THE BOOK--Playing The Percentages In Baseball

I can’t believe I used to love Tom Boswell, the popularizer of Total Average.  This is the worst-kind of reporting or analysis you will find:
http://www.washingtonpost.com/wp-dyn/content/article/2008/05/22/AR2008052203868_pf.html

And I blame baseball’s easy availability of numbers for this.  To think that you can report the number of HR hit in a vaccum, without being fully aware that hitters, pitchers, parks, climate, bats, and balls all have an influence on it, and make your conclusions as if “all other things equal” is a fact.

Context, context, context.  All data is meaningless, completely meaningless, without context.

I had read this article before, and scanned thru it again now.

Today’s ballparks are more homer friendly, and the number of pitchers throwing 25 or more innings has grown 25% while the number of batters with 50 ab has grown only 12% - but I will concur that the major factor is likely the ball.

You only looked at HRs in this study, but I have also looked at BABIP, and have seen the same jump in 1993-1994, not only for the league, but also in pitcher’s hitting.

If pitchers see a spike in their base hits while batting, I doubt it’s the parks or PEDs.

As for “just-enough” HRs, Jay Jaffe’s article at BP looked at average fence distance, but he didn’t consider fence height. In the best example, according to KJOK’s park configuration table, Fulton Co Stadium raised the fences from 6 to 10 feet, without changing the distance, and as a result cut the HR rate by a third.

Yes, Bowell’s piece was very disappointing.  He used to do much better work, and sometimes still does.  The PED issue leads him into poor thinking, as is the case with so many sportswriters of his generation.

His sweeping (and wrong) description of uniform standards for the 90 years prior to 1995 prompted me to scan the historical trends on offense.  I had forgotten how sudden and large the drop in BA was in the 1940s.  Has anyone analyzed and explained this?  What was the primary change that drove BAs down by 20-30 points?

RIP sabermetrician Thomas Boswell:
http://voices.washingtonpost.com/baseball-insider/2008/11/mvps_howard_k-rod_not_pujols_p.html

I’m assuming this is on-topic because of Tango’s #16.  Whether it is or not, it’s more of a rant than anything else.

What I like about these MVP debates (and there is a lot to dislike about them) is that they force people to reveal what they believe truly constitutes value.  And so often those decisions are miles removed from the standard rhetoric.

For example, traditional baseball wisdom tells us that pitching and defense, particularly “defense up the middle”, is what wins games and pennants.  Home runs may be flashy, but they aren’t that important in the grand scheme of things.  Batting average is still very important, and strikeouts are crippling to an offense.

And so the MVP voters are presented with a choice between two teammates: one an outstanding fielding second baseman who also happens to be a fine hitter, and the other a lumbering first baseman who is not superb in any aspect of the game other than hitting home runs, hits for a subpar average, and racks up raw strikeout totals nearly unprecedented in the history of baseball.  Everyone falls over themselves to vote for the first baseman.

Tom Boswell is also revealed to be a fraud in his own way.  Boswell wrote all of that crap about TA capturing the essence of baseball, which is bases and outs and whatnot.  And it’s become fairly obvious that he doesn’t believe a word of it, at least not anymore.

Patriot:  I think there are two types of baseball conventional wisdom.  There’s the CW of very casual fans, which says the guy with the most RBIs is most valuable.  (HRs are very important too.) What you describe is what I would call the pre-sabermetric view of “sophisticated fans”:  BA is more important than power, Ks are a sin, SBs and small ball are important, etc.  Bill James arrived and we all learned that power was pretty important after all, SBs not so much.  But we also learned that RBIs were hugely context-dependent, so the unsophisticated version of the CW also took a hit. 

Boswell is falling back on the crudest version of baseball CW here.  (He has to, given Pujol’s 100-point BA advantage, low Ks.) Like James, he now seems to somewhat regret the statistical revolution in baseball analysis.  Is this just a function of getting old?  Discomfort at seeing others do it better?  I wonder.....

And Joe Posnanski attends the wake as well:
http://sportsillustrated.cnn.com/2008/writers/joe_posnanski/11/19/boswell.mvp/?eref=sircrc

***

Please, post whatever eulogy you like here. 

As a teenager, I couldn’t wait to get the Inside Sports each year where Boswell had his Total Average list of all players in baseball.  It was great stuff.

Total Average is the very foundation of baseball analysis, the thing that is just one or two stepping stones from Linear Weights.  You were so close Tom Boswell.  So close.

Sorry to see you go.

Any one know how to handle a resurrection?  They even brought back Kirigi and Elektra didn’t they?

Total Average is the stat that inspired James to say, “The world needs another offensive rating system like Custer needed more Indians...What we really need, as I wrote three years ago, is for the amateurs to clear the floor.” (Which is funny, and there’s a kernel of truth to it… but I really wonder how that’s any different from the negative response James’ early work got from some reactionary quarters.)

It seems like VORP is the stat that’s inspired Boswell to say the same thing. Which… I just don’t get it. Is there some Moses complex that makes the Old Guard sabermetricians unable to enter the promised land?

I agree with James, from a theoretical standpoint, that amateurs should clear the floor.  But, this is true in any industry, not just sabermetrics.

And not technically “amateurs”, but “amateurish”.  There’s plenty of amateurs on the PGA who deserve to be there based on their play.  Sidney Crosby went from being “amateur” to almost rookie of the year in a matter of months.  Victor Wang may be an “amateur”, but clearly he belongs in the sabermetric world.  There’s plenty of professional academicians that should similarly get out of the way.

At the very least, stay on the sidelines until you are ready to dance.

Total Average is a stepping stone to understanding how offense works, as is Runs Created.  The both have their place. But each takes a back seat to Linear Weights and BaseRuns. 

So, what needs to clear the floor is the misuse of statistics, and the old methods that have been clearly superceded by things better.

If Bill James said that Total Average should clear the floor, then, at this moment, Runs Created should follow along.

Ugh. I’ve gotta be honest, I hate the “stay ouf of the way” attitdue. I understand the feeling behind it, and I agree with that. It does a disservice to the people who know what they’re doing to have crappy information floating around. With as easy as it is to publish on the internet, anybody who knows how to use a spreadsheet and a little html can be a “sabermetrician.” Fine. I’m with you there.

But it never ever hurts to have new people trying. A lot of the new stuff might be crap. And, rightfully, the established people will point out its flaws. This has all sorts of positive effects. It helps the new guy learn from mistakes, it might stimulate thought in the community (maybe the guy had all wrong methods and results, but the idea behind it is one that needs examination).

If the “promised land” is ever actually going to be reached, I don’t think it happens without the people who are misusing statistics.

JD: The problem is not with those who play the field, but with those who are so stubborn that they are right when they are wrong.

I’m all for amateurs with two-left feet to dance.  But, if I come along and tell them to use their right foot half the time, and they insist that two-left feet is the way to go, then THOSE people need to clear the dance floor.

Our job here is as educators.  But, ONLY for those willing to be educated.  The others do more harm than good.

So, I agree with you that the promised land will be reached with the amateurs… as long as they are willing to grow.  Indeed, this blog is filled with “amateurs” who’ve probably made a misstep or two (me and MGL included), but we all move forward.  Those who want to stay in quicksand, well, there’s nothing I can do for them.

I wish we could have convinced Bluzer to abandon his “absolute average (ABSO)”, which is total average divided by 3.

Bluzer: “...because a good hitter needs to hit for average, hit for power, and get on base, we are going to use the Average of Batting, Slugging, and On-base averages, and he’ll say, “You know, that makes sense.”

Bluzer is probably instructive here - I don’t mean to pick on him, but he’s a good example of what we’re talking about here.

Here’s the ABSO post:

http://www.insidethebook.com/ee/index.php/site/comments/abso_lutely_not/

And… ABSO is wrong. It started off with an idea, and that’s great. But once that idea was contradicted by evidence, it was on Bluzer to either come up with better evidence for his conclusions or alter his position. He did neither, and that makes Bluzer wrong.

But looking at OPS in there, it’s pretty close to a wash - depending on the context/era you could make a case that OPS is better than ABSO or vice versa. Not that it matters (you can do better than either of them) - but OPS still persists.

Or look at RC/VORP:

http://www.insidethebook.com/ee/index.php/site/comments/fixing_vorp/

Same thing! I’m sure if you took Total Average out there you’d get similar results.

The problem is not with being wrong; it’s being unwilling to change your position in the face of evidence that you are wrong - and that persists whether or not you’re an amateur. In fact it’s worse when Bill James (Runs Created) or Baseball Prospectus (VORP) or the old guard sportswriters (RBIs) etc. does it, because they have a lot more reach/influence/etc. than folks like Bluzer. They deserve a higher, not lower, standard.