Bill James Likes Jason Kubel, But Not Denard Span

Or:  Marcels loves Denard Span, but not Jason Kubel.

Usually, most of the major projection specialists (Bill James, Zips, Chone, Pecota, Marcels, THT) don't vary too much in terms of how they expect individual players to perform.  This makes sense, since they all have their formulas and calculations for players, positions, player types, that help sculpt career arcs and have an effect on what numbers they expect.  And in the end, they're all pretty accurate and no one system is definitively better than any of the others in a real, meaningful way.

When I was glancing through 2009 projections for Minnesota position players tonight, this is exacty what I came across.  Bill James and Marcels both have their predictions available already, and I wasn't surprised by the similarities.  Joe Mauer is predicted by James to hit .323/.412/.463; .319/.403/.455 by Marcels.  This is just a 17 point difference in OPS.  For Justin Morneau, the difference is eight points.  Delmon Young, 26; Michael Cuddyer, 15; Nick Punto, 16; Alexi Casilla, 23; Mike Redmond, one; Carlos Gomez, another one; Brendan Harris, 7; Brian Buscher, 15.

Then I came across two guys who fell oustide this comfort zone.  That difference for Jason Kubel is a massive 56 points.  Denard Span's margin is even larger, at 83.  Clearly a combination of youth and less experience can lead to differing opinions, like we see with Young and Casilla, but considering the ideas for Gomez's '09 were almost identical there's certainly more to it than age and time served.

What's the diff?  Take a look:

The first question we need to ask is:  What's the difference between how Marcels and James collate their numbers?  Well, we can answer that pretty briefly.  Marcels (aka The Monkey System) uses what Tom M. Tango calls "an over-under", using the last three seasons (the most recent weighted more heavily), regression to the league mean and age as their three guidelines.  Bill James uses a lot of the same guidelines, but at times has been dubbed "optimistic" when people talk about his offensive projections.  He appears to use a bit more personal touch for hitters, but also takes the whole body of work into consideration.

So when we look at Marcels' numbers for Span, knowing that they weigh the last season heavier than the previous two, it's little wonder why they expect him to have such a great season...there wasn't any other stat line to lend to a regression.  As a result, his playing time is the same, his batting average is the same, and while there's mild regression in on-base percentage, OPS points are made up by a little bit of power.  In 2008, Span's OPS was .819; Marcels projects .823.  I think most of us would be pretty happy if Denard had this kind of a season.

This also helps us understand why Kubel's line through Marcels is so "pessimistic".  OPS marks of .665, .785 and .805 over the last three seasons, and a projection of .781; even considering he'll only be 27 in 2009, his bad (2006) and mediocre (2007) performances drag him down from a good 2008.

James, on the other hand, doesn't have as much faith in a player of Span's profile:  no pop, and until this last year, not much history of taking walks.  As a result, he's not losing out on enough hits to really matter, but he is losing out on extra bases and free passes.  There go two better measures of an effective hitter than batting average, slugging and OBP.  In the end it looks like James sees Span as a contact hitter with speed, but without any other weapons Denard is predicted to regress harshly next summer.

With Kubel it's a different story, as after his first full, good season, James has seen enough to believe Jason is on the cusp of turning into the hitter he was on a trajectory to be before the problems with his knees.  James projects improvement across the board, and this has to be the type of season most of us expect from Kubel in '09; more power, increased plate discipline, essentially the kind of season you'd expect from a guy who hit .320/.385/.499 in the minor leagues.

Clearly, I'm hoping Marcels and James each win a battle here.  If Span can have a season like Marcels thinks he'll have, and stay healthy, the leadoff role will be setting the table for the bigger guns all season long and Span will probably score over 100 runs.  And if Kubel can have the season James thinks he'll have, he'll be a good supplemental offensive force hitting fifth or sixth.