Baseball for the Thinking Fan

Login | Register | Feedback

btf_logo
You are here > Home > Baseball Newsstand > Baseball Primer Newsblog > Discussion
Baseball Primer Newsblog
— The Best News Links from the Baseball Newsstand

Tuesday, November 20, 2007

The Book Blog: MGL: The Best and Worst Pitchers

Amazing, Chan Ho Park saw his name on the list…and decided to dump the Dodgers.

The worst pitchers are tough to isolate if only because they tend to be in and out of the majors and they tend to change roles a lot, as would be expected of a bad pitcher.  Rather than using an IP qualifier as Tango did in his “worst pitcher in baseball” thread, I’ll just wing it.  If it looks to me like said pitcher has been around at least a little bit in the majors the last few years, I’ll include him.  If it looks like he is permanantly gone (although you never know) then I probably won’t - like Ponson. 

One more thing.  I make rough league adjustments, so the best pitcher in the NL is probably not the best in baseball.  AL pitchers are substantiallty better than NL pitchers, by I think around .25 runs per 9.  So, for example, a pitcher in the NL who is projected to be .25 runs better than the average NL pitcher is only an average pitcher in the AL and a slightly (around .125 runs) above average pitcher in the NL and AL combined.

Repoz Posted: November 20, 2007 at 02:06 AM | 25 comment(s) Login to Bookmark
  Tags: sabermetrics, special topics

Reader Comments and Retorts

Go to end of page

Statements posted here are those of our readers and do not represent the BaseballThinkFactory. Names are provided by the poster and are not verified. We ask that posters follow our submission policy. Please report any inappropriate comments.

   1. jwb Posted: November 20, 2007 at 02:21 AM (#2620706)
I thought I saw "Bayless" and wondered if mgl had crossed up his "Worst Columnists" and "Worst Relievers" lists.
   2. Amit Posted: November 20, 2007 at 02:28 AM (#2620712)
Mota dodged that bullet.
   3. JPWF13 Posted: November 20, 2007 at 04:14 AM (#2620802)
I think Sabathia got jobbed
   4. Jack Keefe Posted: November 20, 2007 at 04:32 AM (#2620827)
Well Al there is this stathead MGL and his spread sheet says that the worst reliefers is Borowski, Meese, Venereafro, Childless, Herman F. German, Skip Bayless, and Gwin Kolb now note who is not on the List or even close Al it is a red leather day after 4 yrs. in the Bigs and nearly to my Abritation Years I have finely got off the list of Worst Pitchers and this means there is nowhere to go but up or perhaps back down Al.
   5. mgl Posted: November 20, 2007 at 04:38 AM (#2620839)
Sabathia would have easily made the list if it were "Worst performances in an LCS." And Bayless definitely would make MY list of worst columnists/commentators/etc., although that list would be very long with lots of ties. In fact, there might be 50 or 100 tied for worst.

Before anyone jumps down my throat for any or all of the selections, they are based on my (computer generated) pitcher projections which are probably not much worse or better that any of several good ones out there (ZIPS, CHONE, PECOTA, Shandler, etc.). The projections are based on raw component stats over the last 4 years, adjusted for park, defense, and opponents.

Although the lists are somewhat in order from best to worst on the "best" lists and worst to best on the "worst" lists, they are all close enough such that on each list, one pitcher is probably just as good as another. As well, there are plenty of pitchers who didn't quite make the arbitrary cutoff point on each list but certainly could have been included. Given the innaccuracy in pitcher projections in the first place (although they really are not that bad), the difference between a pitcher and those few above and below him on an ordered list are really not that discernible.

As most everyone is focused on "What have you done for me lately?", most people's perception of whether a pitcher is good or bad is going to be heavily colored by this year's performance. In many cases, a pitcher's projection, and hence how good he "is," is going to be quite different from how well he pitched this year. Not to mention the fact that there can also be a wide discrepancy between ERA (and ERA+) and other garbage stats, and some kind of ERC (component ERA - or an "ERA" generated from component stats). This is especially true with relievers, simply because of the small number of IP per season they throw.
   6. David Concepcion de la Desviacion Estandar (Dan R) Posted: November 20, 2007 at 04:42 AM (#2620846)
Sabathia does seem like the missing name there--great peripherals for multiple years running in the tough league.
   7. JPWF13 Posted: November 20, 2007 at 04:58 AM (#2620868)
RE; Harden & Halladay especially Harden-
as you said:

I don’t project playing time (other than classifying a pitcher as starter, reliever, or both/either) and I don’t go beyond next year, so my “best” and “worst” players are always by “rate” and don’t consider longevity beyond next year, chance of getting injured, etc.



I can't see either man as being better than CC because of the durability issue
Snarky answer aside, was the postseason the reason you rank Beckett higher than CC? Or is it based solely on regular season?
   8. Danny Posted: November 20, 2007 at 05:12 AM (#2620878)
MGL takes quality of opposition into account. Sabathia threw nearly 40% of his IP this year against (arguably) the 3 worst offenses in the league (MIN, KCR, CHW). That could be part of his absence.
   9. Boots Day Posted: November 20, 2007 at 05:21 AM (#2620889)
Mark Redman surely deserves to be on this list, but it's worth noting that Clint Hurdle got a 3.20 ERA out of him in three starts (and two relief appearances) in September, after he'd posted an ERA over 11 with the Braves. Redman was on a very short leash for the Rockies, never going more than five innings even in the game when he just allowed a single unearned run.
   10. greenback calls it soccer Posted: November 20, 2007 at 05:43 AM (#2620907)
Mulder probably shouldn't be on that list, unless there are indicator variables for shoulder surgery and the like. His results reflect a talent level that, for better and for worse, does not exist now.
   11. mgl Posted: November 20, 2007 at 06:08 AM (#2620926)
I have no idea what the current talent level of Mulder is. It is VERY difficult to project pitchers who have had surgery, multiple injuries, etc., without at least some scouting reports, pitch f/x data, etc. Even then, who knows. My model just looks at performance blindly, as if every pitcher was never hurt. It does, however, make some kind of adjustment (I forgot exactly how) for days spent on the DL over the last 4 years.

As I said, I am not including durability, only rate. So sure, if pitcher A is expected to throw 220 IP and pitcher B, 180, even if pitcher A is a little worse than pitcher B, he is probably "worth" more. That is obvious.

I don't use post-season stats in the projections, although I suppose I could, as long as quality of opposition is accounted for, which it is in my model.

Good for Hurdle that he never let Redman go for more than 5 even when he was pitching a good game. A bad pitcher should pitch as few innings as possible per game and then a reliever should be brought in. Unless the game is very low leverage early (a blowout), in which case it does not matter what you do and you might as well save your pen. Since how a pitcher is going in a game has little predictive ability as far as how he will do subsequently (see our research in The Book), with some caveats and qualifications, a reliever, any reliever, is ALWAYS going to be better than a bad starter after the starter has thrown to the lineup a couple of times or more.

As far as the fact that "Hurdle got a 3.20 ERA out of him in 3 starts," the likelihood that Hurdle or anyone else (other than Redman himself) had anything significant to do with that, as compared to the ease at which ANY pitcher can post a 3.20 ERA (or 6.20) in 15 ip, is quite small.

I am not a big Sabathia fan. Unfortunately that is heavily colored by his poor post-season performance, both physically and mentally (approach), at least versus the Red Sox. Fortunately, my projection model is 100% objective so it doesn't matter what I think.

Sabathia is 22 on my list among starters. Some on the list above him almost for sure do not belong there or are retired or hurt. Reasonably, he is 14th or 15th, with Lackey, Haren, Escobar, and Oswalt just ahead of him. That is without league adjustments. (Those league adjustments really screw things up. I wish we would get parity between the leagues!)

If we adjust for league, we can probably remove Oswalt and Harang, who are above him, and maybe Sheets. So that would put him at around 12th to 13th on the list, below:

Peavy, Jake
Webb, Brandon
Santana, Johan
Carpenter, Chris
Harden, Rich
Smoltz, John
Sheets, Ben
Halladay, Roy
Bedard, Erik
Beckett, Josh
Escobar, Kelvim
Haren, Danny
Lackey, John

Sabathia

Carpenter and Harden are questionable because of injury, especially Carp. In fact, we can probably remove him from the list altogether. I'm not even sure Sheets is 100% healthy.
   12. RJ in TO Posted: November 20, 2007 at 06:11 AM (#2620927)
I can't see either man as being better than CC because of the durability issue


Okay, I can see the Harden issue, but Halladay has 4 top 5 finishes in Innings Pitched in the last 6 years. On one of the other years, he was piling up a ton of innings (over 7 per start) before he took a liner off his leg. Only in one of the last 6 years did he miss a significant amount of time with a non-freak injury.

Hell, if it wasn't for appendicitis, he probably would have thrown more innings than CC this year too. As it was, he was only able to get up to third in the league.

I really can't understand where the "Halladay isn't durable" myth came from.
   13. RoyalsRetro (AG#1F) Posted: November 20, 2007 at 06:23 AM (#2620937)
Mark Redman surely deserves to be on this list, but it's worth noting that Clint Hurdle got a 3.20 ERA out of him in three starts (and two relief appearances) in September, after he'd posted an ERA over 11 with the Braves.

That's why he deserved "Manager of the Year"!!!

I thought Franklyn German might be a decent minor league FA pickup...maybe not.
   14. S. Ransom Posted: November 20, 2007 at 08:25 AM (#2620980)
MGL - As much as I am an admirer of your work, any worst pitchers list that does not include a single Tampa Bay Ellipses Ray - particularly on the bullpen side - is deficient at best. Your league adjustment is too kind to a team that had five (five!) relievers pitch more than 40 innings with ERAs above 7.00. Casey Fossum allowed 109 hits and 15 HRs in 76 innings. Shouldn't that get at least Honorable Mention? Maybe we should have a group prize - kind of like how the entire newspaper gets the Public Service Pulitzer. You can award a bullpen disaster prize to the D-Rays' bullpen for a historic collection of ineptitude below and beyond the call of duty.
   15. larkin4HoF Posted: November 20, 2007 at 09:57 AM (#2620994)
The question I have is whether the Reds staff was so bad that they were ineligible-I can't imagine any other way that Matt Belisle and Todd Coffey did not make this list.
   16. zoobird Posted: November 20, 2007 at 01:28 PM (#2621022)
I play in a fantasy baseball league with daily transactions and a stock market trading mechanism that made it possible to use hitters with especially easy matchups each day. Some of the pitchers I especially looked for (based on low K/9, high BB/9. and low gb% were:

Speigner
Trachsel
Dumatrait
De Salvo
L. Hernandez

Thoughts on where these guys stand among the worst mlb pitchers?
   17. zoobird Posted: November 20, 2007 at 06:39 PM (#2621393)
15 - Keep in mind that the Reds pitchers are hurt a ton by their home park.
   18. mgl Posted: November 20, 2007 at 07:40 PM (#2621455)
#14 and #15, your opinions are colored by two things: One, this year's performance and ERA, neither of which, in and of themselves, are in my model. If I were listing the worst performances of this year, then maybe some of the Reds and/or TB pitchers would make the list. But I am not. I am listing those pitchers whose projections for next year are the best and worst, and that involves "crunching" the raw stats of pitchers from the last 4 years. Plus, one "reason" for the Ray's (and to some extent, Reds') poor pitching this year was their wretched defense. Defense is factored OUT in my model. That being said, Fossum is a pretty bad pitcher in general. He could be given an honorable mention.
   19. mgl Posted: November 20, 2007 at 07:47 PM (#2621468)
#16, it depends on how your fantasy league works (what you get points for). Most fantasy leagues give points for wins, K's, WHIP, ERA, saves, etc., with no context adjustment of course (you want pitchers who pitch in hitters' parks, who are on good teams, etc.). That is not at all commensurate with my "overall" context neutral ratings. Anyway, you picked some real winners there. Some of them have limited major league experience of course, so you are better off looking at scouting and prospect reports. My ratings are awful for all of those pitchers, especially Dumatrait and Speigner. The rest are bad too, although Livan seems to be one of those pitchers who has a skill at outperforming his ERC (component ERA).
   20. Amit Posted: November 20, 2007 at 07:53 PM (#2621479)
mgl - is the same true of trachsel? (i.e., skill at outperforming ERC)
   21. zoobird Posted: November 20, 2007 at 07:58 PM (#2621487)
19 - Yeah I'm pretty sure that Dumatrait and Speigner are the type of pitcher that makes a mess of any statistical analysis of major league talent that assumes a normal distribution, because they simply don't belong in the majors. If I had to guess, they're both probably about average for high-A ball. Dumatrait at home in Cincinnati was an especially tasty match-up...pretty much any opposing hitter with any power at all was worth using against him.
   22. Russ Posted: November 20, 2007 at 08:04 PM (#2621499)
Carpenter and Harden are questionable because of injury, especially Carp. In fact, we can probably remove him from the list altogether. I'm not even sure Sheets is 100% healthy.


I think one thing that a projection list like this misses is the uncertainty in the prediction (one thing that, whether or not they get it right, is very good about what BPro puts out there). I would guess that a guy like Haren or Harden may have a better mean (or average) predicted season, but the variability would be much larger than Sabathia's (who is extremely consistent). I would venture that the range of Sabathia's optimistic and pessimistic ZiPS projections would be much smaller than Haren's or Harden's (if only due to career IP). I think that lack of uncertainty is what biases our minds to thinking that Sabathia should be higher on the list. He could still have a worse mean projection than Haren and Harden, but still be a better pitcher to have on your team. As MGL has said repeatedly, you have to be sure you clearly state what your criterion is (or critieria are). This is a "best projection" list, not a "best bet for success" list.
   23. mgl Posted: November 21, 2007 at 05:13 AM (#2622271)
I don't know that uncertainty really matters, other than perhaps what the philsophy of a team is. It doesn't really change a player's value. If you ask me whether I would rather have a million dollars or a 50% chance at 3 mil (and 50% nothing), I'll take the latter. Most people would take the former. It is a personal decision which also objectively depends on lots of things. For example, with the money offer, if you had none it is probably "correct" to take the guaranteed mil. If you had lots, what's another million? Etc.

Some teams might prefer a player with a mean value of 3 WAR but a 20% chance of being a bust. Another team might prefer a player with a guranteed 2.5 WAR.

I also don't think that it is very easy to measure volatility in a projection other than as a function of the amount of past history (the less the history, the less certain we are of our projection, even if the mean is good). I think that by and large the Pecota "bands" are nonsense. There is so much random fluctuation in a pitcher's year to year performance that is kind of swamps what we like to think of as uncertainty due to health, etc. I am talking about performance and not playing time. And even with playing time, I am sure that Sabathia, Buehrle, and guys like that will have a higher expected mean IP in the future, but I don't think the uncertainty around that is much different than for a pitcher with a history of injury, and hence a lower mean playing time projection.

I don't know that Trachsel has that skill. I have never heard it said about him. And I doubt that there is much of that kind of skill among pitchers in general.

I don't know much about the low minors, but guys like Dumatrit and Speigner are WAY better than high A. Any pitcher who gets a cup of coffee in the majors, no matter how bad, is usually an average AAA pitcher at least.
   24. Russ Posted: November 21, 2007 at 05:26 AM (#2622284)
For example, with the money offer, if you had none it is probably "correct" to take the guaranteed mil. If you had lots, what's another million? Etc.


This is exactly the kind of thing that I'm talking about. For a good team, I can often see the benefit to trading off a bit of mean projection to have more certainty for a certain "minimal" level of production. Think of it as hedging risk... you want to have some solid performers in the portfolio to keep the bottom from absolutely falling out. You can do that in a couple of ways... you can have a lot of risky performers (let's say a bunch of AAAA types) and then just hold on to the ones who work out for that year. Or you can have a significantly more expensive, mediocre vet who will lock in a certain level of performance.

Someone said it perfectly in another thread... when your team is good, variance is bad. When your team is bad, variance is good. What's neat in baseball is the playoff threshold. Basically there isn't much difference between running a 65 win team and a 75 win team in terms of revenue. But there's a huge difference between an 80 win team and a 90 win team. Each one of those wins between 80 and 90 has a lot of marginal revenue attached to it. It may be that you reduce your optimistic wins from 98 to 95 in order to shrink your pessimistic wins from 82 to 85 by going with Mr. Joe Veteran at first base rather than Mr. Could Be Cecil Fielder, Could be Jeremy Giambi (an exaggeration, but you catch my drift).
   25. zoobird Posted: November 21, 2007 at 09:37 PM (#2622954)
I don't know much about the low minors, but guys like Dumatrit and Speigner are WAY better than high A. Any pitcher who gets a cup of coffee in the majors, no matter how bad, is usually an average AAA pitcher at least.


I just looked at their minor league stats, and you're right - they're mediocre, but not as bad as I thought I remembered.

Another two more of my favorites that I forgot - Joel Hanrahan and Matt Chico. Washington really had a lot of bad starting pitchers this year between those two and Speigner.

You must be Registered and Logged In to post comments.

 

 

<< Back to main

BBTF Partner

Support BBTF

donate

Thanks to
Harveys Wallbangers
for his generous support.

Bookmarks

You must be logged in to view your Bookmarks.

Hot Topics

NewsblogOTP Politics November 2014: Mets Deny Bias in Ticket Official’s Firing
(4218 - 10:57pm, Nov 23)
Last: JE (Jason)

NewsblogMatthews: Cashman sleeps on the street, says all is quiet on the free-agent front
(24 - 10:50pm, Nov 23)
Last: JE (Jason)

NewsblogKemp drawing interest, raising chance he's the Dodgers OF dealt - CBSSports.com
(26 - 10:48pm, Nov 23)
Last: Pops Freshenmeyer

NewsblogOT: Monthly NBA Thread - November 2014
(994 - 10:39pm, Nov 23)
Last: Howling John Shade

NewsblogAstros interested in Robertson: source | New York Post
(17 - 10:37pm, Nov 23)
Last: McCoy

NewsblogBraves shopping Justin Upton at a steep price | New York Post
(35 - 10:34pm, Nov 23)
Last: Rickey! trades in sheep and threats

NewsblogOT: NFL/NHL thread
(8649 - 10:26pm, Nov 23)
Last: AuntBea

NewsblogPirates DFA Ike Davis, clear path for Pedro Alvarez - Pittsburgh Post-Gazette
(21 - 10:13pm, Nov 23)
Last: zonk

NewsblogRed Sox trying for mega-free agent double play: Panda and Hanley - CBSSports.com
(13 - 9:45pm, Nov 23)
Last: DCA

Hall of MeritMost Meritorious Player: 2014 Ballot
(10 - 9:09pm, Nov 23)
Last: MrC

NewsblogPablo Sandoval leaning toward Red Sox, to decide next week — Padres have highest offer, all offers on table (including SF Giants’) - John Shea
(32 - 9:04pm, Nov 23)
Last: the Hugh Jorgan returns

NewsblogMike Schmidt: Marlins' Stanton too rich too early? | www.palmbeachpost.com
(31 - 6:53pm, Nov 23)
Last: cardsfanboy

NewsblogOT:  Soccer (the Round, True Football), November 2014
(451 - 6:43pm, Nov 23)
Last: JuanGone..except1game

NewsblogCashman in wait-and-see mode on retooling Yanks | yankees.com
(22 - 6:14pm, Nov 23)
Last: ReggieThomasLives

NewsblogSunday Notes: Arroyo’s Rehab, Clark & the MLBPA, Doc Gooden, AFL Arms, ChiSox, more
(16 - 5:53pm, Nov 23)
Last: ReggieThomasLives

Page rendered in 0.4185 seconds
52 querie(s) executed