I was in Las Vegas this past weekend, and was asked by a couple of poker buddies about the struggles of Felix Hernandez. Being a few drinks and a couple hundred hands of crazy pineapple into the wind, I made the bold claim that I’d rather have Cole Hamels than the King. Oops.
The point I was trying to make is that is that when you have limited information about a player–and almost any young pitcher meets that criterion–even a small amount of additional information can make a large amount of difference in his valuation. The problem, of course, is that I picked a bad example to support my case. This is not so much because of Hamels’ injury, but because the information we’re getting about King Felix is not all bad.
Let’s take a point-by-point comparison of Hernandez’ numbers on the season to date against his attractive preseason PECOTA projection.
Actual PECOTA 1B 18.5% 15.3% 2B 3.9% 3.6% 3B 0.0% 0.3% HR 3.9% 1.5% BB 9.1% 9.1% K 22.0% 21.7% HBP 1.7% 1.0% GB% 69.1% 66.7%
The three most important indicators in this table are strikeout rate, walk rate, and groundball percentage–and PECOTA has dead-on nailed Hernandez’ performance in each of those categories. The key differences are in the number of base hits that he’s given up, and the number of home runs.
Hernandez’ BABIP on the season to date is .359. That’s the sixth-worst performance in the league among pitchers with at least 40 innings pitched. Although groundball pitchers give up a few extra base hits, that performance is way out of line with what we’d expect out of Hernandez, and almost certainly reflects his bad luck. It’s not like Hernandez is a Carlos Silva type who throws meatballs and telegraphs his pitches.
The inflated home run rate is a bit more of a concern–Hernandez is at nine home runs and counting before Memorial Day, when PECOTA projected him to give up just 11 on the entire season. But it’s much less of a concern than it would be if Hernandez’ groundball ratio had deteriorated with his dinger tally. I decided to watch videotape of the nine home run pitches that Hernandez has thrown this year to see if I could pick up on anything systematically wrong:
|13-Apr||Broussard||93 MPH fastball down heart of plate, 345′ parabola into right field bullpen.|
|13-Apr||Belliard||Hanging curve in middle of hitting zone; probably 395′ deep into left field bleachers|
|18-Apr||Nevin||Slider low and away on 3-1 count — decent pitch that just carried well. About 370′ to RF — barely cleared fence.|
|18-Apr||Wilkerson||97 MPH heater right down the middle on 3-1 count; extended on for 385′ line drive shot to RF.|
|29-Apr||Patterson||Curve that broke late and hung slightly in bottom half of hitting zone — not a bad pitch. About 365′ over RF scoreboard.|
|4-May||Thome||High 93 MPH fastball; got out in a hurry about 365′ to RF power alley.|
|16-May||Melhuse||High fastball that didn’t move much, crushed about 410′ to right field for grand slam.|
|16-May||Swisher||Classic hanging curve, timeable pitch, just over center field wall (390′).|
|21-May||Roberts||Hanging curveball, golfed a little bit to RF (375-380′).|
These are legitimate home runs–only Ben Broussard‘s shot went less than 350 feet. Seven of the nine were on obvious mistake pitches–either hanging breaking balls, or fastballs where Hernandez got too close to the power-happy portions of the strike zone. If there’s a pattern to be discerned, it’s that Hernandez is a bit of a momentum pitcher. He has struggled when behind in the count–opponents have hit .379/.471/.644 (!) against him after a first pitch ball–as well as with runners on base.
What does all of this mean? Well, it means that Hernandez is behaving like a 20-year-old pitcher with all of five months of big league experience to his name. I think opponents are reading the scouting reports and recognizing that, while Hernandez is not averse to throwing breaking balls early in the count, he rarely throws them for strikes. I think Hernandez has such great stuff that he’s never had to make many adjustments. And I think he and Rafael Chaves will sooner or later come up with the necessary counter-adjustments, whether it means making his slider a bit more of a strike pitch, throwing more first-pitch fastballs, or something else.
In short, I think Felix Hernandez is going to be fine. For some confirmation, I did some scenario testing by means of PECOTA. First, I plugged Hernandez’ 2006 PECOTA projection in, and rolled everything forward by a year. Then I ran another projection by prorating Hernandez’ year-to-date numbers over the entire 2006 season. As you can see, there is rather little difference between the two:
Five Year Projection Based on 2006 PECOTA IP ERA VORP 2007 (Age 21) 189.3 3.31 +41.4 2008 (Age 22) 194.7 3.42 +35.3 2009 (Age 23) 193.7 3.34 +38.1 2010 (Age 24) 195.7 3.30 +40.7 2011 (Age 25) 212.0 3.33 +38.7 ------------------------------------------------ Totals 985.4 3.34 194.3 Five Year Projection Based on 2006 Year-to-Date Results IP ERA VORP 2007 (Age 21) 191.0 3.57 +35.7 2008 (Age 22) 199.7 3.80 +28.3 2009 (Age 23) 205.7 3.55 +34.7 2010 (Age 24) 204.0 3.55 +37.4 2011 (Age 25) 210.0 3.58 +33.9 ------------------------------------------------ Totals 1010.4 3.61 169.9
Even if Hernandez pitches this poorly for the rest of the season, it would only increase his medium-term ERA forecast by about a quarter point per season, and reduce his VORP by about 13% per season. And it’s unlikely that he will pitch this poorly all season, since the categories in which he’s underperformed are the most influenced by short-term luck. Note also that Hernandez’ projected innings pitched total has actually improved a bit, simply because he’s been free of injury thus far. If Felix Hernadnez were a stock, his share price might have declined by six or eight percent since the start of the season–not more than that. No, I wouldn’t trade Felix Hernandez for Cole Hamels. Justin Verlander–that might be a different story.