THE BOOK--Playing The Percentages In Baseball

Beltran?  He’s the one I found that best fits that path although his 2004 *looks* like it should be worth more at a glance.

Yea Beltran fits the curve, but his 2005 is definitely above replacement level. It might be above average too, considering home park.

(unless I’m confused and these numbers are being presented as “runs above average")

I’m pretty sure that’s only offense and relative to average.  But I could be mistaken.

I’m guessing it’s because the mlb’s offense has been down, but people’s perception of obp and slg hasn’t changed.

Yes, those are runs relative to (above or below) average (a replacement offensive player would be around -18, but that would depend on his defensive position since that would be -18 including his defensive run value).  NL offense is not down much - it is mostly in the AL.  In fact, NL offense is about what it was in 05 and down a little from 06 and 07.  But total offense does not change lwts much anyway, since it is like OPS+ - always relative to league average.

Beltran becuase his HRs are down.

Beltran’s SLG from 2001 through 2007 was .515 (and that includes his difficult 2005 season when his SLG was just .414).  In 2008 his SLG is .453.  League slugging is down about .011 ticks from last season.  Beltran has 189 TBs this season; if he was at 215 TBs he’d be at his 01-07 SLG standard of .515.  That’s 26 TBs difference—six or seven homers worth.  My bet is that if he had six or seven more homers this season he would not be perceived as having a down offensive season.

Beltran is actually a tiny amount up from his 2001-2007 standard in OBP (.364 vs. .362), and a little more so relative to the league, with the league down a little in OBP this season. So Carlos is up a few TOB from his norm.  If you value the OBP component of OPS highly enough compared to SLG (OPS itself of course values them as equal), Beltran’s very small uptick on OBP will largely balance his downtick in SLG.

Of course it is Beltran.  I briefly saw the Lupica article (did not read it as I won’t waste my time reading Lupica articles - he is a certified moron) which was talking about how disappointing an offensive season Beltran was having, as if everyone was in agreement with that.  Then I looked at his offensive lwts so far this year and compared it to his lwts projection going into the season, they were exactly the same, and I thought, “What the hell is he talking about?”

On a related note, let’s say we have a player who posts two full seasons (1200 PA) of .900 and .882 OPS at around ages 28 and 29 or so.

The next year, he posts an OPS of .860.  That is all we know about him.

Is that a somewhat disappointing season or a sub-par one?

An interesting thing about Beltran is Carlos himself keeps talking about how he’s having a tough season—in repeated interviews he’s frequently bemoaning how his timing has been off, etc.  And I keep keep looking at the numbers thinking that he’s being rather too hard on himself.  I think that helps contribute to the consesnsus among the writers that you see implied in the Lupica story. Plus, as always, he’s one of the great defensive center fielders.

I am pretty sure that the “reason” people think he is having a disappointing season is:

1) His garbage stats (HR, BA, RBI) are not that impressive.

2) 2006 still stands out in their minds as a comparison point and almost any season will pale in comparison.  06 was and is extremely unlikely to be even close to his true talent level of course.

3) His OBP is high this year, one of the stats that few casual fans look at very closely.

4) The Mets are floundering so anyone is fair game as a scapegoat.

5) League stats are down this year, so literally, everyone, on the average (if you can say, “everyone, on the average"), looks like they are having an “off” year.

6) For some reason, his OPS is understating his lwts and of course lwts is the more accurate number in terms of theoretical offensive production (OPS tells us nothing that lwts does not tell us, and occasionally OPS significantly under or overstates a player’s context netural offensive performance because of the “flaws” in the stat - which illustrates an important point, BTW, which is that when you use an approximation or proxy for something, even a very goos one, you have to be always vigilant that in a small percentage of cases, that approximation will do a very bad job of what you are trying to represent and you have to look at the “rigorous” method or metric).