Thursday, August 9, 2007

Joe Morgan

The guys at Fire Joe Morgan go to town on Morgan's recent ESPN chat. As usual, they are both insightful and very, very funny. And as usual, Morgan makes a dismissive comment about OPS and sabermetric principles:

Statistics are overrated. What you do to help your team win is what it's all about. These stats like OPS, it doesn't tell you what you do for the team. To my opinion, to help the team, you drive in runs or score runs. That helps the team. That's how you should be judged.
Not surprisingly, the FJM guys take down that bit of nonsense in a thorough and efficient way. But it's not the first time they or others have had to do it, and it likely won't be the last. A lot of people spend a lot of time writing things about Morgan's strident, almost willing ignorance about statistics and, in many cases, the writers have chalked it up Joe's alleged lack of intelligence or proficiency with numbers, or what have you. I think these explanations, however, are barking up the wrong tree.

Morgan first went on record criticizing sabermetric concepts a long long time ago, back when they were far from mainstream, or at least farther than they remain today. He was one of the principal critics of Moneyball, and even before that seemed to bristle against the increasing use of OPS and other measures of baseball ability that had been filtering into baseball analysis.

I think Morgan's initial criticism constituted a knee jerk response, but not one that was all that different from what any number of other writers and broadcasters were saying at the time. OPS was for statgeeks in their parents' basements. A Nexis search for "slide rule" w/s "baseball" would definitely gotten you more hits in 2003 than it would today. Since those days, however, many of the initial naysayers and doubters have come around. WGN and Fox broadcasts show OBP, and the people who once laughed at the geeks are using some sabermetric principals in their analysis and commentary. But not Morgan. Morgan still sticks to his guns. Why?

Because what defines Morgan the most is not his ignorance, but his stubbornness. I think that now, in a private moment, he would agree that Julio Lugo and his RBIs aren't nearly as valuable to the Sox as Placido Polanco's OBP is to the Tigers. I think he would admit that an RBI is a team-dependent occurrence and therefore doesn't tell you a lot about an individual player's value. He won't do this in public, however, because he is simply incapable of reversing himself in public.

We've seen this recently with both his mistake relating to the hit he didn't have against the collapsing 1964 Phillies, as well as the comment he made about Luis Castillo being unfamiliar with the swirling winds in Wrigley. In both instances, ESPN ordered a correction -- an explicit order in the former, and an apparent one between innings in the latter -- and both times he did so in the most reluctant way possible and at great personal discomfort.

Say what you want about Morgan, but he is not stupid, either intellectually-speaking or with respect to his overall baseball IQ, which many have demonstrated to be among the highest among anyone who has ever played the game. He is, however, bullheaded and unwilling to admit a mistake.

For what it's worth, I think it would be easier to accept a broadcaster that is simply stupid or ignorant than it is to accept one who, like Joe Morgan, is unwilling to change his opinions about that which is before his very eyes. The former can't help himself. The latter can't help anybody.


64cardinals said...

Hey, I'm a stat geek myself. Love 'em, love discussing 'em, love writing them down for just the hell of it. I problably couldn't live without them.

Now for the hard part. I have to do something I didn't think I would ever do.

I kind of agree with Joe Morgan. The stat craze has gotten so out of hand that people forget about the other aspects of playing ball.

Scouts routinely sign guys on potenital, not on any specific numbers they put up. Players are routinely called up to the majors without having gaudy statistics in the minors. Or left down when they do.

There are other intangibles that make ballplayers great. The will to win, the ability to play hurt, field smarts, etc. Not all of this can be attributed to numbers on a scoreboard.

Some managers routinely go with the numbers and it works. Some routinely go with the numbers and it works. But stats can't look into a players eyes and see what a manager sees.

The stats are good tools. But they are not the end all, be all of a player's ability. It has gotten out of hand to where the stat geeks have taken on more importance than the accumulated knowledge of 136 years of professional baseball. Stats are important, but they are a tool. They shoudn't be the decieding factor.

Please don't make me ever agree with Joe Morgan again.

64cardinals said...

Should have said "some go with the numbers and it works, and some go against the numbers and it works".

Shyster said...

I actually agree 100% with that, 64. I think that intangibles and makeup and a scout's gut feeling (as well as what he simply observes with his own two eyes) are very important. To the extent that it Morgan's position -- or anyone else's -- I wholeheartedly agree.

My problem (and FJM's among others) with Morgan is that when he does wade into stats, he values the stats that don't necessarily tell you the most about a player's individial value and is openly dismissive of those that do.

I'm fine with Joe Morgan saying "you can't just look at the stats; you have to look at the whole player." I'm not fine with Joe Morgan saying "OPS tells you nothing; RBIs tell you everything," because I simply think that's wrong.

64cardinals said...
This comment has been removed by the author.
64cardinals said...

See, now I'm conflicted. I agree that OPS is an important number. But I still have to agree with Joe Morgan. Ahhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhh!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

At the end of the day, the only stat that truly counts is Runs. And, to me, the 2 most important statitistics are who scored them and who knocked them in. Yes, RBI's might be more of a team issue. BUt it is the individual batter who does something to get that runner across the plate, albeit with help (sac, sb, go, etc.)

I also agree that OPS is a very important number, as the players with a high OPS are the players most likely to score the runs and drive them in. Again, who scored and who knocked them in.

I have a problem with OPS, however, in that many of the stat geeks continually claim that it is the most important statistic, but then claim batting average doesn't matter at all. Not to insult anyone's intelligence, but this is how I see it.

OPS = OBP *(or +) SLG

OBP = (H + W + HBP) / (AB + W + HBP)


TB = ((1*S)+(2*D)+(3*T)+(4*HR))/AB

In all of this statistics, AB and H play an important part. In fact, without AB and H, you can't have a sluggin pct. With out AB and H, all you have for OBS is walks. By this rationale, Put Burrell is better than Albert Pujols because he has more walks.

So what it comes down to with OPS is that the guy who walks the most is the best.

I can't take a statistic seriously if all the major advocates of it discount some of the stats that make up the stat they are trying to trupmet.

If batting average (H/AB) isn't important, then by defintiton, OPS can't be either. Its just not allowed.

Of course, this is just my opinion. I'm probably wrong.

64cardinals said...

OBP = (H + W + HBP) / (AB + H + W + HBP)

Diesel said...

Dude, dude, dude ...

I think your heart is in the right place, 64, but you (like a lot of people who feel the same as you) completely misunderstand the "stat geeks." Let me clarify:

No one ever said that BA was unimportant or irrelevant. What people like me do say is that BA is a poor indicator of a batter's real value, since it only counts one thing (hits against at-bats), without really distinguishing the value of the hits themselves. If Luis Castillo gets 10 hits in 20 official ABs, he has a BA of .300, which is considered by many people to be "good." If Pat Burrell has 7 hits in 30 ABs, he has a BA of .233, which is considered by many to be "mediocre."

But what if all of Castillo's hits are singles, and all of Pat Burrell's hits are HR? I can't imagine anyone preferring 10 singles to seven home runs, because home runs are the single best thing a batter can do anytime he comes to the plate. So, if we know what the hits are, as opposed to simply counting the hits, we have a much better idea of an individual batter's value.

That's why statistics like OBP and SLG are more telling of a player's value than BA.

However, I don't know one real "stat geek" who still believes OPS is the best stat available to judge a hitter's value. More advanced stats like VORP, WARP, Run Shares and EqA do the job much better, by adjusting for park effects, strength of opposition, era, baserunning and position played.

Sure, lots of people just laugh off funny-sounding things like VORP, but the bottom line is that those things are extremely useful and reliable tools. Just ask the front offices of the Blue Jays, Red Sox, Devil Rays and Indians, all of whom have hired former writers from "stat geek" mecca

It's only the most strident of "stat geeks" who claim that statistics are superior to baseball scouting. You're correct in saying that scouting is still the backbone of prospect evaluation, and I don't know of a single reputable sabermatrician who says otherwise. However, scouting is costly and imprecise, and is dependent much more on intuition than it is anything concrete. Our eyes can lie to us; numbers, when put into proper context, are much more reliable. So, the fusion of traditional scouting with statistical analysis is by far the best approach. Just make sure that your scouts can see instead of simply imagine, and your stats are the right ones.

And, I promise you, runs, RBI and BA are among the least useful stats you can find, unless you're playing fantasy baseball.

Take care,