Baseball-reference has a stat that is rarely ever used: bequeathed runners.
It's the number of runners on base when a pitcher leaves the game. Subsequently, there's also a stat on how many of these runners score. These runs are added to the starter's ERA, even though the relief pitcher was pitching and, ultimately, allowed the runs to score.
Want a full year of data? Look at the stats from 2009.
Roy Halladay is the only starter who didn't have a single bequeathed runner. Think about that. He started 32 games and left every single game without a runner on base. Of course, leading the league in complete games helped. His 2.79 ERA is a true representation of how he pitched.
On the opposite end are Jonathan Sanchez and Zach Miner who had 30 bequeathed runners apiece. In Sanchez' case, seven of them scored— his ERA ended up being 4.24.
I'm not one to change stats, but I'll do a "what if" here. If none of Sanchez' seven bequeathed runners would've scored, his ERA would have dropped to 3.86. And if the Giants had an awful bullpen and all 30 of them would have scored, his ERA would skyrocket to 5.51.
That's a 1.65 range that his ERA could have fallen to. Since less than half of the 30 scored, he was fortunate to be on the low end of that, but he could just easily have been on the high end.
Miner had a very similar ERA at 4.27, but 11 of his bequeathed runners scored. Since he didn't pitch as many innings as Sanchez, his range was higher. His ERA could have been anywhere from 3.22 to 6.14; a difference of 2.92!
Miner is an unusual case, since he only had 5 starts; the majority of bequeathed runs come from starters. A pitcher's bequeathed runners become another pitcher's inherited runners, a more common stat that is used to gauge relievers. Miner had 29 inherited runners and 10 of them scored; so I guess you could say his ERA is pretty accurate, as his 10 runs scored from inherited runners cancel out his 11 runs scored from bequeathed runners. Are you as confused as I am?
Who had the worst luck? Probably Roy Oswalt. 11 of his 12 bequeathed runners scored. His ERA would have fallen in the range of 3.57 to 4.17. The low end of that was more in line of what was expected.
The best luck? A.J. Burnett. Only 1 of his 19 bequeathed runners scored. His ERA range was between 4.00 and 4.83.
Another Yankee, Joba Chamberlain also had good luck, only 1 of his 14 bequeathed runner scored. His ERA range was from 4.69 to 5.49. The Yankees bullpen was really good (3.91 ERA) so this is understandable.
I could keep going on with other examples, but you get the idea.
What does this all mean? Not much. It's a very interesting statistic, but l ike I said, I'm not one to change stats.
This could, however, be a useful tool in fantasy baseball strategies.