Good Question... NEXT!
July 9, 2005


Howard Bryant's "Juicing of the Game" came out this week with much fanfare.  Peter Gammons hailed it as the book the baseball establishment didn't want published.  But the fact of the matter is that it is a disappointment for anyone looking to find insights or revelations into the life of steroid use in Major League Baseball.  Authored by the long time Boston Herald columnist, "Juicing..." is much more an adroit interweaving of about 50 newspaper stories over the last 10 years, chronicling MLB's public relations' successes and disasters during the steroid era than it is any investigative breakthrough.  It's a sad commentary on the state of American journalism that probably the most in depth look into steroid use in baseball was done right here at longgandhi.com.  It's ironic, to say the least.  And this is just sports.  Imagine how much has been buried over the last 6 years on topics of real importance.  What happened to real journalism?  When did fluff, second rate humor and the contrived fiction of reality TV replace integrity and a moral compass in our news media.  When did America become so jaded that real issues became just an alternative form of entertainment to pro wrestling and game shows?  How did our attention spans become so short that a regurgitation of the news now passes for investigative journalism?  It's not enough that there's still a lot we don't know, but now we seem to be forgetting everything we do know as well.  Don't get me wrong - "Juicing..." is very well written, entertaining and a worthwhile archive resource.  It just doesn't say anything that hasn't already been said on the topic of steroids in baseball.

Anyway, I wanted to look a little bit at the way we measure baseball.  Stats are all important in fantasy baseball and becoming increasingly influential in the way organizations build their teams.  But how much can we trust the numbers?  More specifically, can we trust the numbers enough to accurately tell us how valuable a player is and perhaps give a relatively accurate appraisal of how valuable he will be? 

For example, Bill James once described triples as a factor of speed because the yearly leaders in triples are almost always the players who are perceived to be the fastest runners in the game.  But are they really?   If triples occur primarily as an aspect of speed, could we then draw the conclusion that BJ Surhoff is a faster runner than say Eric Davis or Darryl Strawberry was?  He has more career triples than either.  Dmitri Young has the same number of triples in his career as Alex Rodriguez despite playing for two fewer seasons.  I doubt you'd find too many scouts who would say that Young and Surhoff are speedsters, yet their triples totals suggest an as-yet untapped resource of speed. 

Or could it be that triples aren't so much a factor of speed, any more than doubles are, but that the park or the type of hitter play a more important role?  Could it be that both Young and Surhoff have played in parks that tend to increase triples?  Or the fact that they are predominantly a line-drive hitters while Davis, Strawberry and ARod hit more flyballs.  Fielders have less time to react to a line-drive than to a flyball giving the hitter slightly more time to run when a ball scoots by.   In Saturday's Orioles/Red Sox game, Miguel Tejada and Larry Bigbie both hit triples to center field.  Johnny Damon made an ill-advised dive for Tejada's drive that resulted in the ball rolling to the wall and on Bigbie's flyball, misjudged the distance to the wall resulting in an awkward collision that sent the ball rolling back toward the infield unabated.  Maybe opposing outfield defense, especially considering that teams play an unbalanced schedule, plays a significant role as well.  So the question then is where was speed required?

How do we measure speed?  Outfield range, stolen bases, infield singles, extra bases taken on base hits and doubles on soft or slow hits to the outfield can all display speed. And triples have an aspect of speed in them at times.  Yet only two of those six events have reliable statistics attached to them.  It seems to me that the standard measures more identify the players who don't have speed more than they do players who do.  No one stole more bases than Rickey Henderson, yet because he took such a big swing he didn't get that many infield hits.  So was he fast (as measured by stolen bases) or slow (as measured by infield hits)?

Perhaps faster players score more runs than would be expected and/or slower ones score fewer. Still, that and other "speed stats" like range and SBs might be as much attributed to savvy as speed.  Sure, the players who accumulate extremely high stolen base totals can be safely described as speedy.  But once the numbers get closer to the norm, can we really say that Paul O'Neill - 22 SBs with only 3 caught stealing in 2001 at age 38 - was speedy?  Jeff Conine always amazes me how often he goes from first to third on a single.  Is he that fast?  I don't believe so.  He just has an uncanny sense of how much time he has to run wherever and to whomever the ball is hit.

As a longtime Padre fan and a game scorer for Major League Baseball, I've seen Chris Gomez hit plenty of balls to the gap in right center, then watch in horror as he almost always tries for second base.  Most of the time, the fielder gets to the ball and guns him down because he's slower than a 3-year old eating vegetables.  Gomez still gets credit for a single in the boxscore and that's pretty much the extent of his statistical record on the play.  On rare occasions the fielder doesn't field it cleanly or makes an awkward throw allowing Gomez to reach safely and be credited for a double.  A moderately fast runner probably makes it to second without much trouble.  A more self-aware slow runner probably doesn't attempt Gomez' gambit.  He doesn't get the extra couple of doubles per season, but he also doesn't create the score of extra unnecessary outs.  The stats show Gomez as a guy who hits a fair amount of doubles, but fail to reveal the extra outs he creates with his baserunning... there is no statistical consequence except for possibly when his team loses.  But the losses get nailed to the pitcher's record, not Gomez'. 

So the most obvious physical attribute - speed - escapes qualitative measure in baseball.  We can watch a player run and know whether he is fast of not, but if we limit outselves to just looking at the numbers we still don't know how much his speed will help.  Which means we really don't know when Luis Castillo is going to steal 60 bases or just 20, or who was more likely to steal 30 bases this year - Shannon Stewart or Torii Hunter?  Alex Sanchez is extremely fast, but he gets so many bad reads on flyballs that teams hate to play him on defense.  That limits how often he gets to use his speed on offense.  Just looking at the numbers, however, doesn't reveal that very important bit of information. 

One might conclude that I'm trying to say that stats don't matter.  Nothing could be further from the truth.  Statistical analysis is a very important and, at least when it comes to major league front offices, appears to be a sometimes neglected aspect of understanding the game.  However, with so many sources for statistical analysis on the web, it's not enough to only look at the numbers to get an edge on your fantasy competition.  The methods used to gather the data and the measures we ascribe value to are far too primitive to offer definitive understanding.  We still use the judgments of official scorers interpreting vague rules to determine a significant number of results.  What is "reasonable effort"?   How is it different when a player makes a mental miscalculation yet gets to ball only to fail to cleanly field it (often ruled an error) any different than a player who makes the same mental mistake yet can't get to the ball (often ruled a hit)?  My point is this: not only are statistical projections guesses, but analysis of already established numbers leaves significant room for interpretation.  Until statistical analysis is tied to a more accurate method of data gathering (perhaps video analysis) one must still watch the games to understand the real value of individual players, both in reality and for fantasy games.  Surprise fantasy seasons occur largely because last year's statistics didn't reveal the skills and talents that were already on display for those who watched.  As Yogi Berra once noted, "you can observe a lot just by watching."