Pitching by the Numbers: Believe it or not?

I was planning a simple blocking and tackling ERA piece that looked at the pitchers with WHIPs that didn’t jibe.

But nothing is quite so simple anymore. Fortunately, it is better, however. And we only have to add one other stat to get the most likely outliers, i.e., those pitchers whose ERAs currently are most inflated or deflated. That stat is one I’ve trumpeted here regularly since I stumbled upon it last year. Then, I was trying to figure out why Jeremy Hellickson’s ERA consistently was better than his commonly cited peripherals indicated. Instead, I found the reason why Johnny Cueto outperforms expected ERA (and still does). His consistently low ISO was what was driving the low ERA, not his BABIP numbers, which varied widely while ERA held steady.

So what we’re doing with these charts is looking for where the ERA doesn’t agree with the WHIP or the ISO (slugging average allowed minus batting average allowed). We have a good sense of what a good WHIP and ERA are because we play those categories. But ISO is a new stat, at least as in being applied to pitchers. I’m doing my best to make it more widely used, however. And to help us, I’ll note that the median ISO for all pitches who qualify for the ERA title through Thursday is 1.42. And I’ll also note the ISO ranks for the true outliers following the charts. This first chart is made up of all pitchers who seem to have ERAs too high for their WHIPs.

So we see that most of these apparent ERA outliers, guys I definitely would have bet on in prior years, are not really that because their ISOs are below average. More extra-base hits obviously lead to more runs. But there are exceptions.

Telling you to bet on Lee doesn’t help much. Sale is also a pitcher not available cheaply. So the fact that his ERA is higher than it should be (slightly above average in ISO meaning his ERA and WHIP should be more closely aligned) is not very helpful, either. The big buy/waiver-wire get is Villanueva, still owned by little more than a third of Yahoo! owners. His ERA is solid but should be better given his WHIP (WHIP is more projectable than ERA because the sample of baserunners and innings is bigger/more meaningful than the sample of runs). Villanueva is ranked 41st in ISO allowed (out of 110) – solid.

Yes, Darvish and Strasburg’s ERAs should be better. But you are not going to be able to trade for those guys anyway. But Strasburg owners shouldn’t read anything into his current ERA. And there is no reason to disbelieve that Darvish will remain elite in fantasy (and reality). Finally, Maholm is still available in about 20% of leagues and his ERA right now should be even better (23rd in WHIP and eighth in ISO allowed).

Now let’s look at the pitchers who seem to have too good an ERA in light of their WHIP.

Again the strength of the ISO stat is demonstrated clearly. Most of these apparent outliers are not really outliers if you look at the low quality of the hits they generally allow. The exceptions, meaning pitchers who should currently have a significantly higher ERA, are Guthrie (78th in ISO, 44th in WHIP, 16th in ERA), Wilson (58th, 94th, 67th), Moore (64th, 39th, 13th), Ogando (55th, 62nd, 40th). Given their respective ISOs and WHIPs, I’d say Sale should have an ERA about where Ogando’s is and vice versa. Harrell, Vargas and Sanabia are all overachieving in ERA but are not mixed league options anyway. Most interestingly, Moore’s number of baserunners and high hit-allowed quality mean his ERA should be higher right now and, presumably, going forward. Note this does not mean I think Moore is a dog. He’s still good, just not this good in ERA unless he allows fewer baserunners and/or weaker hits.