Sunday, September 16, 2007

 

I Want My Ha-An-Leee

Just a super post from Avi on the current winner of the Josh Beckett/Hanley Ramirez deal. Baseball fans will want to start their Sunday off with this one.

Comments:

A couple of questions:

1. What is VORP? Does it factor in ballpark effects?
2. In regard to that, when the author says "Lowell has contributed more than Beckett" is he factoring in BP effects? Because unless Fenway has turned into a pitcher's park since baseball died the author has been smoking a mighty, mighty spliff in making that comment this season.
3. Typically, when a team makes a prospects for veteran deal it's with winning in mind. Does the author plan on giving points, or is this just some kind of fantasy baseball thingy?
4. Why on earth would Florida play this kid out of position? I've never seen him play, since baseball is dead to me, but if he makes you pine for the days of Hubie Brooks, well you've got more hell than a little bit.

I think maybe people still don't like giving Boston their due. Gone are the days when a Reggie Smith leaves it all out on the field and can't win jack diddley because the BoSox go cheap on the pitching.

This pieece doesn't even mention the fact that Ramirez' value decreases when they move him to one of the corner infield slots or the OF (unless I missed it).
 


I’ll try to answer the questions from LT, and keep it brief. Not sure how successful I’ll be at either.

1. What is VORP? Does it factor in ballpark effects?
Short answer: yes, VORP adjusts for ballpark effects. It’s Value Over Replacement Player, or the number of runs contributed beyond what a replacement-level player at the same position would contribute if given the same percentage of team plate appearances. For each player, replacement-level is calculated based on playing time at each position.

VORP’s advantage that it facilitates comparing players at different positions, and comparing pitchers to batters. Caveats include: it doesn’t measure defence, and it’s cumulative (not a rate stat) – it describes what a player produced in the playing time available. If two players of equal production rates have unequal playing time, the one who batted more will have the higher VORP. There’s a pretty decent explanation on wiki.

2. Re: “Lowell has contributed more than Beckett” – you’re smoking a mighty, mighty spliff in making that comment this season.

Technically, I was talking about their contributions over the 2006 and 2007 seasons. Beckett made a league average contribution last year, while Lowell’s performed well in both seasons.

That said, the statement holds true for the 2007 season.

I was surprised to discover that Lowell’s contribution outshines Beckett’s – I was expecting to find that Lowell was merely an acceptable price to pay for Beckett’s production.

In the American League, Lowell’s VORP ranks him second among third-basemen, behind only (obviously) A-Rod. Now he’s a long, long way behind, but hopefully this conveys how good a season he’s had (currently .326/.380/.500 by traditional metrics).

It’s the measures beyond VORP that lead to the conclusion that Lowell’s outshone Beckett.

For that I went to Runs Above Average. Lowell has generated 30 Batting RAA this season, and 7 Fielding RAA. Beckett meanwhile has produced 32 RAA.

There are two things going on here. First, Lowell’s defence, which is not going to show up in a box score anytime soon, is quite good – to the point that it’s about 20% of his value. Second, Lowell has taken up 10.1% of Boston’s at-bats this year. A superlative everday position player is simply going to contribute more to the team’s success than a starter who’s in every 5 games.

A final aside: since you long ago abandoned baseball, you may not realize how thin the 3B position is in the American League. By VORP it’s A-Rod (92.8), Lowell (43), Figgins (39.8), Beltre (30.2) and Glaus (20.6). Yes, the 2007 version of Troy Glaus is the fifth-best batting third baseman in the American League. It’s embarrassing.

With apologies to Alex Gordon, the exciting, talented 3B are in the National League: David Wright (69.5), Miguel Cabrera (66.8), Ryan Braun (49.2), Aramis Ramirez (41.6), Garrett Atkins (25.0) and Ryan Zimmerman (23.7). Old timer Chipper Jones (67.4) is right up there too.

3. Typically, when a team makes a prospects for veteran deal it's with winning in mind. Does the author plan on giving points, or is this just some kind of fantasy baseball thingy?

I confess you’ve lost me. The centerpiece of my conclusion was that Boston has increased its short-term production by about 3 wins (2007), albeit at the expense of losing a star whose projected five-year production and marginal value far outshine what’s anticipated from the veterans they acquired. To me, this is win-win.

That Boston missed the playoffs last year is obvious – and Beckett’s first-year inadequacies are part of what went wrong. That they’ll make the playoffs this year, and perhaps win the division, can in part be attributed to the net 30 runs above average that they’ve gained from this deal in the 2007 season.

4. Why on earth would Florida play this kid out of position?...This piece doesn't even mention the fact that Ramirez' value decreases when they move him to one of the corner infield slots or the OF…

Florida isn’t playing Hanley Ramirez out of position: he is, and always has been, a shortstop. He’s just not very good with the glove.

All is not lost. Ramirez played his age-22 and 23 seasons in the major leagues. He’s still learning, and I would think that given how high the scouts are on his tools that he could mature into a league average defender. His rookie year defence wasn't as bad as this year's, so there may be an element of bad luck in this too.

Florida could move Ramirez, but not without making other changes. Third base is blocked by another elite player, 24-year-old Miguel Cabrera. The Fish are currently using Mike Jacobs at first base, which is nothing to write home about. The more significant block to that change might be the conventional wisdom that first should be occupied by a power hitter. Ramirez has a solid slugging percentage, but it’s driven by his speed (doubles, triples) more than out-of-the-park power. Having a first baseman who steals 50 bags in a season may be too out of the box for baseball management.
 


Ramirez has a solid slugging percentage, but it’s driven by his speed (doubles, triples) more than out-of-the-park power.

His power is nothing to complain about, though. He's been steadily improving on that. I honestly think he could be the next 40-40 player. Imagine a shortstop doing that even 25 years ago. Inconceivable.
 


avi: I don't know as much about baseball as you do, but they're playing this kid out of position unless his range factor is Ozzie Smith area. Does he get to a lot of balls? Is he improving?

I think the problem comes in defining "replacement level" in your study. Just because the AL has a weak third base group in a single year can't really counter the value of a Beckett in Fenway.

I'd say just by making the statement "Mike Lowell is superior to Josh Beckett in 2007" you could safely consider the equation a failure.
 


I'd say just by making the statement "Mike Lowell is superior to Josh Beckett in 2007" you could safely consider the equation a failure.

Uh, why? I'm confused as to why you think this commonly accepted metric is unacceptable compared to what you've been seeing.
 


Andy: What we have here (as described by avi) is a short term dip in quality at a specific position. While it may appear that Lowell's value is increased to enormous proportions, fact is that he's a good not great 3b who is having a fine season in all areas in a league where there is none.

Listen, I don't follow baseall much at all. But any reasonable person is going to point out (perhaps more respectfully than I have in this thread) that along with the checks and balances that are in this VORP thing there needs to be a "does this make sense?" clause that allows the user to reward a Josh Beckett kicking out the jams in Fenway.

Is Fenway still a hitter's park? Are rh batters still hitting the monster?

I know you'll come back and say "LT all those things are considered" and I'm still going to argue that the dip in 3b quality is being compensated for and then some in avi's study.

It is what it is, Andy.
 


Where to begin? Maybe I'll split this into the different issues LT raised:

Ballpark factors
As mentioned before, VORP is adjusted for park effects; and yes Fenway is still a hitter's park, although only in one respect -- doubles. This season it's actually ranked #1 for hitters; last year it was 13th.

Tropicana field is a very tough park for batters -- 5th toughest in 2007. Hanley gets credit for doing so well given that handicap.

Beckett at home this year is 9-4 in 16 starts with a 3.98 ERA. On the road he's 10-2 in 12 starts with a 2.23 ERA.

A temporary dip in 3B quality?

No. I regret bringing up the thin crop of third basemen in the American League. It's not relevant to the numbers I'm referencing -- I was just pointing out that Lowell is vastly superior to most of his league rivals (and that in this one area of offense, the NL is looking really good).

Lowell is looking so good in all of this because he's having an incredible year -- he's batting 47 points higher than his career average, his OBP is 36 points higher, his SLG is 33 points higher and his fielding (7 runs above average, 26 runs above replacement) is making a noticeable impact on the standings.

The Runs Above Average figures I've been quoting (they're not my invention) are adjusted for season and all-time. The adjustments are described here and here.

Does this make sense?
Well it clearly doesn't to LT. I'm not sure why. Baseball and life are both full of information that leads to counterintuitive conclusions.

Rather than rely on one statistic, I pulled together a group of numbers that let us assess player performance.

VORP, which you've picked on for some reason, suggests that Beckett (52) has outproduced Lowell (42.4) in 2007, but it doesn't factor in defensive contributions. Runs Above Average, which does factor in defence, bumps Lowell past Beckett. If all you looked at were adjusted ERA and OPS, you' say that Beckett is doing better relative to his peer group (pitchers) than Lowell is to his (batters).

My preference, because it includes defence and allows comparison of pitchers and batters straight-up, is to look at RAA. You might want to start advocating in favour of VORP, since it supports your preferred conclusion.

The forecasts are pretty clear that Beckett will deliver more production and value than Lowell over the coming five years, and as I mentioned, I'm sceptical that Lowell will do this again. I'd trade him if there was a reasonable offer on the table.
 


Avi: What does your defense rating include? Is range factor an important element, or does it reward fielding average? How is Lowell's range factor, anyway?

I don't think it's a rational argument to say Beckett is less than Lowell. If you look at "established level of ability", if you look at this season, if you look at career, what you have is one of the best arms of his generation up against a guy who has landed in just about the best spot possible for his ability and team need.

Beckett today, tomorrow and yesterday. Any study that comes up with a different answer isn't asking the right questions.

jmo.
 


I don't think it's a rational argument to say Beckett is less than Lowell.

Avi's entire argument has been rational. Yours has been "seen him good." To say that any argument that comes up with a different answer than the one your predisposed to believing is wrong is the height of irrationality. The point of the article was to compare Hanley and Beckett. One of the surprising side-effects was to realize how well Lowell has been producing. Avi isn't creating these numbers, LT, or fabricating them to fit his own beliefs. He's using some of the most modern and sophisticated metrics out there, and happily admitting that it's showing him things he previously wouldn't have believed. There's not a bit of me that wants to admit that Beckett is of more value than Hanley, because I love Hanley. But looking at this evidence, it's pretty hard to argue that the Sox, at least for now, came out ahead in that trade.
 


Andy: I have to see what is in the equation. It's SO unbelieveable that alarm bells have to be going off. Frankly, I'm suprised that no one else is asking this question. Is the SABR group gone?

For instance: when taking Beckett's current season and comparing it to the rest of the AL, is he taking Beckett's numbers out of the equation in order to get a true number? Also, these defensive numbers Lowell is putting up, do they value range factor over fielding percentage?

Mike Lowell being better than Josh Beckett is like saying Rusty Staub was better than Tom Seaver.

He may have been for a month, but cripes all Friday if you have a measuring stick that makes Lowell worth more than Beckett when Beckett is 18 games over .500 in a two year span and has the secondary numbers to back it up.

Lowell's 15 errors are a nice total if he's still getting to everything, but is he getting to everything?

On another note, I just looked up Hanley Ramirez. Holy shit.
 


Is the SABR group gone?

These aren't Avi's numbers, LT. They are established numbers used by sabermetricians. VORP was created by Keith Woolner, who writes for Baseball Prospectus. You aren't going to find a bigger group of number crunchers than those guys. In fact, I'm sure Avi grabbed these numbers off of that site. Here again is the wikipedia entry on VORP, which should show you how it is formulated.

And I think it bears repeating that Avi isn't saying Lowell is better than Beckett. He's only saying he's performing better than him right now. He said, "The forecasts are pretty clear that Beckett will deliver more production and value than Lowell over the coming five years, and as I mentioned, I'm sceptical that Lowell will do this again."

As a Red Sox fan, and a guy who watches the team as much as possible, I'm actually not that surprised by how valuable Lowell has been to the Sox. He has been hitting the ball very well, and his defensive play is consistently excellent (though John Dewan's +/- system says it isn't as good as other metrics say it is). It's long been guessed at that the managers using advance statistics in baseball have moved on from finding hidden value in hitting to finding hidden value in fielding. When Lowell was acquired, I immediately assumed that was what the Red Sox management team (which includes Bill James) were thinking. I know now that the deal was orchestrated by Larry Lucchino, not Theo Epstein, but at the time I assumed they were trying to get an underrated defender for a critical defensive position. Anyway, as Sox fan, Lowell's value doesn't really surprise me.
 


And to loop all the way back to the beginning, my original point about Lowell's surprising value was that over the 2 seasons we're talking about the comparison with Beckett is favourable.

You can't possibly look at Beckett's season last year and defend it -- he was league average, no better. For pete's sake, he had a 5.02 ERA and Boston's run scoring machine still got him 16 wins. His K rate dropped to 7 per 9ip, and his walk rate rose to 3.26/9.

We seem to have ended up down a rabbit hole, and I'm not sure I'm interested in helping dig us back out. I think I'll head on up to the game day thread and we can save the baseball talk for another day.
 


Looking back, I appear to be wrong. :-)

Thanks for taking the time, guys.
 

Post a Comment

<< Home

This page is powered by Blogger. Isn't yours?