March 19, 2008
Last week's starting pitcher rankings generated a lot of e-mail that had questions along the same lines. The main question was something to the effect of, "Why is X not in the top 50? I think he is better than Y and Z." There are a few general responses to this that I have come up with in between articles.
First, I used the Player Forecast Manager as a guide for creating the tiers of starting pitchers in order to split them up more effectively. This helped in the sense that I was organizing pitchers within different portions of the draft, rather than from one single list like the PECOTA spreadsheet. The negative that came out of this is that I stopped the list at 50 pitchers; if you look at the rankings, at a certain point, the pitchers seem like they can be switched out with another you might find later on the list without much change in value.
Somewhere within the third tier (in the twenties) the values all start to appear somewhat consistent, with a few exceptions due to fewer projected innings despite more talent. This is the reason I can think of that kept certain pitchers like Brad Penny and Chien Ming Wang out of the top 50; it isn't that they aren't top 50 pitchers, it's just that I stopped at 50, and they would have come in a bit later. The middle portion of the draft for pitchers is somewhat generic, or at least hard to differentiate between. I'm not sure how much of this is how the rankings should actually work, but that's the message I'm seeing from the PFM in regards to ranking pitchers in standard draft leagues. The second issue is just with PECOTA's regressions; again, this isn't so much a problem as it is automatically making certain pitchers lines seem cautious, or at least a step back from what we can expect from them. That helps to blend the values of the less extreme talents together.
If this seems bothersome, I'd like to hear some suggestions on how to improve the pitcher rankings for the next time around. Breaking them up into tiers with an assist from PFM did the trick in a sense, but when the pitchers value starts to blur into too much similarity, it can obviously cause some problems since the rankings only go so deep. E-mail me your thoughts so we can try to piece together a better system for the rankings using PECOTA, PFM, and our own noggins.
With that, it's time to look at the closers.
Rank Name Saves W K ERA WHIP Beta 1. Jonathan Papelbon 40 5 84 2.59 1.07 0.63 2. Francisco Rodriguez 41 5 96 2.79 1.21 0.78 3. J.J. Putz 35 4 70 2.62 1.07 0.85 4. Huston Street 42 5 89 2.52 1.02 0.83 5. Joakim Soria 36 4 81 3.16 1.19 0.88 6. Billy Wagner 35 5 67 2.82 1.16 0.71 7. Rafael Soriano 32 5 66 3.00 1.14 0.81 8. Mariano Rivera 30 5 56 2.60 1.12 0.99 9. Joe Nathan 32 4 68 2.92 1.14 0.81 10. Takashi Saito 30 4 64 2.81 1.15 0.95
Papelbon and Rodriguez are the obvious top guys for me, due to their being on winning teams, their ridiculous lines, and their reassuring Beta scores. You can't do much better than those two, though Huston Street's line certainly tries to do so. I'm not sure I believe Street is capable of throwing that line out there, and if the A's end up worse than their PECOTA projected record, you can be sure he will end up with fewer saves than that to boot. Putz is a solid alternative, since his numbers are also excellent and he'll post saves as well as Street.
Joakim Soria doesn't have the ERA forecast of the pitchers ranked in front of him, but his saves and punchouts are right there with them. Billy Wagner and Mariano Rivera are getting up there in age, but as we've seen recently, they can still get guys out at an excellent rate. PECOTA has no argument about this, posting very good seasons from both of the elder closers. Rafael Soriano, with the closer gig under his belt in Atlanta, is in line for a great campaign, though again his ERA is a bit higher than many of the top closers in the rankings.
Joe Nathan and Takashi Saito would be ranked higher if PECOTA had more faith in them to save a ton of games. They are both in the 30 range though, so they are towards the back of the top ten. You can't go wrong with any of those guys on your team though, so if you draft closers early, that's who you want to look at.
Rank Name Saves W K ERA WHIP Beta 11. Matt Capps 34 4 57 3.10 1.17 0.91 12. Jose Valverde 38 4 74 3.39 1.26 0.84 13. B.J. Ryan 28 3 60 2.98 1.19 0.71 14. Francisco Cordero 36 4 75 3.36 1.27 0.89 15. Chad Cordero 28 4 58 3.63 1.25 1.00 16. Bobby Jenks 29 3 59 3.32 1.26 0.88 17. Brad Lidge 24 4 64 3.54 1.29 0.93 18. Manuel Corpas 27 4 51 3.70 1.35 0.89 19. Eric Gagne 21 3 53 3.41 1.28 0.85 20. Bob Howry 16 4 58 3.55 1.21 0.94
Matt Capps has been a name I see in many e-mails, and he's ranked #11 due to his solid line and potential for high saves totals. Jose Valverde isn't expected to dominate like last year once he regresses in a few spots, but he's still the #1 closer for at least one team in a 12-team league. B.J. Ryan would be ranked higher if I thought he would pitch at his normal effectiveness all season, but it's to be seen how he fares returning from surgery. Francisco Cordero is moving to a park where homers may be an issue, but otherwise, his line is fantastic.
The park doesn't matter so much for Chad Cordero's homer issues, and if the Nats see a boost in park-related offense in their new venue, he's going to lose a little more value. Bobby Jenks has succeeded despite having a home park out to get him though, so he's one of the top options in the league. The list of closers in hitters parks continues with Brad Lidge, who could be in for a world of hurt in Philadelphia despite his optimistic forecast. If his issues with homers don't correct themselves soon, we'll look back on this line and wonder how it came to be.
Manny Corpas is set to replicate his success as a rookie, posting a second solid season as a closer in a row. I wouldn't bet much higher than #18 on him though, thanks to his park and the limited sample of pitching we have seen from him so far. PECOTA's Beta is reassuring, though his line isn't the best we've seen. Eric Gagne would get more love from me if I hadn't watched him tank mechanically in Boston for months after the trade deadline. If Mike Maddux can heal what ails him, then he's better than this spot. Similarly, Bob Howry would be ranked a bit higher if you could guarantee him the closer job all season in Chicago. As of now, Carlos Marmol is listed on our Depth Charts as the closer, but his forecast isn't as neat as Howry's either, and MLB.com lists Howry as the top reliever at present.
After these pitchers, we start to get into the realm of guys who do not have secure closer jobs, or are bad pitchers with lots of saves. Joe Borowski isn't expected to repeat last year, but he's one of those guys. C.J Wilson is one of the next-in-line closers, but he's expected to post a high ERA for a reliever along with an unappealing WHIP and just 27 saves. The aforementioned Marmol would be a more attractive pick if he secured the closer job in Chicago.
My own strategy is usually to buy low with closers so I don't waste a lofty draft pick on one, and then to hope for someone to snag a team's closer role as the year goes on. This isn't the best strategy for everyone, since some people aren't comfortable punting saves until they find that guy, but it works with the way I build my own team. It's also a more effective strategy if you are in a league where you know you can pull off a trade down the stretch, offering a superfluous player of your own to a team in need. Closers are necessary in some form in standard 5x5 formats, though, so figure out if drafting high or low is what works best for your team-building, and use that spot on the list to your advantage.