April 30, 2008
With the first month of the season over as far as this column's publishing date is concerned, I wanted to take a look at the leaders in one of my favorite pitching statistics, QuikERA, or QERA for short. For those unfamiliar with the statistic, let's take a look at the definition:
QuikERA estimates what a pitcher's ERA should be based solely on his strikeout rate, walk rate, and GB/FB ratio. These three components-K rate, BB rate, GB/FB-stabilize very quickly, and they have the strongest predictive relationship with a pitcher's ERA going forward. What's more, they are not very dependent on park effects, allowing us to make reasonable comparisons between pitchers on different teams.
QERA is good to use when we have small samples (like one month's worth of play) since, as the definition states, it uses components that stabilize quickly. That doesn't mean using it is perfect after 20 innings pitched, but it's more useful than plenty of other options. This way, you get a good idea of whether you want to sell high or hold onto a pitcher.
I made a spreadsheet with every pitcher who has thrown a minimum of 20 innings and calculated their QERA; some of the pitchers you expect to see are on the list (Josh Beckett, John Smoltz), and there are others that I've covered recently (Cliff Lee), but there are still more surprising names to see, and it's this last group that's the one we'll take a look at today.
Jonathan Sanchez has posted excellent strikeout numbers, punching out over 31 percent of the batters he's faced in 28 innings pitched; between the strikeouts, a lower walk rate, and a 41 percent grounder rate, Sanchez has a QERA of 2.98, 0.56 points better than his actual ERA. His walk rate has been the concern in the past, because despite filthy stuff Sanchez would get himself into trouble handing out free passes to roughly five hitters per nine. Home runs were also an issue last season (1.4 HR/9 in '07), a surprise considering AT&T Park deflates offense, especially homers. Sanchez has cut his walk rate from 4.9 to 3.5 and shaved his home run rate to a rounded-up 1.0. Sanchez was also beat around to the tune of a .367 BABIP and 9.9 hits per nine innings in '07, but he's reduced those figures to .307 and 7.1 this respectively this year. With only a month of data to work with, it's difficult to say if that BABIP will stick, but there's nothing outstanding about his batted-ball distribution that makes you think it will change much in either direction. My main cause for concern is the Giants defense, which currently ranks 28th in Defensive Efficiency with a .679 mark.
Sanchez has done a better job of keeping his walks down by attacking with his fastball more often-according to Pitch f/x data, Sanchez has used his fastball nearly 81 percent of the time versus the 71-72 percent he used the past two seasons. He's using his changeup less often but utilizing his slider more often to compensate, and it's worked out for the lefty so far. Given the Giants defense probably isn't going to help him much in the five months to come, I think we'll see his hit rate jump a bit, but since he's already over his QERA a bit, I wouldn't worry about a large ERA hike.
Scott Baker probably hasn't been on the radar this year due to his 4.50 ERA and because he's starting for the Twins, but according to QERA, you should start paying attention. Baker's earned a 3.41 QERA, over a full run better than his actual ERA, and he's done this on the strength of 8.1 K/9 and an excellent 1.5 BB/9. In 2007, Baker had 6.4 K/9, but only handed out walks to 1.8 per nine; his career rate in the majors over 310 2/3 innings pitched is 1.9, so the 1.5 mark is close to where he'll end up. The reason he has outperformed his ERA via QERA is due to the homers he's allowed; in 30 innings, Baker's let the opposition go deep 6 times for a HR/9 of 1.8. We've seen this from Baker before as well, when he allowed 1.8 HR/9 over 83 1/3 innings pitched in 2006. In 2007, he held opponents to 0.9 HR/9, so we can split the difference and say Baker's ERA may end up around 4-4.10 rather than 4.50. With the lofty strikeout rate and low WHIP-Baker's already allowing a hit per inning, so we don't need to adjust that much-he's going to be a useful starter for you, regardless of his win total.
Though he's on the DL now and may not return to action until the first or second week of May, Wandy Rodriguez has pitched well during his first four starts. He's over his head with a 2.31 ERA, but his 3.13 QERA deserves an explanation. In 2007, Rodriguez posted a 4.09 QERA against a 4.58 ERA, so the leap of faith to this year's performance isn't as large as it looks. Rodriguez has jumped his strikeout rate from last year's 7.8 to 9.3 this year; that's a significant jump given his 7.7 career rate, but we can assume he's improved a little in that area. The area I notice more is the walk rate, which has fallen from 3.1 (and 3.5 for his career) to 1.9 this season. If he's improved both his walk and strikeout rates a bit from last year-not at the level he currently sports over 23 1/3 innings, but let's say 8.0 and 2.7 or so-then he's going to have a QERA in the high threes. Chances are good Rodriguez is available to you because he pitched on the Astros, is currently on the disabled list, and wasn't considered as a guy who could post an ERA in the high threes during draft and auction time. The thing that makes me optimistic is that he's switched his changeup usage, favoring more curveballs. Rodriguez's change was beat around last year for a .308 batting average, but his curve worked more for him with a .219 opponent average. He's using that almost 32 percent of the time now, mixing it in with is fastball and an occasional changeup. I'm not advocating he ditch the change and become a two-pitch starter, but it's good to see him using his better pitches more often.
Performances from many starters in San Diego are often dismissed as products of Petco Park, but if you look more closely, you can see some pitchers who are for real. Randy Wolf's ERA is 3.68, but his QERA is actually a tad better than that at 3.55. Wolf has succeeded by striking out 8.6 hitters per nine, which would be the fifth time in his career he struck out at least eight per nine, and also by keeping his walks to a minimum (2.5 per nine). Wolf has had control issues in the past, but usually during the years where he was having arm trouble, so a 2.5 rate, while maybe a bit low, is at least believable with a month's worth of data. Wolf has also induced grounders 39 percent of the time, which isn't special, but is enough for a guy who spends half his time pitching in the expanses of Petco. Wolf's problems in the past were often home run-related, though his former home park in Philadelphia didn't do him any favors. Now that he's in San Diego, where homers go to die, the one weakness in his game outside of arm health has been removed from the equation in roughly half of his starts. Wolf is also throwing his fastball more often this year, and hasn't thrown cutters much, if at all, since 2006. No one pitch has lost a significant amount of use in exchange for the seven percent increase in fastballs, but all get fewer looks per start. If the increase in fastballs is part of the reason his control has improved, then he should continue to work this way.
All four of these pitchers are worth having on your roster, even if the team backing them or their ERA was keeping you from grabbing them initially. It's tough to have a good-looking QERA unless you're striking out a decent number of hitters while keeping walks down to a reasonable rate, so taking a chance on any of these pitchers isn't a major risk. QERA is an especially useful stat in use here at Baseball Prospectus, so do yourself a favor and refresh this custom stats report often, while keeping this formula in mind.