Lies, Damn Lies, and RBIs
Junk science invades Major League Baseball; the fans fight back.

By Dan Lewis, a freelance sports writer
March 31-April 1, 2001

atting average. Home runs. Runs batted in. These three statistical categories make up baseball's "Triple Crown" — an

E-mail Dan Printer-Friendly

honor given to any player who leads his league in all three over the course of a full season. Triple Crowns are rare — Carl Yastrzemski earned the most recent one back in 1967 — yet baseball fans across the nation watch these three categories religiously. The lure of statistics and the easy availability of the data — and over a century of complacency — make these three categories the holy trinity of batting prowess.

Talk about a false messiah. We can measure batting ability much more accurately. And we should.

Home runs are easily the best measure of the bunch, yet are still to be avoided. A player who slugs forty long-balls is a more attractive hitter than one who cannot seem to get the ball out of the infield, but in the end, homers are a one-dimensional stat. Last year, Blue Jays infielder Tony Batista was tied for fourth in the American League with 41 home runs. He also managed to strike out 121 times while failing to walk 40 times. The happy-medium label of "good player" fits nicely. Going by home runs alone, he's a superstar; by some of his other numbers, he's a benchwarmer.

What of those other categories? Batting average is the third rail of common baseball statistics — local radio hosts will laugh at those who doubt the statistic's utility. Its simple formula, hits divided by at-bats, attempts to depict how often a player will do his job; that is, how often the batter will successfully procure a hit. It fails by not considering the times when a player draws a walk or gets hit by a pitch. In either of those cases, the batter does not record an out, and deserves at least some credit. After all, does it make a difference if a batter leads off an inning with a long single or a six-pitch walk? Not in the slightest — either way, he gets the job done.

Finally, there is what the Simpson's grumpy comic-book salesman would likely call the "worst stat ever": runs batted in. Players are awarded RBIs almost any time their at-bat ends with a runner scoring. If Mark McGwire were to hit a grand slam, he'd be credited with four RBIs — one for himself, and one for each of the runners previously on base. The problem with RBIs is clear; McGwire had no control over the fact that the bases were loaded. Hitting a home run with no one on base is no easier than doing so with the bases full (in fact, it is arguably more difficult to hit the solo shot — but that's an entirely different topic). Yet Mac gets extra credit anyway. Make any sense? Didn't think so.

These notions are hardly original. In 1954, twelve years before Yaz hit the trifecta, Branch Rickey penned an article for Life magazine titled "Goodbye to Some Old Baseball Ideas." Rickey, the former Brooklyn Dodgers general manager, was not discussing his decision to sign Jackie Robinson; rather, he was defenestrating hard-and-fast categories while introducing two new metrics: "On Base Average" (now "On Base Percentage," or OBP) and "Extra-Base Power" (EBP). The former attempted to measure a batter's ability to get on base; the latter takes a stab at showing how hard the batter hits the ball. One could add the two numbers, Rickey reasoned, and the end result would show how effective a batter was in relation to others.

"On Base Percentage" takes into account only a few more criteria than does Batting Average, but OBP conveys far more information. With little exception, a player's OBP tells you at what rate he successfully reached base, whether it's due to a fluke single or a 400-foot home run. As a corollary, OBP also shows how often a player gets out: simply subtract the number from 1.

One of the best statistical tools ever developed, OBP has not earned widespread respect. In 1999, more than fifty years after Rickey's article, Topps issued its annual set of baseball cards, but omitted the stat from the back of their card.

Thankfully, EBP has earned some respect — kind of. Over the years, Rickey's gauge has morphed into what us stat-heads call "Slugging Percentage," whose abbreviation is the somewhat cumbersome SLG. (Baseball number crunchers are not only enamored with numbers, but acronyms too. Baseball Prospectus, the ultimate stat-head resource, has categories called everything from "EqA" to "WAA." Bill James, the father of modern-day "sabermetrics" — a.k.a. the scientific study of all things baseball — has dozens more.)

Slugging percentage has a very clean formula: total bases divided by at-bats. A single is worth one toward the numerator; a double, two; a triple, three; and a home run, four. Walks are omitted, but in this case, that makes sense. Slugging percentage only cares about how hard a player hits the ball, paying little heed to how often he hits it. This makes a lot of sense in the end, as a player who slugs a home run in one at bat but flies out the next is roughly equal to one who rips back-to-back doubles.

And, as Rickey prophesized, adding OBP and SLG makes good statistical sense. There are people out there who do not need a wallet insert to figure out how much to tip a waitress, and some of them decided to see if there was a correlation between this sum (it, too, has an acronym, OPS, short for "On-Base Plus Slugging) and runs scored. Ninety percent of the time, OPS is a good proxy for runs scored. Batting average weighs in at slightly over 75%. Wow.

The baseball world is making its way — slowly — toward accepting OPS as a reliable way to rate batters. Rob Neyer, a senior writer for ESPN.com, is most responsible for popularizing the statistic, referring to it in his column more often than Darryl Strawberry gets arrested. In fact, his official ESPN.com message board is half-jokingly titled "Rob Neyer and his Adventures with OPS." OPS is quick and dirty yet amazingly accurate at the same time.

There are other measures that will lead to even better correlations, but let's face it, no one wants to spend hours on end multiplying the number in column B with the one in row A, then dividing it by the sum of…well, you get the idea. Just about every stats page on the web, from ESPN's to the sabermetrician's dreamsite of baseball-reference.com has OBP and SLG readily available. Simple addition is the only math skill that is required, and estimation is allowed. An OPS greater that 1.000 is extraordinary — only eight players have lifetime values above that line (led by Babe Ruth, Ted Williams, and Lou Gehrig) — but anything over .850 is generally good enough to ensure a starting job. Under .700 and you'll need some phenomenal intangibles like a well-respected fielding ability.

It has taken the sport half a century to begin shunning traditional statistics in favor of scientifically sound ones. We may be a long way until the science of baseball makes its way into news columns and onto the airwaves, but the metamorphosis has begun. Just expect another Triple-Crown winner first.