February 21, 2005
Top 50 Prospects - Pitchers
Tomorrow, we'll announce our Top 50 Prospects for 2005. The list, which is included along with capsule comments and stats for all prospects in Baseball Prospectus 2005, is compiled under the leadership of Rany Jazayerli. The discussions that go into its assembly provide some insight into the process. Today, the pitchers.
Rany Jazayerli: If you've got a favorite pitcher who you'd like to trumpet, go for it. If we hadn't already gambled on a Mariner pitcher as our overall #1 prospect once upon a time, Felix Hernandez might be a candidate for that slot.
Derek Zumsteg: To everyone who hasn't seen King Felix: if you're in a PCL town, and he doesn't break with the team out of spring training, go. If you're not, consider a spring or summer trip to lovely Seattle when the Rainiers are in town.
Will Carroll: Hernandez reminds me of Pedro Martinez on paper, but I haven't seen him pitch.
RJ: There's really no clear #2 pitching prospect. Jeff Francis would be in any other park. The collision of Francis with Coors Field will be fascinating, no doubt, but I've given up on the notion that I have any idea how Coors Field will affect a player.
A few people have expressed worries about Scott Kazmir (Devil Rays): his mechanics, his control, his new team. On the other hand, Matt Cain (Giants) translates very poorly; Francis has Coors Field to deal with; and the fact that Yusmeiro Petit's (Mets) scouting report is decidedly average should count for something, especially since he only has two starts above A ball.
I guess what I'm saying is, if not Kazmir at #2, then who? Which isn't to say that there's not a huge gap between Hernandez and everyone else on this list.
WC: Kazmir's mechanics suck, but the organization might save him. I think he's too high, but I'm not sure by how much.
Nate Silver: We can compare Kazmir and Francis:
1. Strikeout Rate:
Kazmir: roughly 80th percentile versus MLB competition, according to PECOTA baseline.
2. Walk Rate:
Kazmir: <10th percentile.
Ding, ding, ding. Enormous edge for Francis in a very important department.
3. Groundball/home-run rates
Both rate as very slight flyball pitchers. However, Kazmir has had somewhat better home-run rates, ranking in about the 65th percentile according to PECOTA, versus Francis, who is around the 30th percentile. I expect some convergence in these numbers, and there's one further thing about Francis that's worth noting:
In other words, he's turned into more of a groundball pitcher in environments in which that would in fact be more advantageous. The sample sizes are smallish, but the numbers are certainly pointing in the right direction. This is a slight advantage to Kazmir, but not a huge one.
4. Hit rates.
You'll sometimes see a pitching prospect who has good strikeout and walk numbers but a high hit rate on balls in play, which might be a sign of a guy that has advanced technique but lacks raw stuff. Both Francis and Kazmir have been pretty tough to hit, so there's no edge here.
Kazmir: 93-97 MPH fastball, plus slider, workable change.
I don't think it's debated that Kazmir has somewhat better raw stuff. I just don't know how important that is when they've posted comparable strikeout rates at the upper levels of the minor leagues (and the major leagues, for that matter).
6. Injury history/mechanics
Kazmir: Some minor ailments, has been used very cautiously. Smallish for a pitcher (6'1"/170), and Will has some questions about his mechanics.
Slight edge to Francis.
Kazmir is three years younger. This is a point in his favor, but it's not nearly so important as it would be for a hitting prospect; the relationship between pitcher development and age is there, but it's pretty flaky, and there's even the injury nexus argument that younger pitchers are more likely to get hurt.
Lou Piniella versus Coors Field.
I suppose I think it would be slightly unscientific to punish Francis for being a Rockie. The Rockies have not really had very many good pitching prospects; it's true that Chin-Hui Tsao got hurt, but that was before he'd pitched substantially at Coors Field. Breaking-ball pitchers are supposed to have trouble at Coors but I don't know if Francis is really a breaking ball-guy; he's more in the Greg Maddux-lite class of guys who do it with movement and location. Coors is a risk factor, certainly, but it's hard to assign an appropriate value to it.
On the other hand, we have the Devil Rays. I know that Will has had some nice things to say about their medical staff, but they have not exactly had a tremendous record in developing pitching prospects. I'm willing to give a little bit of extra credit to a pitcher with the A's, and a lot of extra credit to a pitcher with the Braves, but apart from that, I think it's mostly speculation.
So I see it as follows:
I realize that there are more checkmarks on Kazmir's ledger, but I think the command advantage for Francis is overwhelmingly more important than anything else, and that's why I think he'd make a better #2.
Joe Sheehan: I think this strongly underrates the negative that is pitching in Coors Field. The Rockies have never had a pitcher be healthy and effective for three straight seasons.
That's real, and if the argument against it is, "well, Francis is different," that's the same argument that gets made for every damn pitching prospect, and it's essentially meaningless.
I seriously doubt Francis will be both healthy and effective across consecutive seasons in the short and medium term unless he ends up with another organization.
Chris Kahrl: I think you had it right in your first and second paragraphs, because that's what's generally true for all young pitchers, and this is an exercise in speculation.
In the same way that I think naming a guy as a Top 50 prospect who wouldn't be a prospect anywhere except possibly Coors is silly, I think not giving a guy who's the second-best pitching suspect anywhere else in baseball due props is silly. It reinforces the "Coors is different and special and needs to be treated differently" rigamarole that has yet to be proven to have anything to do with the Rockies' actual problem, which is that they suck on the road because they suck.
NS: One thing I was trying to suggest in my original message is that we ought to err on the side of the negative hypothesis in this case, which is that pitchers aren't adversely affected by Coors above and beyond the ordinary adjustments for park effects. It wouldn't surprise me if there are additional psychological or even physiological effects that in fact have a statistically significant impact on pitcher performance, but it also wouldn't surprise me if there weren't, and to my knowledge there have been no studies on the matter one way or the other.
RJ: I think Nate and Joe both make good points here. The problem is that the Rockies have never had a prospect of the caliber of Francis, so we really don't know what the nexus of a great pitching prospect and a horrible pitching park will look like.
The one thing I can suggest we do is to look at the best pitching prospect the Rockies had before Francis: not Tsao, but Jason Jennings.
On the one hand, Jennings has a career 5.03 ERA, and his walk total has increased in every season. On the other hand, he has made 98 starts over the last three years, and despite growing pains he has not, to the best of my knowledge, missed a single start due to injury since he was called up in August of 2001. When you factor out the intentional walks, Jennings' BB/9 peaked at 3.7 per 9 last year, which isn't terrible. His translated career ERA of 4.53 is almost exactly league average, and his WARP1 scores the last three years are 5.5, 4.1 and 4.1.
In other words, when you strip park illusions from the picture, Jennings has been a league-average innings muncher, which has a fair amount of value.
Now compare Jennings' minor league career with Francis':
Jennings: 409 IP, 8.75 H/9, 2.46 BB/9, 8.36 K/9, 0.51 HR/9, 3.83 ERA
Both reached the majors at age 23, late in their third pro season. Jennings has a marginal edge in HR/9--his greatest asset coming through the ranks was his ability to keep the ball down--but Francis has sizable edges everywhere else.
Having thought this through some more, it appears to me that Jennings has proven an elite pitching prospect can stay healthy and reasonably effective in Coors. If you were to translate Jennings' performance to sea level, I think his value over the last three years is well within the error bars of a projection made in 2001.
Based on this, I'm comfortable with the idea that, altitude or no altitude, Francis is a perfectly defensible choice as our #2 pitcher.
JS: Jason Jennings, road, '02-'04:
4.61 ERA, 306 2/3 IP, 331 H, 213 K, 130 BB.
He's not a good pitcher. He's not even league-average (I concede he's made all his starts over three years. Yay. He's 6'10" and 450 pounds). If he's the argument for ranking Francis #2, it's not a strong case.
I think, as we seem to do more and more each year, we're falling into the traps that we started out trying to avoid. Nate's input aside, there's a lot of scoutspeak and hand-waving in these discussions. We're performance analysts, and just because scouts and AGMs and PDDs will talk to us doesn't make us qualified to evaluate skills or the statements we hear.
The biggest mistakes we make come when we believe our own hype and start to think we can make judgments on things we cannot. Our core competence is performance analysis of baseball players, teams and executives, using 20 years of research and scholarship on those issues. Our list should reflect what we know about the players in the context of that information, which includes things like the insane attrition rate of young pitchers and the physics and physiology issues that affect all Rockies' pitchers.
CK: As long as you're summoning basic ideological principles to support your point of view, I'd suggest you then take the time to demonstrate what we apparently know about attrition rates with young Rockies pitchers specifically, with relevant info on how that's demonstrably different from all young pitchers. Otherwise, that looks an awful lot like fact by assertion.
RJ: I have to side with Chris on this one. The attrition rate of young pitchers is something we definitely need to be more vigilant about...but we're comparing one pitcher to another here, so that's not the issue. The issue here is, as Joe puts it, "the physics and physiology issues that affect all Rockies' pitchers," which isn't our core competency at all, actually--that sounds an awful lot like scoutspeak.
As Nate pointed out, in the absence of compelling evidence to the contrary, we have to go with the null hypothesis here: namely, that Coors Field, while a terrible park to pitch in, does not do any more damage to a pitching prospect above and beyond the standard park effects.
JS: OK. It's not unreasonable. I strongly disagree, but I'm the TNSTAAPP guy.
RJ: I do think there's some good evidence that, historically speaking, teams that play in pitcher's parks have always done a much better job of developing pitchers than the converse, even after factoring park effects into the equation. The Dodgers vs. the Cubs, for instance.
But is that enough to knock Francis down more than a peg? I don't know. If Kazmir were still a Met, I'm sure I'd be arguing to keep him ahead of Francis. And it's possible that there's a good argument for Kazmir (or against Francis) yet to be made that will convince me. But after looking at the issue as much as I can, I can't penalize Francis so much for his ballpark that he comes out below Kazmir.
JS: The Rockies have been around for 12 years. Their best young pitcher ever has 104 starts and a road ERA in the high 4.00s, with so-so peripherals. Pretty much everyone else has gotten hurt or been blasted out of the league within three years.
Maybe Francis is different, but the whole prospect-hype industry is based on the next guy being different.
Dayn Perry: I understand the injury caveats, but I think Cole Hamels (Phillies) needs to be on this list. 4.0 K/BB and, most impressively, no homers allowed as a pro. He's at least an HM. Personally, I'd probably put him in the 30-40 range.
NS: I agree with this sentiment as well. I'm convinced that the demise of Cole Hamels has been slightly overreported. He's working with Tom House, and the extent of his injury problems is probably better described "reasonably severe and handled cautiously" rather than "extremely severe." That might seem like too fine a point of distinction, but in consideration of his upside, and the fact that once you get past the top-tier names a lot of these pitching prospects have injury concerns anyway, I think some sort of ranking is justified.
Jonah Keri: Hamels' performance record merits Top 50 status. If we're looking for a tie-breaker, the genesis of his initial injury was a freak accident, not--as far as we know--criminal abuse by an overzealous pee-wee coach. He should be on there.
[Ed. note. This exchange came before Hamels' midwinter hand injury.]
RJ: Dayn, you're a Cardinals fan--what's the scouting report on Anthony Reyes?
DP: His numbers have been awesome, but he had multiple elbow problems at USC. Middle of the list sounds about right.
RJ: In the six years we've been doing this list, I think we've put a grand total of two relievers on the list (Francisco Rodriguez and Ryan Wagner.) But all three of these guys can make a legitimate case this year: Chad Orvella (Devil Rays), Jesse Crain (Twins) and Huston Street (A's).
Orvella gets the least attention, but has the sickest numbers: in 84 pro innings, he's surrendered 59 baserunners and struck out 131. I don't think I've ever seen that kind of dominance before, even in a sample size that small. Crain has been an outstanding minor-league reliever for three years and should have been pitching significant innings for the Twins in October. Street was an outstanding college closer for three years and should have been pitching significant innings for the A's in September.
All three are terrific prospects. It really comes down to: how terrific a prospect do you have to be to make a Top Prospect list as a reliever? What do people think about Orvella, especially? His scouting report is good, but not good enough to explain his numbers.
Clay Davenport: I don't think Orvella's numbers now, good as they are, are any better than Crain's were last year.
IP H R HR BB SO H/9 R/9 HR/9 BB/9 SO/9 Orvella 03-04 86.1 48 17 7 11 132 5.01 1.77 0.73 1.15 13.77 Crain 02-03 111.2 57 25 0 36 143 4.59 2.01 0.00 2.90 11.53Crain has 25 more innings, a better hit rate, and a zero home run rate. Orvella is a little better in strikeouts and walks. Offsetting that, Crain pitched farther up the ladder. 65 of his innings were at Double-A and Triple-A, compared to nine for Orvella. Counting the short season A leagues as a "1," Sally and Midwest as "2," high A as "3," Double-A as "4," and Triple-A as "5," Orvella's inning-weighted level is 2.29, Crain's is 3.42, more than a full level's difference.
I'm noticing that if you take Triple-A out of their lines, they have about the same number of total innings pitched: 84 2/3 for Orvella, 85 2/3 for Crain. Crain still has more innings at Double-A and a higher average level. Crain leads in hits allowed (33 vs 47), runs (15 to 16), home runs (0 to 6). Orvella still leads in walks (10 to 26) and strikeouts (130 to 110).
I still don't see that big of a difference between them. If Crain wasn't good enough for the list last year, then Orvella, basically a guy who blew away the Sally League, shouldn't be this year, unless the overall quality of the list has declined.
NS: Everything's gotta be just about perfect if we're going to rank a relief pitcher. I'd rank Orvella somewhere and not the other two.
CK: I thought we'd learned something from Ryan Wagner, but let's face it, optionable relief "prospects" are about as luck of the draw as they come. Skip two bad weeks--in too many situations, two bad outings--and it's back to the minors, at which point, you can be overlooked for a year because of health, new flavors of the week in minor league bullpens, or somebody somehow getting in favor, like a Vinny Chulk or Yhency Brazoban.
Orvella might make a nifty Mr. Irrelevant or HM, but philosophically, I think it isn't sensible to place groomed relievers among the fifty best prospects in organized baseball (and Jered Weaver).
RJ: There are two things to say about the relievers:
1.) All things considered, Crain may deserve to rank ahead of Orvella and Street, simply because he's been dominant for two seasons instead of one.
2.) You could make the case that none of them deserve to make the list at all.