How to Lie With Statistics

When the Sox fell to the Orioles 2-0 last night, starting pitcher Ryan Dempster earned the loss.

Jun 14, 2013; Baltimore, MD, USA; Boston Red Sox pitcher Ryan Dempster (46) throws in the second inning against the Baltimore Orioles at Oriole Park at Camden Yards. Mandatory Credit: Joy R. Absalon-USA TODAY Sports

No. “Earned” is not the right word. It would be more correct to say “due to the quirky nature of baseball, a perfectly good performance from this pitcher did not result in a win due to a lack of offense” but that might be a bit too wordy for some box scores.

Dempster allowed only five hits over 7.2 innings. He gave up five walks and had four strike outs. He is now 4-7 with 4.21 ERA. He gave our exhausted bullpen a much-needed rest. This performance could easily have lead to a win if the offense had managed to get some runs.

I don’t know about you, but the win-loss stat for pitchers is totally useless. To give an accurate portrait of the hurler it must be accompanied by ERA. Look at John Lackey: he’s 3-5 with 3.14 ERA. He has a “losing” win-loss record which could lead you to believe he’s the same terrible Lackey from two years ago. But his ERA is much better which gives a glimmer of hope. To my mind, ERA and Quality Starts reveal a more accurate picture of the pitcher.

Tonight Lackey goes against Freddy Garcia who is 3-3 and has a 4.47 ERA. He has a better win-loss record than Lackey but a higher ERA. So, who’s the better pitcher? Who will get the win? For all we know the Sox will hit 15 HRs while Lackey gives up 12 runs and earn a win. Or Garcia might give up only one run and get a loss when Ortiz hits a 2-run jack. In the end, isn’t this why we watch the game?