Sports

WHITROCK: In defense of knowledge

Statistics don’t tell the whole story.

Or so I’ve been told. Part of the trouble with having more than a passing interest in sports is figuring out which arguments are valid and which are not.

The sabermetrician, in between glances at the gaudy figures decorating Albert Pujols’ stat sheet, will tell you the relevant information for any game is contained in the numbers. Statistics quantify performance. Through careful examination, a sufficiently valid approximation of reality can be obtained.

The guy sitting next to you at the bar, on the other hand, doesn’t want to hear about weighted On-Base Average (wOBA) or Wins Above Replacement Player (WARP). The numbers don’t matter, he says: Derek Jeter is a clutch player, Alex Rodriguez is a choker and that’s that (unless it’s a Red Sox bar, in which case both players are unwelcome).

Who is right and who is wrong? Or is it really that simple?

Truth be told, it isn’t that simple. The numbers tell us a lot, but they don’t tell us everything. And everything is important. Of course, some information is more important than other information. And some things don’t necessarily need to be quantified, depending on how precise you need to be.

Consider LeBron James, for example. Joe Average can sit in front of his television, watch James play 48 minutes of basketball and conclude that No. 23 is one of the best players in the league. Few would argue the validity of Joe’s conclusion.

But how valuable is that conclusion? Say Dwight Howard and LeBron James were both free agents, and you have enough money to sign just one of them to a contract. Which player is the better choice? In this instance, it’s not enough to say that either Howard or James is outstanding. If Joe Average is the general manager, telling his boss that Joe based his choice on watching a few games and a gut feeling isn’t going to receive a positive response.

So Joe goes to the stat sheets, but he doesn’t look very deep. Howard led the league in rebounds and blocks, but James is strong in points, rebounds, assists and steals. A weak attempt at evaluating each player’s value yields an equally weak conclusion ‘-‘- that is, an inconclusive one. At this rate, James or Howard will be paid using money from Joe’s salary.

Statistics themselves aren’t the problem. Finding the right stats is what causes so much trouble. A pitcher can win a game while giving up 10 earned runs or lose a no-hitter due to fielding errors. Pitchers shouldn’t be evaluated using wins and losses ‘-‘- they should be evaluated using items in the pitcher’s control.

Sometimes we don’t have the right tools to properly encapsulate a performance. Hockey hasn’t progressed far beyond the absolute basics: goals, assists and penalty minutes for skaters, and win percentage, goals-against average and save percentage for goalies. Basketball is farther along, with WARP, Player Efficiency Rating (PER) and other tools. Baseball is the farthest along of all.

The solution to problems created by poor statistics ‘-‘- or poor use of statistics ‘-‘- isn’t to abandon all things numerical. Where one stat doesn’t work, use another. If no appropriate tool exists, describe the problem objectively and convert that description into something quantifiable.

Claiming one player is more valuable than another or more clutch than another is one thing. Proving it? That’s something else entirely ‘-‘- but presenting empty arguments in lieu of verifiable evidence is lazy. More importantly, it’s unnecessary.

Want to show A-Rod isn’t clutch? If the premise is correct, proving it shouldn’t be hard. Find a quantifiable measure of each game situation’s importance ‘-‘- Tom Tango’s Leverage Index (LI) is a good one ‘-‘- and weigh the impact of his on-field actions using that index, and determine the individual value of each outcome. Okay, maybe it’s not simple addition, but it’s doable.

Sometimes limitations are self-imposed. Other times, the data aren’t available. For example, NBA teams track deflections made by individual defenders, but the values aren’t released to the public. Re-watching every relevant game to get deflection data for a player of interest isn’t a realistic option, so deflections have to be factored out of analyses until a publicly accessible source decides to count them.

Improving our collective understanding of sports is an evolutionary process, but the process can’t continue if it isn’t embraced. The tools are in place to fully utilize the available data. This isn’t an attempt to take the human element out of sports ‘-‘- it’s an attempt to objectively describe the events that define each contest. An increased understanding of the implications of those events can only grow our appreciation of sports, not hinder it.

Statistics are neither an answer to everything nor a useless assortment of numbers. No, statistics are a resource, a body of knowledge to be harnessed. When we learn how to use them ‘-‘- and more importantly, how to use them appropriately ‘-‘- the game will still be there.

We’ll just be smarter.

Website | More Articles

This is an account occasionally used by the Daily Free Press editors to post archived posts from previous iterations of the site or otherwise for special circumstance publications. See authorship info on the byline at the top of the page.

Comments are closed.