Say It Ain’t Sosa
Barry Bonds is the MVP, statistically speaking.

By Dan Lewis, a freelance writer living in Connecticut
October 13-14, 2001

eventy-three home runs, 177 walks, and an .863 slugging percentage — all new records — make Barry Bonds the Greatest Show on Earth. Could this be the best single-season performance by a player? A performance so great, Bonds can only be mentioned along the likes of Ruth, Williams, and Mantle?

Depends on who you ask.

Average, home runs, and runs batted in make up the trifecta for batters. When aiming for greatness, batters are expected to excel in these categories. In March, I argued for a broader set of metrics when judging a players greatness, a set used by the most hardcore stat-heads. Adding the measures of on-base percentage (OBP) and slugging percentage (SLG), Bonds's year certainly becomes one for the ages. Not only did he set a new record with his astronomical slugging percentage, but he is also one of only five players (since the inception of the World Series) to get on base more than half the time. Bonds's "on base plus slugging," or "OPS," tied the record set by the Babe in 1920. Wow.

And then there's the little issue — a very little issue, mathematically speaking — of runs batted in. Bonds finished the year with only (only?) 137, which does not break into the top 100 of all time. Even worse, it was only good enough for fourth in the league. This blemish has led sportswriters to downplay Bonds's year, and some have even stumped for Sammy Sosa to win the Most Valuable Player (MVP) award.

The MVP imbroglio will likely come down to Bonds versus Sosa. Neither of their teams made the playoffs. Both finished with a .328 batting average. Both cleared the formerly magical plateau of 61 home runs. Indeed, OPS and RBI have come to odds.

A few weeks ago, the Washington Post's Thomas Boswell assailed Bonds's march to greatness with logical fallacies and word-games masquerading as evidence. It would be bad enough if he only had called those in the OPS camp "nerds and purists." But he went too far in proclaiming that "there's no way a knowledgeable baseball writer could vote for Bonds over Sosa for MVP if neither's team makes the playoffs." Even worse is how he supports this ridiculous claim.

Boswell relies on the oft-cited, never-accurate statistic of "runs produced." This is runs scored plus RBI, minus home runs. The stat claims to measure how many runs a player produced. And it fails. The root statistics — runs scored and RBI — both rely on the abilities of teammates. If Sosa hits a home run with no one on base, he'd get one "run produced." If he does it with the bases loaded — and note, he has no control over the batters ahead of him — he gets four. If Bonds doubles and is driven in by the next batter, that's a run produced. If he doubles and is stranded on second — similarly, an event out of his control — that's a big zilch.

On the other hand, Sosa's home run — arguably easier to hit with the bases full than empty — is worth the same, in either situation, on the OPS scale. So is Bonds's double. In fact, OPS takes into consideration what may be the most important variable there is: how often a player does not make an out.

Basketball, hockey, football, and even lacrosse use clocks to determine when a game ends. Once the clock hits zero the game is over. The team with the most points wins. Baseball's clock is the out system: a team gets 27 outs, and, unless the game is tied, a winner is declared. As long as a player does not record an out, he costs his team no time. On base percentage — one half of OPS — determines how often a player allows the clock to tick down.

So, how much better was Bonds than Sosa? In a full season, players have about 650 plate appearances, just over four per game. Sosa had an OBP of .437, and based on the 650 number, is responsible for 366 outs. Bonds — and his Ruthian OBP of .515 — cost his team only 315 chances. The Giants, due solely to Bonds's abilities, were given 51 more chances to score runs. Having Barry Bonds over Sammy Sosa is like being allowed to play almost two more games than everyone else. And that is only half the OPS picture.

Slugging percentage is also telling. Sosa and Bonds finished the year with the same batting average, which is simply hits divided by at-bats. But all hits are not created equal. A home run, clearly, is better than a single. That's where slugging percentage comes in. It also has a simple formula: total bases divided by at-bats. Because the BA and SLG have the same denominator, and because Sosa and Bonds have the same batting average (hey, even nerds like a short-cut), whoever has the higher SLG has more effective hits. Bonds .863, Sosa .737. It's not even a contest.

And the MVP vote should not be either.

BACK TO NRO