Smart Stats, Dumb Stats Why OBP and OPS are very hot, and batting average and RBIs are out of style

Measuring the value of a player used to be as easy as reading his
numbers on the scoreboard, in the Sunday morning paper or on the
back of a baseball card. The information age, however, has
revealed the limitations in such traditional stats. Without
venturing into the dizzying acronymic labyrinth of VORP (value
over replacement player), BRARP (batting runs above replacement
position), DIPS (defense independent pitching statistics) and
other advanced specialty stats that could scare off a jet
propulsion engineer, here is a look at what's in vogue--and
what's not--in the vast array of baseball stats.

WHAT'S IN

1. On-base percentage. An official stat only since 1984, OBP
(hits plus walks plus hit by pitches divided by at bats plus
walks plus hit by pitches plus sacrifice flies) is enormously
important because it tracks how often a hitter wins his duel with
the pitcher by not making an out. Contrary to the adage, great
hitters do not fail seven out of 10 times--that's a .300 OBP,
which is poor. Great hitters fail six out of 10 times (a .400
OBP), and extraordinary ones like Barry Bonds, at least for the
past three seasons, fail five out of 10 times.

2. OPS (on-base plus slugging). Adding on-base percentage to
slugging percentage tells the story of how well a hitter gets on
base and how much damage he does with his hits. It reveals how
much better Alex Rodriguez (.996) was last year than fellow
shortstops Edgar Renteria (.874), Nomar Garciaparra (.869), Derek
Jeter (.843) and Miguel Tejada (.808).

3. Strikeouts per nine innings. "Pitching is defense," Red Sox
general manager Theo Epstein says. "You can't separate them." The
outcome of balls put into play often depends on the range and
skills of the fielders or just plain luck (think bloop hits and
bad hops). Strikeouts, though, are defense independent; the
higher the rate of strikeouts, the less that pitcher relies on
defense and luck. A declining trend in strikeouts per nine
innings may be fatal for a pitcher, as was the case last year for
Mets lefthander Tom Glavine.

4. WHIP (walks plus hits allowed per inning pitched). The fewer
base runners allowed, the fewer opportunities for the opponent to
score. The Giants' Jason Schmidt was the only qualifier last year
with a WHIP lower than 1.00 (0.95). Bad news for the Orioles and
the Devil Rays: Six of the top 10 are in the AL East (Boston's
Pedro Martinez and Curt Schilling, Toronto's Roy Halladay and the
Yankees' Mike Mussina, Javier Vazquez and Kevin Brown).

5. Stolen base percentage. It's not how many bases you steal but
how successful you are. Playing for one run early in a game is
overrated in today's power game. The stat-minded A's and Blue
Jays, for instance, ranked 29th and 30th in steals last year
because they know not to risk the opportunity for a big inning.
"If you're stealing at less than a 75 percent success rate,
you're better off never going at all," wrote Joe Sheehan of
Baseball Prospectus. The team with the most steals, Florida
(67%), was below that mark; only three teams exceeded it.

WHAT'S OUT

1. Batting average. Wait a minute. Isn't the player with the
highest batting average considered the batting champion? Isn't
batting average one of the jewels of the prestigious Triple
Crown? Well, yes, and people once listened to music by having a
metal stylus skim across a rotating vinyl disc. The problem with
batting average (hits divided by at bats) is that it reveals
nothing about the type of hits (bunt singles count the same as
grand slams) or how many times a player reached base by walk or
by hit by pitch.

2. Runs batted in. RBIs are heavily dependent on a player's spot
in the lineup, how often his teammates batting in front of him
get on base and how well they run. Any player who hits in the
middle of a lineup and stays healthy should drive in 100 runs.
Half of the 20 AL players with 600 at bats last year did so,
while two others had 99. Carlos Lee of the White Sox and Jay
Gibbons of the Orioles, for instance, accumulated 100 RBIs even
though they made outs at a worse rate than the average major
league player.

3. Runs. Like RBIs, runs scored are strongly influenced by lineup
position and teammates' abilities. Moreover, the stat doesn't
reveal how responsible the player was for creating the run. The
man who scores after reaching base on a fielder's choice, for
instance, gets the same credit as the man who scores after
leading off an inning with a triple.

4. Won-lost record for pitchers. Many a 20-game winner owes his
milestone to generous run support from his teammates and, these
days, a strong bullpen. Andy Pettitte won 21 games for the
Yankees last season with the help of 7.04 runs per nine innings,
the second best support in the majors among ERA qualifiers. New
Yankee Javier Vazquez, who otherwise had better numbers (3.24 ERA
and 241 K's to Pettitte's 4.02 and 180), won only 13 games for
Montreal while getting 3.98 runs per nine innings, which ranked
83rd among 92 qualifiers.

5. Errors. Everyone knows a home run when they see one. An error,
however, or the mere possibility of one, is a Rorschach blot.
Official scoring is dreadfully inconsistent from park to park.
Moreover, errors tell nothing about a player's range or throwing
ability. Former Angels outfielder Brian Downing, for instance,
once held the AL record for consecutive games by an outfielder
without an error. He often was replaced late in games for
defensive purposes. --T.V.

SI.com
More on baseball's stat services plus scores, team schedules,
complete rosters and analysis from Tom Verducci at
si.com/baseball.

COLOR PHOTO: J. PAT CARTER/AP RISK MANAGEMENT Luis Castillo and the Marlins racked up a major-league-best 150 stolen bases last year, but did they run too often?