having a glut of them are over. They already have another guy with game-changing speed in Rafael Furcal, and until Martin develops a little more power, and/or Kemp develops into the star he's supposed to be, old standby Jeff Kent is their most potent run-producing righty bat. Andruw should change that. And – according to the Fielding Bible – his range is far superior to Pierre's, and comparing their arm strengths is like comparing the light from a laser to that from a lightning bug. So, good signing, LA!
Back to last year's widely questioned signing of Juan Pierre. Pierre is now entering his age-30 season. His career line is .301/.348/.374. And, while people could suggest that he put up about half his stats in run-friendly environments in Colorado and Chicago, he probably benefited less than anyone from the conditions in Coors, and his first two full seasons in Florida were better than his last 2 full seasons in Colorado, anyway. But after that 2nd full season in Florida, Pierre was entering his age-27 season...
Now, much is made of players “slowing down”, and it can be shown that SB totals decline with age, but for ultra-fast, skilled base thieves, they very frequently maintain their levels of stealing, even as they age. The SB totals usually go down slightly as their batting prowess declines and they reach base less often, but their SB/opportunity often stays high if they stay in shape. BUT... that “batting prowess decline” really took a bite out of our hero's game. Since the 2004 season, Pierre hasn't posted an OBP over .331. And, now, in the neo-Moneyball era of fandom for baseball, being crazy fast isn't viewed as much of an asset as it once was. Besides, managers run less these days, so “crazy fast” no longer translates into 80+ SB.
Since the mid-80's, the phrase “one-run strategy” has really fallen out of vogue. “Smallball” is still cited sometimes, but much less frequently than in the past. This makes sense, as the runs/game (RPG) average in the NL (always the more-likely league to use smallball) has gone from slightly over 4.00 RPG to well over 4.50 RPG in that span. In fact, as recently as 1981, NL teams averaged only 3.91 RPG. This climbed to 5.00 RPG in 1999/2000. It's dropped a bit since then, but there's still a much different balance to the game than there was as recently as the 80's. The logic behind one-run strategies is probably similar to the football strategy of kicking a PAT instead of going for 2, even in situations where the team has a higher-than-even chance of making the 2PT conversion...getting something is better than risking the chance of getting nothing, even if the “expected value” is lower. And, if a manager has a lead, he's “in control” of the game... relievers can be brought in for matchups, defensive replacements can be made, etc., etc. If the win expectancies of various situations (historically available at Walk off Balk) are examined, it's clear that there are some situations which warrant even the most conservative managers trading outs for a better chance at 1 run (one-run strategy).
Anyway, one-run strategies are Pierre's forte. Looking at his 3-year OBPs of .326, .330, and .331, it's difficult to see why he's used in the top 2 spots in the order, even if his “reached on error” stats are included. But the thinking is that while his teams score fewer runs, they will more-often score “important” runs, which should help sway the team's ability to score “1 run”, and thus have an impact on “1-run game” wins and losses. Does this balance out? Pierre has batted 1 or 2 virtually his entire career. Here are the “symptoms” that should be expected of using him atop a lineup:
- The teams would score fewer runs than expected by simply looking at BA, OBP, and SLG.
- The teams would do better in 1-run games than in overall results.
- The teams would have a better record than the “Pythagorean Theorem” indicates.
- The teams would have a better 1-run record than the “Pythagorean Theorem” indicates.
Here are the Batting Average, On-base Percentages, and Slugging Percentages for Pierre's teams for his full seasons (and a rudimentary version of Runs Created per 27 based only on BA, OBP, and SLG – listed as “ExpRuns”):
Team
Year
Runs
Gm
BA
OBP
SLG
ExpRuns
Rockies
2001
923
162
.292
.354
.483
955
Rockies
2002
778
162
.274
.337
.423
777
Marlins
2003
751
162
.266
.333
.421
755
Marlins
2004
718
162
.264
.329
.406
718
Marlins
2005
717
162
.272
.339
.409
753
Cubs
2006
716
162
.268
.319
.422
727
Dodgers
2007
735
162
.275
.337
.406
746
That works out to a “run deficiency” of almost 94 runs over his career, or more than 13 runs per season(!) For a rough estimate, that's worth about a win and a half per season that Pierre has cost his teams due to the inefficiencies of lineups with Pierre at the top, with regard to compiling high run totals.
Team
Year
1-Run
Overall
+/-
Rockies
2001
18
28
.391
73
89
.451
-2.7
Rockies
2002
18
19
.486
73
89
.451
1.3
Marlins
2003
30
23
.566
91
71
.562
0.2
Marlins
2004
20
17
.541
83
79
.512
1.0
Marlins
2005
20
23
.465
83
79
.512
-2.0
Cubs
2006
15
26
.366
66
96
.407
-1.7
Dodgers
2007
28
20
.583
82
80
.506
3.7
TOTAL
-0.2
From this table, it seems clear that Pierre's teams have done almost exactly the same in 1-Run games as they have done overall. How about if the 1-Run results are compared to the Pythagorean Theorem W/L records? The Pythagorean Theorem-derived W/L record should be worse than the overall W/L record if the 1-run strategies are turning fewer runs into more wins (than expected by the Pythagorean Theorem).
Team
Year
1-Run
Pyth
+/-
Rockies
2001
18
28
.391
82
80
.506
-5.3
Rockies
2002
18
19
.486
70
92
.432
2.0
Marlins
2003
30
23
.566
87
75
.537
1.5
Marlins
2004
20
17
.541
83
79
.512
1.0
Marlins
2005
20
23
.465
79
83
.488
-1.0
Cubs
2006
15
26
.366
70
92
.432
-2.7
Dodgers
2007
28
20
.583
82
80
.506
3.7
TOTAL
-0.7
Nothing in these results seems to indicate that the Pierre's teams have been especially good at winning 1-run games more than the team should be expected to win, based on their runs scored and allowed totals.
Obviously, some aspects of Pierre's game haven't been treated here, such as his defense and caught-stealing (though the CS are indirectly factored in when comparing ExpRuns based solely on BA, OBP, SLG for the team. But two conclusions can be drawn with some certainty: 1) Teams using Pierre to leadoff score significantly fewer runs than a “typical” team with the same overall BA, OBP, and SLG stats. 2) Teams using Pierre to leadoff don't gain a significant advantage in 1-run games. In short, giving tons of outs to Juan Pierre at the top of an order is damaging to a team's offense. Not really a surprising conclusion, but sometimes the “popular opinion” isn't supported by objective statistical research.
Questions and comments for this article may be submitted to Rob McQuown at
. Past articles for this author can be found under “Staff Writers” at the Baseball Digest Daily site.
