PREFERING War TO WAR

By Murray Chass

November 14, 2013

If I am guilty, as charged, with demonstrating disdain for the stats patrol (notice I didn’t say geeks or nerds), I blame their arrogance for prompting such feeling.

My dictionary defines arrogance as “disposed to making claims to unwarranted importance,” and that describes the vast majority of statistic zealots who inundate me with critical e-mail.

There are no two ways with the stats patrol. It’s either their way or no way. No method other than theirs can be right. Those who don’t accept their numbers can’t possibly know what they’re talking about.

Mention intangibles, and those who aren’t dictionary challenged might know that an integer is a number but an intangible?

Only numbers count with the stats patrol. If it’s not a number, it doesn’t fit into the formulas that determine WAR and other such rankings. Baseball-reference.com, an in-depth baseball research site, has the results of each year’s Hall of Fame voting, including columns of statistics for each player on the ballot.

The column headings include WAR, WAR7, JAWS and Jpos. Nowhere is there a column headed I for Intangibles, but they exist and often in a meaningful way: how a player performs in critical circumstances, for example.

Aside from their lack of recognition of intangibles, the stats patrol, in my view, has undertaken a mission to undermine one of baseball’s most significant statistics. Wins for pitchers, the heretics claims, is a vastly overrated and misguided statistic.

Writing in response to a recent column about Jack Morris and the Hall of Fame. the writer described himself as a passionate 19-year-old fan.

“One of the main things you bring up is pitcher wins, and how the sabermetric community does not value them much,” he writes. “This is true, but there is a reason for this. A win is awarded when a pitcher goes at least five innings and the game ends without his lead being destroyed. The issue with this is that the pitcher cannot control how his offense plays. In theory, a pitcher can allow one run through nine innings and still lose the game, and a pitcher can allow 7 runs in 5 innings and get the win!

“This means that part of a pitcher getting a win is team dependent! How can you measure the value of a pitcher based simply on whether or not his team hits for him or not? This includes the discussion of wins in the World Series as well. It is not something that a pitcher has complete control over. Even the defense plays a role.”

Another stats patroller writes,

“Wins are the combination of pitching, defense, hitting, and baserunning, so giving Morris full credit for 254 wins is not correct. Don’t give Morris full credit for having Whitaker, Trammell, Parrish, Gibson, Chet Lemon, along with Aurelio Lopez and Willie Hernandez at the end.”

So what should we do about, say, Cy Young’s 511 wins? Do we wipe them out of the record book, the history book? Do we change the name of the Cy Young award maybe to Cy Young and teammates award?

Do we eliminate the term 20-game winner from the baseball lexicon? Does the standard become the 13 wins with which Felix Hernandez won the Cy Young award a few years ago? How do we treat Warren Spahn’s 13 20-win seasons and his 363 career victories?

Why when Max Scherzer wins the Cy Young award does everyone write and talk about him, in the first sentence, as a 21-game winner? Where is the stats patrol when we need them to make sure we get it right?

On an MLB.com program this week, Mitch Williams, the former reliever and now television analyst, disputed the idea that wins don’t mean anything. “Wins mean something” he said. “I’m pretty sure the World Series was won with 4 wins.”

The stats patrol obviously let Williams fall through the cracks.

Then there is the revisionist criteria aspect of the stats world. The stats guys use the fancy statistics they created to judge players who played before the statistics were created.

How can you judge players on WAR when their careers preceded WAR and they didn’t know from WAR? They knew their batting averages and their earned run averages, but they didn’t know their number of wins above replacement.

Do not fear. The stats patrol has all the answers.

Context, a reader explained, “is taken into consideration. Morris is being judged in comparison to his peers, not arbitrarily.”

Another reader explained, “Stats are historical, so going back and using them to compare players with new formulas, is used to dig deeper into players from past era’s.”

The idea of digging deeper into players is aimed at comparing players from different eras. Lots of fans are obsessed with engaging in such pursuits. I do not share their curiosity. I wouldn’t waste a minute on that effort.

I know a lawyer who is compiling a list of the 1,000 best players in history, ranked from 1 to 1,000 no less. If his book is published, I will be happy to look at it and see if it is worth writing about. I have never asked him if he has used sabermetrics in ranking the players.

I would think more of his work if he has used his judgment rather than metrics.

In the collection of Morris e-mail I received, one began this way:

“I’m not sure how familiar you are with Win Probability Added, but it shows how much each plate appearance affected the chances of winning the game.

“If Morris gave up a run when up 4 runs, he would lose very little value. If he gave up the run with only a one-run lead, it would be a big drop in value.”

That did it for me. I decided I would be better off playing a game of War with one of my grandchildren than reading about WAR.

Comments? Please send email to comments@murraychass.com.