Don't just use stats for their own sake

When I was a freshman in college, I had a friend who was on some kind of philosophy kick or something (maybe it was a class she was taking), so she used to walk around and ask in a joking, whiny voice, "What does it meeeeeaaannnn?"

As part of my job involves trying to make things relevant for people, I ask myself and the people who work for me that a lot. In honor of her last name, I call it "Sullivan's Question." In my business, if you can't make something meaningful for people, you're wasting your time.

I got thinking about Sullivan's Question outside of work recently when I came across Jay Jaffe's Sports Illustrated blog about sabermetrics starting to make their way into baseball broadcasts (he likes the idea), which linked to a New York Times article on the same subject.

I will never buy into sabermetrics completely, but I also don't want to be so closed-minded that I can't see where they may have value. I think MLB Network's new show "MLB Now" could be a really good show it it features smart conversation between a true believer in sabermetrics in Brian Kenny and Harold Reynolds, who's more of a skeptic, so I hope it isn't just a forum for them to automatically disagree all the time.

But both the Jaffe post and the Times piece point out what will be the major issue with bringing advanced stats into the booth.

"Now, as the two (Robert Ford and Steve Sparks) settle into the Astros’ broadcast booth, they and their colleagues across the country face a balancing act. How much do listeners want to know about these advanced numbers? How much is informative? And how much would prompt the audience, a group that spans all generations, to tune out?
Listeners and announcers alike say that striking the right balance will be a challenge."

The challenge of that balance is going to be making those statistics meaningful to people, to answer Sullivan's Question, and not just spouting off numbers for the sake of it and hoping people figure it out. (I have a term for that, too, the "Graveyard of Numbers.")

Some are easy. WHIP (walks and hits per innings pitched) and OPS (on-base percentage plus slugging) are simple addition. I happen to think they can be superfluous in that people can figure out if a pitcher doesn't allow a lot of baserunners, a batter gets on base a lot and/or tends to do a lot of damage without them, but I don't find them offensive and they can perhaps be helpful.

However, those stats are easy for fans to grasp because they're simple math. What about the harder ones? Even though I can't think of them by name off the top of my head, I believe there are several stats intending to strip out luck and ballpark factors and quality of the team someone plays for; can those be implemented in a way that intelligently spells out for fans something they may already know instinctively if the math behind them is complicated?

Or what about something like WAR (Wins Above Replacement), the formula to which I find incomprehensible? In the future, will it be enough to just cite someone's WAR and people understand what that means, or will the math involved be its undoing?

My first instinct was to say no, that Mike Trout's 10 WAR last year, which from what I've read is sensational, could be meaningless to fans because they wouldn't be able to understand where those numbers came from.

But then I thought about it some more, and realized there are numbers in life that are relevant and meaningful to people who aren't familiar with the statistical methods used to acquire them. For example, how many people who don't know the science of television ratings or political polling read the numbers and think "Wow, Show X is really popular" or "Gee, Candidate Y looks like he's going to win"?

So it's possible, over time, for fans to become as comfortable with advanced statistics on the TV or radio as we are with terms like batting average that we've used forever and know what they "mean" (this point provided by Poopsie, my ambassador to analytics, with whom I am chatting on Facebook as I write this), but at least for the time being, it's probably going to take a skilled practitioner to answer Sullivan's Question and give them meaning.

I think we can all agree that John Sterling, quoted in the Times piece as avoiding advanced numbers, is NOT that guy.

I will close with a brilliant example that Poopsie just gave me.

"I would LOVE for a ground ball base hit to roll into left field, and then the announcers be able to call up video of that shortstop giving up ground ball hit after ground ball hit all in that same location.
The announcer could say 'According to data complied by Baseball Info Solutions, this guy only fields 30 percent of grounders hit 7 feet to his left, a league average SS gets them 60 percent of the time.'"

Now that would work.