The Baltimore Orioles won again today, September 16th, 2012, in Oakland. Their 9-5 victory over the surging Oakland Athletics kept Baltimore within one game of first place and in a position for a wild card spot by three games in the loss column.
Not only does this win put the Orioles in a solid position with only 16 games remaining, but it also defies the new conventional wisdom regarding a team’s run differential.
Essentially, Baltimore is giving new statistics the bird.
To be fair, the concept of examining run differential is an interesting and many times informative statistic. If a team scores more runs than it allows over the course of the season, chances are they are a good team. And a large run differential could be a valuable supplement to the win and loss totals in evaluating the quality of a team.
Just this year, the teams with the best run differential in the each league are, as of this writing, the Texas Rangers in the American League and the Washington Nationals in the National League. They are also the teams with the best record.
The problem is not with the statistic, which is a useful tool, but rather with the mentality of some fans when a new method to crunch numbers is introduced.
Whenever there is a new stat du jour, there always seems to be a group of people who will act as if any old system of analyzing baseball is meaningless. Basically they have the key to understanding the game and others are fools.
Remember when OPS was first introduced? Then that evolved into OPS+ and for about a week people cared about VORP before WAR became the only stat worth mentioning?
Those stats were not only introduced, but as Sports Illustrated's Joe Posnanski wrote, old stats like batting average were declared meaningless.
The run differential band wagon got crowded fast. Michael Wilbon of The Washington Post and ESPN's Pardon the Interruption said in 2008 "The one stat that I now believe in above all others (is) run differential." Also that year Jesse Spector of the New York Daily News favored run differential to a win loss record, writing "a team's run differential should be able to measure the relative strength of a club's lineup and pitching staff, telling you who the best team really is."
That's all fine. But when the praise of this stat gets to be absurd is when it becomes the only method of evaluating a team.
With the Baltimore Orioles in second place for most of the season, their run was written off by many writers solely because of their bad run differential.
Carl Bialik of WSJ.com wrote that the main reason for the Orioles good start was luck and that they would fade away. Keith Law of ESPN refuses to give Baltimore any credit and gave the ultimate quote to explain why fans of new stats can sometimes be so loathed:
"There’s literally nothing that the Orioles can do to convince me that they are a good team."
People I have interacted with on Twitter have said bluntly that the Orioles are lucky and will fade away.
I pointed out to one Twitter follower that the Orioles, that day tied for first place, had a worse run differential than the reeling and awful Boston Red Sox. (They still do as of this writing.) I asked if the Red Sox were a better team than the Orioles.
The person replied yes and I informed that reader that he was insane. I think I have the stats to back up that claim.
As a wise anonymous person once said, "Luck is a loser's excuse to a winner's opportunity."
Evaluating a team solely on one single statistic is insane, even if that stat can be reliable. How can a team be lucky for 146 games?
To say a team that is one game out of first place with little more than two weeks to play is inferior to a team on its way to 90 losses with a horrible pitching staff and inept lineup because they don't measure up in one column is the clearest example of how people can rely too heavily on the new stats.
The Orioles find ways to win, which is the whole purpose of the game. And like the 2007 Arizona Diamondbacks, who had the league's best record and went to the NLCS despite a negative run differential, they are defying expectations.
What was the main reason for Keith Law's bullheaded quote?
"It’d be nice if [The Orioles] outscored their opponents on the season."
It would be nice. But they are beating them.
To the people who put too much weight on run differential, remember that while it is a good stat to use in evaluating a team, it does not take things into consideration.
There are times to allow a run in a game. Conceding a run late. Having the infield play back in the ninth or letting a meaningless run score from third.
There are times when a reliever is left in a game to take the proverbial bullet for the team. Perhaps a pitcher would be bombed and let up several runs, but it would allow the team to have a more rested bullpen for the next day. Sometimes a position player comes in to pitch for the same reason.
In a series against Chicago this year, the Orioles won a pair of close games but lost an 8-1 blowout. Bad for the run differential but the Orioles won the series.
Perhaps this over emphasis on run differential comes from so many people playing fantasy baseball. There it truly does not matter when a team scores a run. In fantasy baseball, a grand slam by a player on a losing team in a 10-4 blowout means more than someone hitting a two run walk off homer in the 10th.
But the Orioles are not playing fantasy baseball. They are playing in reality. They have played more than 90% of the season in a way where they win close games and concede the blowouts and it has been working.
And putting more premiums on blow out wins brings up that awful BCS mentality. "Sure the team won, but not by enough" has no place in baseball.
A streak here and there can be marked up to goo fortune but not with 9/10 of the season in the bag.
There are just 16 games remaining, a small sample size where luck and good fortune truly can make a difference.
If the Orioles are just two games better than the Yankees over 16 games, then talk about luck and crunch numbers to your hearts content.
The Orioles would have more wins, which in the end is how champions are determined.
And do not worry. A new stat will be invented soon that will make run differential irrelevant.