I’ve been a baseball fan for as long as I can remember and I still don’t understand how errors and earned runs work. For example, in tonight’s Seattle-Oakland game, the bottom of the 4th inning went like this:
J. Vidro singled to shallow right
K. Johjima singled to right, J. Vidro to second
J. Lopez safe at first on C. Gaudin’s fielding error, J. Vidro to third, K. Johjima to second
Y. Betancourt flied out to right
I. Suzuki walked, J. Vidro scored, K. Johjima to third, J. Lopez to second
A. Beltre struck out swinging
R. Ibanez homered to deep right, I. Suzuki, J. Lopez and K. Johjima scored
J. Guillen grounded out to second
The pitcher was not charged for any of the 5 runs scored. The way I though it worked was if you assume that the error was not made and instead an out was recorded, Beltre would have been the last out of the inning. Therefore, the pitcher would be charged with 1 unearned run and 4 earned runs.
But that’s not how it works and the more I think about it, the more confused I get. I can see the flaw in my original thinking but I can’t figure out the solution.
If the error hadn’t happened, first base would have been open when Ichiro walked so he wouldn’t have driven in a run. Beltre’s strikeout would have been the third out of the inning, so Ibanez wouldn’t have had a chance to hit the homer. So, none of the runs is earned.
While this is certainly true, I think this just highlights the idiocy of the “earned run” concept. We are to assume that anything after what should have been the third out is somehow not the pitchers fault when it most certainly is (at least to some extent).
Also, in this case it is not clear that the first run wouldn’t have scored without the error. If Vidro advanced to third on Lopez’s out, he may have scored on the fly ball by Betancourt. I didn’t see the game so I don’t know for sure but it seems plausible.
I think its more useful to just use runs and run average (RA) and forget about ERA, especially in an era where errors are much less common than before. I also think it has been shown that pitchers that are good at preventing earned runs are also good at preventing unearned runs so this should give a better indication of true pitcher quality than ERA does.
Jas09, I agree that ERA is dumb. But the scoring rules, as I noted above, are pretty clear about how to score it. Absent the error, Vidro would have advanced to third on the walk instead of scoring, and while it’s possible to score from second on a flyout, I haven’t seen it happen.
There’s an infinite universe of possible outcomes, but the official rules direct the scorer to give the pitcher the benefit of the doubt, and in this case it looks like it was done correctly.
That said, ERA is a terrible measure of pitcher performance. If you’re not already reading Baseball Prospectus (www.baseballprospectus.com), you may find it worth checking out.
But my understanding is that had there been a change of pitchers any time after the error, then any batters that he put on base who scored would have been counted as earned runs against the pitcher. Perhaps had the statistic been called “chargeable runs” or something of the sort since it is purely a pitching stat Whether it makes sense even then is another question. None if these stats is entirely fair. A pitcher who comes in with a runner on first that he allows to score does not get any penalty; it is charged to the guy he replaced. Maybe they should charge him with 3/4 of a run (assuming the runner was on first) and his predecessor with 1/4. That is why ERA of reliever’s is misleadingly low. Any change would make stats incomparable with the past.
Yeah, I’m pretty sure BP is where I found the study I was referring to. Their SNLVAR and related stats are much better than ERA (and I find their reliever stats even better than the ones for starters - using ERA to evaluate relievers is quite ridiculous).
I second your recommendation for any baseball fans out there that enjoy the “mathy” side of the game.
Since Earned Run Average is an individual measurement of pitcher effectiveness, why wouldn’t it make sense to remove the incompetence of his teammates from the equation? If a pitcher throws a no-hitter and loses (which has happened, by they way), how does that run negate the performance of the pitcher? Wins and losses tell only part of the story. A pitcher with an exceptional ERA and a shady W-L record is getting no run or defensive support, and it’s easy to see that. For that reason, I take no real issue with ERA.
Oh yeah, ERA is much preferable to W-L. But that’s a pretty low threshold since pitcher wins are not very useful at all.
The problem is that by removing unearned runs you can remove a pretty large chunk of the runs a pitcher allowed in a season. This can make a pitcher with a bad defense actually look better than he really is. The ability to prevent unearned runs is a good quality to have in a pitcher and should be measured. Also, I think ERA tends to benefit ground ball pitchers since most errors are committed on ground balls.
I don’t have a huge problem with ERA (its much better than the default stats for hitters - AVG and RBI, or closers - SV), but I think that plain RA is a little better and doesn’t have the seeming randomness that the OP complains about. And if you’re going to go to the trouble to remove unearned runs you might as well go all the way and factor in park effects, etc…
Here is the Wikipedia entry for some more complicated alternatives to ERA.
I agree that ERA is a flawed stat. I was only explaining why the run wasn’t counted as earned. I didn’t invent the scoring rules.
There are several problems with ERA:
It assumes that “unearned” runs are completely unearned - that is, the pitcher had nothing to do with allowing them.
It doesn’t take park factors into account.
It doesn’t take defense into account.
When there’s a pitching change, it assigns 100% responsibility for each baserunner to the pitcher who put that runner on. For example, if a pitcher is relieved with a runner on first and two outs, and the reliever walks the next three batters, the run is charged to the first pitcher even though the reliever did more to let the run score.
The concept of errors is so flawed that any stat which relies on errors must also be flawed. For example, if a run scores because the shortstop lets a grounder tip off his glove, it’s not counted as earned. If, in the exact same situation, the shortstop lets the same grounder get by him because he wasn’t positioned well, there’s no error and the run is counted as earned. The same batted ball has the same result, but in one case the run is counted as earned and on the other it isn’t. In both cases, if the shortstop had been doing his job properly the run wouldn’t have scored.
Better stats than ERA include strikeout-to-walk ratio, WHIP (walks plus hits per inning pitched) and DIR (defense-independent runs).