Dobbs and BP Premium articles
Two quick hits:
* If you can, please help Dave with his computer problem. He will need the peace of mind after he sees John Levesque’s latest effort, a love letter to Greg Dobbs, who has “intensity” and is “feisty.”
After reading his column about Mariners clubhouse guy Ted Walsh, there seemed cause for hope that Levesque would put together what Ron Fairly might call a “two game writing streak.” Perhaps he had even read some wise counsel about stepping off the well-trodden road for ideas.
But no, we’re back to elevating the gritty, bad player. Pity.
* Interested in the Mariners’ prospects for injury? Of course you are. Heard of Will Carroll? Of course you have. Subscribe to Baseball Prospectus? Maybe you do. Meet all three criteria? Check out Will’s take on the M’s, especially the Aaron Sele zinger.
* While I’m linking to BP Premium content, scope Nate Silver’s PECOTA system’s projections for the American League. The system expects the A’s to win the division by five games over the Angels, with the M’s finishing last, just behind Texas.
Don’t worry: my spreadsheet can beat up Nate’s spreadsheet.
Comments
35 Responses to “Dobbs and BP Premium articles”
But apparently your spreadsheet can’t count to three.
I kid because I care.
I’ve often wondered why everyone uses spreadsheets? I’ve actually been thinking of trying to do statistical analysis as a hobby because I need some new coding projects. Of course every where I see people using excel. I want to shy away from that. Does anyone keep their data in either a flat file, xml file, or database and write actual software to analyze it? Another reason I ask is because I can’t run microsoft windows, so excel macro’s aren’t going to cut it for me.
Clearly, Dave, you don’t understand nonlinear thinking. I know baseball seems like a linear game with all those lines and box scores, but the fact is, it’s got a spacious non-time kind of time to it.
Either that or I changed the way I organized the post in the middle. I leave that up to the reader.
Oh, I figured your plea for helping me was what caused the two quick hits to become three, but after commenting that you loved the Sele quip, I couldn’t resist.
I understand that some content is to be paid for, but is the purpose of USSM really just to give links without discussion of content that would be great if you subscribed to it? Just asking…
Have you ever come here before? USSM does an incredible job of discussion, and gives us an incredible amount of free information.
I think our ratio of discussion to premium linking is something like 8,000 to 1. I’m okay with that.
Here’s one to ruminate over: If you had to choose, would you rather have Bloomquist or Dobbs in that 25th spot on the roster?
(Don’t ruminate too long unless you have prescription-strength Maalox within arm’s reach.)
I think I’d go with Dobbs, but that’s because I think a left-handed no hit weak glove is better than a right-handed no hit weak glove, and that may be sabermetrically wrong.
I wish there was someone healthy out there who could satisfy the multiple positions, replacement-level left-hander requirements. Where have you gone, Denny Hocking?
DW, as much as it pains me to say it, Bloomquist. Unless Lopez wins the starting SS job, Bloomquist’s suckiness as a glove in the middle infield >>> Dobbs’ suckiness as a glove in the middle infield. Plus, Dobbs’ lefthandedness might actually entice Hargrove to pinch hit him for a decent right hander at some point. Hopefully Bloomquist’s right handed non-stick won’t have that allure.
Bloomquist in a landslide. Dobbs had a .286 OBP against Triple-A pitching last year. Bloomquist’s major league OBP is .319. Dobbs isn’t just worse defensively, he’s also worse offensively, as hard as that is to imagine.
I agree with you DW that Dobbs appears to have more upside than Willie. It would also appear that the issue is moot unless Lopez wins the shortstop job and Pokey goes to the bench. If that happens, Bloomquist is extraneous (although not exactly redundant—the Mariners would have to sign the recently released Joe McEwing to accomplish that).
What is Levesque smoking? – “Dobbs’ greatest selling point is that he’s a left-handed hitter with power. In recent years, the Mariners have had fewer of those than timely hits with runners in scoring position”. Is there any statistical evidence to support thid claim? I thought the knock on Dobbs was that his high average blinded the team to his impatience and utter lack of power.
One homer in the bigs and two in spring training do not make Dobbs into Mike Schmidt.
But Dave: Look at that swing! That gritty, feisty, swing!
Sorry for coming out so strong and making you have to edit my earlier post. My bad.
I’ve often wondered why everyone uses spreadsheets?…Does anyone keep their data in either a flat file, xml file, or database and write actual software to analyze it?
Because (a) spreadsheets are “actual software”, (b) most of us already have a spreadsheet program or access to one, and (c) most of us who can program would rather spend our time working on the formulas, not finding the bugs in our programs (particularly in the reinvented wheel parts of parsing files, computing common stat functions, presenting output, etc).
If you really want to get hardcore and use tools that allow you to sneer at mere spreadsheet users, I’d suggest [url=http://www.mathworks.com/products/matlab/]Matlab[/url] (which is available for just about any platform you’re likely to be using). Of your could try your hand at [url=http://www.wolfram.com/products/mathematica/index.html]Mathmatica[/url], but statistics aren’t really its strength.
#10–
You’ve picked the one bad half-season in Dobbs entire minor league career. Otherwise his numbers look pretty darn good. And maybe we should cut him some slack for his numbers the year after his recovery from a ruptured appendix–oops, I mean Achilles.
Right now he’s leading the team in Spring Training ISO, HR, and RBI. If he continues to hit like that for 2 more weeks, it’s hard to imagine leaving him off a bench with absolutely no pop otherwise. If he doesn’t continue to hit, he’s gone. So what that you don’t want him in the field–that’s what those other (no-hit-um) bench guys are for. What’s the downside?
Take out those two homeruns and all of a sudden those numbers aren’t nearly as pretty. There’s no way to prove they aren’t a fluke. Actually, there’s a pretty good argument that they are. Dobbs has established his level of ability, a small sample size against questionable competition in spring training should do nothing to change our opinion of him, even with another two strong weeks. And I think Spiezio will bring more power off the bench than Dobbs will, although we’re obviously speaking in relative terms.
I forgot to add this. If you want to cut him slack for coming off an injury that’s fine. But the biggest problem with Greg’s performance in triple-A last year was his utter lack of plate discipline. That can’t be blamed on an Achilles.
#16: Dobbs’ other numbers that you say look “pretty darn good” are a result of being at least 3 years older than his league’s average competition his entire minor-league career. He tore up short-season ball as a 23-year-old, was average in low-A as a 24-year-old, and hit for empty average in AA as a 26-year-old before proving to be pretty much completely inadequate in AAA the same year.
If we’re impressed by Dobbs’ team lead in HRs (2 to everyone else’s 1) and ISO in all of 24 ABs, it should also be pointed out that sweet-swingin Ryan Franklin is third on the team with a blazin-hot .500 BA. Maybe we should use him in late-game pinch-hit situations! And if we wanna get technical, the team lead in ISO actually belongs to some dude named G Harris, who has a ridiculous .600 ISO (3-for-3 with a triple).
More to the point, Dobbs hit .275/.338/.431 in the Midwest League at age 24, and then had a great month in AA to finish the season. So it’s more like he’s had one great month, two good half-seasons (counting the short-season NWL in 2001 as a half), three-quarters of a mediocre season, and a miserable half-season (Tacoma last year), being old for his league all the way through.
Incidentally, his walk rate wasn’t all that bad before 2004. Perhaps he switched to a swing-early-in-the-count approach last year? I don’t know.
In any case, I agree that Bloomquist would be preferable if that were the option. More likely, it’s Dobbs vs. Strong, but Bloomquist a lock. But I don’t see Dobbs offering much this year that Spiezio doesn’t; Spiezio’s trying to prove that he can come back from whatever messed him up last year, but Dobbs is trying to prove that he can hit above the AA level.
If Dobbs puts up an OBP over .350 in (preferably well over it) in AAA this year, then I won’t mind taking him seriously as a role player for the next few years. But not until then.
For a somewhat less snarky comment: I just noticed that Bloomquist also has a higher ISO this spring than Dobbs, in pretty much the same number of at-bats. I don’t have a calculator on me, but Dobbs’ ISO is 5/24 and Bloomquist’s is 7/22. A math major I ain’t, but whatever those numbers turn out to be, Bloomquist’s will be higher.
I know it’s fun to read things into Spring Training performances. We’ve all been suffering from a baseball drought for 6 months now, but these stats mean absolutely nothing.
Stop the presses – Greg Dobbs has hit the AP wire!! For those of you that think John Levesque is goofy, checkout this nugget of wisdom; which, after reading it, is suspiciously similar to Levesque’s drivel.
http://story.news.yahoo.com/news?tmpl=story&ncid=1362&e=9&u=/ap/20050321/ap_on_sp_ba_ne/bba_mariners_dobbs&sid=95747885
From the Will Carroll article linked above:
(A) “Signing [Aaron Sele] is like buying an old Fiat just in case your Yugo breaks down.”
😀
(B) “The Mariners may be best known, healthwise, for being armshredders. […] Lots of teams lose young pitchers in development, but no one else does it so consistently.”
Ouch. If Joel and Rafael (several years older than Felix) could suffer moderate-to-serious arm injuries out of the blue, what are the chances of Felix remaining healthy at his tender age?
It appears to me that the interest in Dobbs has to do with people in the Mariners FO who place a premium on tools and scoutability and don’t give much weight to minor league. Dobbs looks like he should be a great player.
The organization looked on him favorably before his injury. Leone is perceived to have proven himself not ready for the blg leagues, so the natural pecking order is restored.
Dobbs draws the same attention and emphasis from the same places in the organziation that Bloomquist does. They’re both mediocre, at best players, who make a good impression with hustle, etc. They look like what ballplayers look like.
For as much as we’ve lauded some improvements the front office has made in the past year or so, the Dobbs lovefest is a basic reminder that the organization still has a pretty big blindspot for certain types of lousy players.
He’s getting so much ink lately that he probably deserves his own post, but make no mistake, Greg Dobbs is a lousy player. His absolute best case scenario is Shea Hillenbrand or Randall Simno. That’s if everything goes perfectly and he maximizes his potential.
Yech.
As far as Pecota’s projections are concerned, it’s worth keeping in mind that Pecota has a very low (in my opinion) projection for Adrian Beltre because Pecota is incapable of differentiating between a healthy active player and a hobbled active player: all Pecota cares about is whether you’re playing or not. If you think it’s reasonable to suppose that Beltre’s productivity was harmed by his appendectomy recovery — and I think it’s reasonable to suppose that — then Pecota is underestimating Beltre’s likely productivity, and possibly by a lot.
But the season will show. I’d be surprised if the Mariners finished in first or last.
I’ll have you know I was once picked second-to-last in kickball.
Two comments deleted: name-calling.
re: #19
Given the FO’s way of thinkin’, maybe Gary Harris is the man who can hold down the fourth outfield spot. If not him [Harris], then Carlos Arroyo could sure give it the ol’ college try.
Now, with the above said, this is what the bench should look like:
C Dan Wilson
Corner Infielder Scott Spiezio
Middle Infielder Calvin Reese (with Jose Lopez starting at shortstop)
Outfielder Jamal Strong
Designated Hitter Bucky Jacobsen (or Justin Leone, ’til Jacobsen is no longer injured)
Furthermore, this is what the bullpen ought to look like:
Lefty Specialist George Sherrill (Matt Thornton should be DFA’d)
Long Reliever Ron Villone (ugh!)
Middle Reliever Scott Atchison (let Julio Mateo work out his problems in Tacoma)
Middle Reliever Shigetoshi Hasegawa (ugh II!)
Setup Man J.J. Putz
Closing Pitcher Eddie Guardado (if he’s still hurt at the start of the season, then let Mateo stay at the Show)
Also, how long until Rafael Soriano and Chris Snelling are placed on the 60-Day disabled list? Hopefully, their being put there won’t be the lead-in to Aaron Sele and Jeff Nelson sewing up their spots on the roster.
I think you meant “Pokey Reese” as the backup MI
Hmm…maybe I meant Ronald Villone at long reliever; it’s all ‘nominal’. But yeah, I’m amazed at the blunt nature of Jeff Heaverlo; I wish all people — including athletes — were that honest and forthcoming.
Andrew:
I’m not sure what you are looking for. E.g., I posted the Marcel forecasts at my site, and they are in csv format. Retrosheet has play-by-play and game-by-game logs, and they are also in csv format. Everything in HTML can be parsed. What exactly have you come across (other than those Excel spreadsheets) that you couldn’t grab from a non-MSFT platform? ( And, my guess is that free stuff like Star Office would read in MSFT products.)
I agree that I would prefer not to code, and let database engines and spreadsheets do as much work as possible. Debugging is one pain in the a$$.
It’s not finding the data that’s the problem. I find I can pretty easily parse all the data I can find. It just seems that I’m going to have to reinvent the wheel in calculating basic stats because all the formulas are locked in spreadsheets. I’m not a big fan of matlab (as I had to use that in college). Plus I run linux full time, and while open office does a pretty good job of reading excel files, it kind of dies when it tries to load complex formulas.
Ideally, I’d like to write python script that goes out to retrosheet or somewhere else and inserts all the data into an SQL compliant database. From there I hadn’t decided what I’ll code the actual interesting calculations in. But for me, coming from a programing background, it’s actually faster to do things in a rapid prototype language like python than to work in excel or matlab. It also has the added benifit that I can set any calculation to fire from a website. Of course then I can GPL the whole thing. If anyone else wants it, they can download the code and an SQL dump to recreate the database.
What I don’t want to do, is too much duplication of other peoples work. If there are open programs that people use to do statisitcal calculations from a database, then I would prefer to start there rather than recreate everything. At worst, I’d like to recreate their database schema so our code would be compatible.
In your case then, Andrew, you are best to join the yahoo group baseball-databank, and set your plan in motion. Tell the group that you want to capture SQL views/queries against the database, so that that’s always consistent. BA, SLG, OBP, LWTS, RC, etc. And then others can include their own calculations as well. I for one have a pitch count estimator which I could include as a SQL view.
Yeah, but Nate can kick your ass at the 15/30 table 😉