I recently had my first opportunity to visit the Giants' gem of a venue,
Pacific Bell Park. As I gazed at the statue of Willie Mays out front, I
recalled what is often said about him -- "the best all-around player
ever" -- and got to thinking about what that actually means and whether it
could be measured.
When we talk about a "complete" player or an "all-around" player, we
typically mean he has no major holes in his game. Presumably, if you
named the categories that were important and measured players in each
category, you would find out in which category the player was weakest. If
you then took all players, and found the player whose weakest category was
stronger than anyone else's weakest category, you'd have one way to crown a
"best all-around player."
I decided to follow this approach and identified several statistical
categories which, while not perfect representations of certain skills, are
decent enough indicators to make it interesting:
I started with the so-called "five tools": hitting, hitting for power,
baserunning, fielding and throwing. I also selected a number of categories
representing other skills, such as on-base percentage, strike-zone judgement, and
differentiating between home-run power and gap power.
While defense is certainly important in judging all-around excellence, we'll
set it aside for now, and focus on offensive metrics, and consider position
and defense more qualitatively as appropriate. The 10 offensive categories
used were:
Batting average (AVG)
On-base percentage (OBP)
Isolated slugging (SLG - AVG)
HR Rate (HR/AB)
Other extra-base hits rate ((2B+3B)/AB)
Walk rate (BB/PA)
Strikeout rate (SO/AB)
Strikeout-to-walk ratio (SO/BB)
Steals per time on first base (SB/(1B+BB))
Stolen base success rate (SB/(SB+CS))
I looked at seasons in which a hitter had at least 400 at-bats, and divided
each of his rates by that of his league average rates to see how far above average he was.
For example, a player who hit .300 when the league hit .260 would be listed
at .300/.260 = 1.1538, or 15.38% above league average.
Years in which caught-stealing and walk data were not available were not
considered, and when hit-by-pitches were not available, they were assumed to
be zero. Nor did I apply any park adjustments. Park factors are fine to use
when you're trying to account for the value of a player's performance in
differing offensive environments, but you'd need a finer granularity of park
factors to try to estimate the impact on specific categories of stats (e.g.,
the park factor for doubles in Fenway Park is quite different from the
park's overall effect on run scoring). To keep the analysis somewhat simple,
we'll use raw rates of production versus league average as our standard.
Each player was rated in the 10 categories listed above, with his weakest
category noted. For example, let's look at Craig Biggio's 1997 season. Note
that the ratios have all been calculated such that higher numbers represent
better performance, even in categories like strikeout rate where higher
strikeout totals are seen as negative.
YR Name AVG OBP ISO HR R23 BB SO BB/K SBA SB%
97 Biggio,C 1.17 1.24 1.30 1.27 1.28 1.28 1.15 1.56 2.67 1.21
Looking at the BB column, we see a ratio of 1.28, meaning that Biggio had a
walk rate 28% better than the NL that year, while his rate of stealing bases
(SBA) was 167% above the league average. His weakest category was strikeout
rate, which was "only" 15% better than average. Still pretty
darn good, but not good enough to crack the list of top all-around seasons.
So, according to this system, who had the best all-around season of all
time? The man whose statue prompted this line of research almost came out on
top, but as it turns out there was someone slightly better.
That distinction belongs to Joe Morgan. Morgan's 1975 MVP-winning campaign
was followed closely by Willie Mays' 1958. Morgan's worst category (batting
average) was 27.25% above league average, edging out Mays's worst category
(OBP) at 26.90%. Morgan hit .327/.466/.508 with 67 steals in 77 attempts,
while playing above-average defense at second base. Mays hit .347/.419/.583
with 31 steals in 37 attempts, falling one home run short of a 30-30 season,
winning a Gold Glove in center field, and was runner-up for MVP to Ernie
Banks.
Both Mays and Morgan clearly fit the bill defensively, playing difficult
positions very well. But if we consider position or defense, we'd have to
throw out the No. 3 candidate, Orlando Cepeda's 1967, as he played first base,
a relatively easy position. Strictly on completeness as an offensive player,
we'll leave him on the list. Joe Morgan's second consecutive MVP season in
1976 was nearly equal to his first, and ranks fourth. Another MVP
season, Rickey Henderson's 1990, rounds out the top five.
Here are the top 25 seasons (worst category is boldfaced):
YR Name AVG OBP ISO HR R23 BB SO BB/K SBA SB%
75 Morgan,J 1.27 1.42 1.62 1.83 1.35 2.28 1.42 3.69 4.48 1.28
58 Mays,W 1.33 1.27 1.65 1.72 1.45 1.32 1.57 2.12 4.42 1.35
67 Cepeda,O 1.31 1.28 1.74 2.22 1.41 1.26 1.29 1.68 1.27 1.42
76 Morgan,J 1.25 1.38 2.41 3.38 1.55 2.26 1.68 4.26 4.07 1.30
90 Henderson 1.25 1.33 1.96 2.45 1.41 1.85 1.35 2.69 4.89 1.32
87 Raines,T 1.27 1.30 1.37 1.23 1.47 1.62 1.79 3.07 2.43 1.28
61 Robinson,F 1.23 1.23 2.01 2.39 1.44 1.33 1.34 1.84 3.22 1.42
65 Aaron,H 1.27 1.21 1.94 2.36 1.57 1.22 1.23 1.51 2.81 1.35
64 Robinson,F 1.21 1.27 2.01 2.33 1.65 1.65 1.42 2.48 2.94 1.36
64 Aaron,H 1.29 1.26 1.55 1.92 1.20 1.35 2.07 2.84 2.62 1.40
91 Larkin,B 1.21 1.18 1.67 1.97 1.34 1.22 1.27 1.57 1.81 1.19
92 Bonds,Ba 1.24 1.44 2.68 3.74 1.66 2.55 1.18 3.49 2.24 1.22
56 Jensen,J 1.21 1.18 1.36 1.35 1.22 1.27 1.86 2.40 1.87 1.36
81 Matthews,G 1.18 1.24 1.38 1.52 1.31 1.65 1.24 2.17 1.24 1.31
97 Walker,L 1.39 1.34 2.40 3.07 1.55 1.32 1.25 1.72 2.08 1.18
54 Robinson,J 1.17 1.22 1.37 1.47 1.28 1.47 2.34 3.63 1.79 1.20
68 Aaron,H 1.18 1.17 2.15 2.95 1.42 1.34 1.69 2.29 3.33 1.40
63 Aaron,H 1.30 1.27 2.26 3.15 1.18 1.45 1.17 1.74 3.22 1.48
62 Mays,W 1.17 1.17 2.35 3.02 1.45 1.30 1.19 1.57 2.03 1.37
93 Molitor,P 1.24 1.18 1.26 1.29 1.19 1.17 1.50 1.75 1.41 1.32
70 White,Roy 1.18 1.19 1.37 1.36 1.36 1.45 1.54 2.32 2.38 1.17
59 Mays,W 1.20 1.17 1.93 2.16 1.67 1.19 1.53 1.84 4.42 1.41
99 Jones,C 1.19 1.28 1.95 2.44 1.28 1.89 1.16 2.39 1.48 1.27
59 Moon,W 1.16 1.21 1.38 1.28 1.37 1.52 1.31 2.08 2.12 1.16
56 Minoso,M 1.21 1.24 1.57 1.51 1.52 1.25 1.88 2.49 2.25 1.15
The only players with multiple seasons among the top 25 are Hank Aaron (4),
Willie Mays (3), Joe Morgan (2), and Frank Robinson (2). So Willie Mays has
a pretty strong claim to his reputation as the best all-around player ever,
although Hammerin' Hank might have something to say about that. The biggest
surprises on the list are Gary Matthews, Wally Moon, and Roy White -- not
exactly players who come to mind as providing all-around excellence.
Who are the best all-around players among current players? Well, it's a
little too early to base a judgment on 2001 statistics, but
we can look at the American League and National League charts from last season, and find just two players who
were above average in every category from the Junior Circuit -- Bobby Higginson and Bernie Williams -- and a surprise "best" from the Senior Circuit.
Name AVG ISO HR R23 BB K/BB SBA SB% AB/K OBP
Higginson,B 1.09 1.42 1.47 1.35 1.14 1.23 1.53 1.21 1.08 1.08
Williams,B 1.11 1.54 1.63 1.34 1.20 1.39 1.40 1.05 1.14 1.12
Ordonez,M 1.14 1.38 1.59 1.05 0.94 1.54 1.83 1.19 1.64 1.07
Valentin,J 0.99 1.30 1.29 1.27 0.97 0.92 2.29 1.31 0.96 0.98
Alomar,R 1.12 0.99 0.91 1.15 0.97 1.29 3.56 1.32 1.33 1.08
Erstad,D 1.29 1.11 1.08 1.12 0.89 1.29 2.15 1.13 1.47 1.17
Kapler,G 1.09 1.02 0.92 1.25 0.89 1.21 1.12 0.97 1.39 1.03
Easley,D 0.94 0.94 0.88 1.05 1.08 1.15 1.64 1.11 1.05 1.00
Anderson,B 0.93 0.98 1.10 0.86 1.56 1.47 1.56 0.93 0.88 1.08
Cruz,J 0.88 1.33 1.50 1.03 1.09 0.91 1.80 1.09 0.83 0.93
Nixon,T 1.00 1.10 0.82 1.37 1.32 1.22 1.06 1.29 0.90 1.06
Rodriguez,A 1.15 1.73 2.16 1.09 1.55 1.36 1.32 1.15 0.82 1.20
Durham,R 1.02 1.01 0.81 1.20 1.11 1.18 2.34 0.96 1.04 1.03
Lofton,K 1.01 0.86 0.81 0.86 1.30 1.81 2.84 1.18 1.35 1.06
Bordick,M 1.08 1.10 1.20 0.99 0.82 0.79 0.97 0.79 0.98 1.00
O'Neill,P 1.03 0.84 0.93 0.77 0.84 0.93 1.52 0.88 1.12 0.96
Everett,C 1.09 1.71 2.00 1.22 0.96 0.76 1.43 1.07 0.78 1.07
Clayton,R 0.88 0.85 0.80 0.85 0.78 0.75 1.54 0.89 0.99 0.86
Jeter,D 1.23 0.84 0.74 0.99 1.05 1.13 1.72 1.23 1.07 1.19
Cameron,M 0.97 1.02 1.02 0.99 1.28 0.97 2.40 1.12 0.73 1.05
Name AVG OBP ISO BB AB/K HR R23 SBA SB% K/BB
Klesko,R 1.07 1.15 1.40 1.59 1.19 1.55 1.22 2.09 1.11 2.00
Beltre,A 1.09 1.05 1.11 1.01 1.25 1.16 1.08 1.22 1.03 1.25
Floyd,C 1.13 1.11 1.38 1.06 1.00 1.55 1.23 2.99 1.29 1.09
Green,S 1.01 1.07 1.23 1.30 0.99 1.16 1.35 2.04 1.20 1.33
Abreu,B 1.19 1.22 1.43 1.51 0.97 1.28 1.55 2.11 1.13 1.54
Finley,S 1.05 1.06 1.59 1.08 1.21 1.92 1.02 1.25 0.97 1.33
Jones,C 1.17 1.18 1.54 1.43 1.77 1.84 1.16 1.08 0.97 2.64
Bonds,B 1.15 1.29 2.30 1.98 1.22 3.01 1.15 0.93 1.14 2.71
Rolen,S 1.12 1.08 1.52 0.97 0.95 1.59 1.35 0.94 1.29 0.92
Hidalgo,R 1.18 1.15 1.95 0.89 0.99 2.33 1.39 1.42 0.99 0.91
Hammonds,J 1.26 1.16 1.17 0.89 1.07 1.30 0.98 1.44 0.97 0.94
Renteria,E 1.04 1.01 0.88 1.02 1.43 0.84 1.01 1.91 0.90 1.46
Berkman,L 1.12 1.14 1.59 1.38 0.95 1.76 1.41 0.84 1.09 1.37
Jones,A 1.14 1.07 1.43 0.83 1.28 1.62 1.10 1.81 1.13 1.05
Kent,J 1.26 1.24 1.58 1.33 1.07 1.66 1.40 0.91 0.83 1.50
Larkin,B 1.18 1.14 1.05 1.11 2.50 0.82 1.35 1.67 1.02 2.76
Vander Wal 1.13 1.20 1.59 1.61 0.82 1.85 1.30 1.27 1.23 1.39
Gant,R 0.95 0.95 1.41 0.97 0.92 1.72 0.90 0.91 0.81 0.88
Lansing,M 0.97 0.92 0.98 0.80 1.46 0.89 0.94 1.32 1.16 1.13
Drew,JD 1.11 1.17 1.11 1.43 0.80 1.31 0.80 1.75 0.95 1.21
Bell,D 1.00 1.02 0.96 1.08 0.85 0.97 1.01 0.77 0.97 0.93
Boone,B 0.94 0.95 1.03 0.98 0.93 1.21 0.74 0.97 0.97 0.92
Sanders,R 0.87 0.88 1.03 0.88 0.85 0.96 1.21 4.28 1.22 0.73
Jordan,B 0.99 0.94 0.95 0.73 1.19 1.03 0.91 1.25 1.21 0.85
Walker,L 1.16 1.20 1.19 1.27 1.53 0.85 1.53 0.73 0.73 2.05
Gutierrez,R 1.04 1.10 0.75 1.29 1.51 0.72 0.80 0.78 1.16 2.03
Being an "all-around" player, as we've defined it, doesn't necessarily make
you great. You can be slightly above average at everything, and be a
slightly-above-average player, but it's not a sure path to Cooperstown.
Higginson leading the list is a mild surprise, but most fans would agree
that Bernie Williams is a pretty complete ballplayer. The other names at the
top of the list (Magglio Ordonez, Jose Valentin, Roberto Alomar) were all
solid producers as well.
We are not measuring value. Nobody would argue that Damion Easley and Brady
Anderson are superior players to Alex Rodriguez. A-Rod's worst category,
relative to league is strikeout rate, but that category, by itself, isn't
nearly as important as home-run rate (Easley's worst) or gap power
(Anderson's worst). And A-Rod is so dominant in other areas (isolated
slugging, HR rate, walk rate) that his overall value as a player towers over
some of the others.
What this method does is call attention to players with no major flaws in
their game, and help us think more clearly about what it means to be an
"all-around" player -- like the great Willie Mays.
The team of writers from the Baseball Prospectus (tm) will be writing twice a week for ESPN.com during the baseball season. You can check out more of their work at their web site at baseballprospectus.com. Keith Woolner can be reached at kwoolner@baseballprospectus.com
| |
|
