Youngsters just as reliable as veterans


We're barely a week into October, and already we've heard enough about
postseason experience to last us a lifetime. Type "postseason
experience" or "playoff experience" into your favorite news search
engine, and you'll get back hundreds of recent articles, most of them
reflecting the importance attached to it by baseball insiders.

Nowhere has it been more clear than in the Giants-Marlins series.
When the Giants were named as overwhelming favorites to win, many of
the prognosticators cited the Marlins' lack of October playing time as
one of the reasons. When Giants manager Felipe Alou decided to start Kirk Rueter in
Pro Player Stadium instead of Pac Bell Park, where he'd been more
successful this year, he identified Reuter's postseason experience --
and Sidney Ponson's lack of it -- as one of his reasons.

But does it really make a difference? Are postseason novices who excelled
playing in front of 30,000 fans and a small TV audience in August
going to freeze up now that they're playing in front of 45,000 fans
and a slightly larger TV audience in October? It doesn't seem likely,
but we can look at the numbers to eliminate some of the guesswork.

On the face of it, the way to answer this question would be to look at
postseason series from the past and see if teams with lots of
postseason experience tend to beat teams with little. But looking at
the question that way really won't tell us anything, because of the
whole correlation-does-not-imply-causation thing.

The Yankees juggernauts of the recent and distant past, for example, had lots of
postseason experience, and they won lots of postseason series. That
doesn't mean the experience was causing the success. It's just as
likely -- in fact, far more likely -- that the Yankees of those eras
simply had better teams than anyone else, and their superiority was
causing the success and the experience.

If we want to isolate the experience factor in postseason performance,
it's better to look at individual players. We want to know if players
turn their postseason performance up a notch or two once they've
gotten some October games under their belts.

Let's start with hitters. We looked at all players of the post-war
era who had 30 or more plate appearances in at least two different
postseasons, and compared their hitting performance in their first
postseason with their hitting performance in their last one -- or
their most recent one, for active players. (We picked the last
postseason rather than the second or something else because it
eliminates the issue of how much postseason experience you need; if
more postseason experience is supposed to be better, the last one is
the one where the player should get the most benefit.) We used OPS to
measure hitting performance. Here are the average numbers for the 87
players who qualify:

First Last
Postseason Postseason Diff
.755 .720 -.035

So the average postseason hitter's performance actually declines a
fair amount with more postseason experience. Forty-eight of the 87 (55 percent) were
better in their first postseason than in their last. Average age was
26.7 in the first postseason, and 32.3 in the last one.

If experienced hitters aren't any better in October than inexperienced
ones, maybe pitchers show more of the expected effect. Here are the
corresponding ERA numbers for pitchers with at least nine innings in
at least two different postseasons:

First Last
Postseason Postseason Diff
2.86 3.50 -0.64

For pitchers, the decline with experience was even more pronounced
than with hitters. Pitchers lost over half a run on their ERAs
between their first postseason and their last, on average. As with
hitters, 55 percent of the pitchers declined with experience, and the average
ages were around 27 for the first and 32 for the last.

What might we be missing with this study? One possibility is that by
choosing a player's last postseason as his "experienced" one, the
numbers are skewed by guys who hung on with Yankee teams for eight or
nine Octobers and we're just seeing the results of extreme old age in
their postseason finale. But we can eliminate those players by only
looking at players who played in exactly two postseasons -- then we're
comparing their first one to their second one -- and in that case we
see almost exactly the same level of decline with experience.

Another possibility is that the pitching decline (but not the hitting
one) is caused by rising league offensive levels over time. We can
fix that by limiting the numbers to just the post-1995 expanded
playoff era, where there hasn't been a huge jump in scoring. In that
case, we see an even larger decline among experienced pitchers than
before, albeit in a much smaller sample.

So at the very least, we can say that there doesn't seem to be any
evidence that postseason experience helps you in October. But is
there a good explanation for the declines we found? Part of it could
be age, but I suspect a bigger reason could be the inflated importance
that managers attach to experience. Many managers will bend over
backwards to favor veteran players over younger ones in October,
regardless of what the regular season showed about their abilities.
As a result, experienced players are used more than they should be,
and in situations where they shouldn't be, and the numbers above may
reflect that.

So the lesson here for managers is: go ahead and throw your youngsters
to the wolves. They're no more likely to choke in the playoffs than
your October-seasoned veterans, and all things being equal, they may
actually be a better bet.

You can check out more work from the team of writers of the Baseball Prospectus at baseballprospectus.com.