Orioles - Promoted Wieters
Matt Wieters - Granted the Baltimore Orioles the right to take the field with him.
Not since Mike Mussina’s debut has there been an Orioles call-up that I have been more excited about. Hopefully, the success of Wieters will encourage the Orioles to remain aggressive in the draft. For those that don’t remember, Wieters was minutes from returning to college, but the Orioles came to the wise decision that they should thank every possible deity that the Pirates chose to draft Daniel Moskos for no particular reason and up their offer for the tip-top prospect Wieters by the equivalent of two months of Danys Baez’s salary.
Wieters was clearly ready at the start of the season and even a not-ready Wieters would have been extremely hard-pressed to not obliterate the 210/304/327 that Gregg Zaun and Chad Moeller have combined to put up for the O’s behind the plate. For delaying the arrival of Wieters, it is not known whether the O’s former catching tandem will be allowed to open up their wrists in a warm bath or if they’re going to be bowstringed later after Wieters’s victory procession.
All kidding aside, Wieters is a tremendous hitting prospect, the first the Orioles have had in a really long time. Much has been made of PECOTA’s very aggressive projection for Wieters this season, but who can blame it? ZiPS likes Wieters a lot, too almost to the degree that my old 386 PC had for Civilization 15 years ago. The amazing thing is that the 305/387/504 that Wieters has put up in AAA this season was accomplished one of the most difficult hitting environments in the upper-minors. Harbor Park has been terrible for power-hitters in recent years, with HR park factors the last 3 years of 60, 60, and 72 (by comparison, Petco’s weighted factor over the last 3 years is 76).
It remains to be seen how successful Baltimore’s strategy of trying to keep Wieters from being a Super-Two arbitrationee will be. After all, the more teams that try to game the eligibility for that extra year of arbitration, the later in the season a quality prospect will have to be brought up to avoid that status.
2009 ZIPS Projection - Matt Wieters
——————————————————————————————————————
AB R H 2B 3B HR RBI BB SO SB BA OBP SLG
——————————————————————————————————————
Year-to-Date 143 23 39 7 1 5 25 16 29 0 .273 .342 .441
Rest-of-Year 349 50 100 13 1 15 52 40 63 1 .288 .360 .462
——————————————————————————————————————
MLB Total 349 50 100 13 1 15 52 40 63 1 .288 .360 .462
Dan Szymborski
Posted: May 29, 2009 at 07:50 PM |
61 comment(s)
Login to Bookmark
Related News:
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. CraigK Posted: May 29, 2009 at 08:41 PM (#3199158)Hey, it can't be any more off than the Ortiz projection. :)
Are you sure? It could be like the Lastings Milledge projection.
Between those two, I think we've ensured that Dan will never use ZIPS like that again.
For a catcher, ZIPS line of 288 .360 .462 is pretty impressive. Last year, as a group, catchers hit .257 .325 .390. Weiters line is roughly that of an All Star.
Pretty much a Russ Martin line.
He's 0-2 right now, but Matt Wieters is catching a shutout.
Did the O's trade him?
The world has ended.
Well, in Dan's defense, he was projecting Wieters on only 550 PA, all in the Minors to boot, which adds additional uncertainty - forecasting is tough enough when you have two or three times that much data to work with.
How about giving him an actual full season w/ the big club before jumping to conclusions?
Well, I think it's pretty clear he's not Pecota-level, best C in the league good. No right now. Maybe in a few years.
Or moves to 3B.
What conclusion? I don't have a conclusion.
He has 500 minor leagues PAs that say he is good
300 or so MLB PAs that say "eh"
I haven't seen him play.
BUT weren't all those people calling him the next Johnny Bench jumping to conclusions in the first place?
He's a "Jump to Conclusions" Matt.
I'm sorry. I'll go.
Including those that will criticize any outlier projection, and its failure, as an indictment of the projection system as a whole. We're all suspicious of those outliers; most of us don't bother using them as a stick to beat someone with. For months on end.
PECOTA's projection was ridiculous, and I think people were right to criticize it. I'm pretty sure they screwed up their translations. The other projection systems were too high on Wieters as well, but it's still not crazy to expect Wieters to end up as an 800ish OPS hitter. I doubt he'll ever sniff PECOTA's projection.
I know this is anecdotal so far at this point, but at the end of the season we can (and I probably will) look at all Eastern League players from last year who played in the majors this year, and I think you'll see the same pattern bear itself out.
This assumes this is about Colin: For what it's worth, and as much fun as I make of him over it, I thought Colin did brilliantly with this. He thought it was wrong, he asked questions first of others around (a couple of Lounge diversions), then of those more and more in the know, addressed direct questions to the principals, and then did his own examination and discovered a systematic issue. I don't know what more one could ask for.
See #31, actually there was an article somewhere (Hardball Times?) that said that BPro's translations for two leagues EL and the Carolina League (both Wieter's leagues) seemed way off- league difficulty was messed up- EL seemed to be as difficult as the PCL/IL, based on their translations, and the Carolina League was as difficult as the SL and the TxL...
Wieters has underperformed other projections, but Pecota was seemingly based on bad translations to begin with- so they massively overshot...
Wieters was a "perfect storm" sort of case (because he had ONLY played in those leagues) but there are dozens of projections affected by these issues to some extent. As noted, I intend to do a followup for THT after the season.
See #31, actually there was an article somewhere (Hardball Times?)
Written by another Plaschke, Mariotti, Justice, Chass-style old-school know-nothing hack, no doubt.
You two friends?
Colin tolerates me. If you call that being friends, and I do because I am desperate for human contact in any form, then he is my best friend ever.
did you ever hear from them?
To drag out the beating stick one more time:
- Steve Goldman, 3/3/09
- Steve Goldman, 9/13/09
I remember back in the 1980s, Bill James made a monumental screw up in how he calculated OWP (offensive winning percentage). OWP was a great idea- but he botched the execution becasue he took a shortcut, he based the denominator not on a league/park neutral number (runs/game) but on the actual r/g in that player's games (he reasoned that since it included both home and away games...)
He then wrote an article claiming that Seitzer was a better hitter (in 1987) than McGwire. The conclusion was absurd, McGwire's OPS was higher, his RC/27 oputs was higher, and McGwire played in a pitcher's park and Seitzer played in a hitter's park (yes KC was a hitters park- a fact James would acknowledge when extolling Royal pitching...)- But Seitzer had a higher OWP...
How did that happen? It happened because McGwire's teammates were better hitters than Seitzer's and the A's pitcher's were worse than the Royals'. IOW he took a great idea- OWP and created something no better than RBI at determining an indivdual's value...
I have never seen James acknowledge that he screwed it up*, nor did he "fix" it (and fixing it was easy), rather he simply abandoned it and moved on to something else...
Look at what Joshua Satin did this year, look at his OBP/SLG compared to the Sallie League average, then consider that his home park had a park factor of .87 (per Dan). One hell of a hidden year, of course 24 in the Sallie League is wee bit old...
* In fact the only comment I have seen from James on the issue took the form of him changing, after the fact, what he claimed OWP was designed to do...
I'm equally certain he didn't mention that Seitzer used 50 more outs. Nor did he park adjust in making the argument. IOW it wasn't as close as he implied.
Yes he did use a goofy OWP at one point -- just didn't use it for Seitzer/McGwire ROY (unless it was in a different place)
Even my 11-year-old self who would have really liked to believe that Seitzer had the better rookie year and unable to recognize the methodical error was unconvinced by that piece. That being said, given the amount of work necessary to write the Abstracts it's impressive how few blunders James made.
Re Blackadder in #41 - so what? They were wrong, then they tried to defend being wrong and ended up more wrong. It's one projection. If the lesson is to be skeptical of single data points, I think we can take that as read. I think that none of those comments made by BP gainsay or even mitigate the fundamental point, which is that player projections (especially statistical player projections) need to be taken with salt. And so we should do so. As MARCEL shows quite well, player projection is just a game.
I'm 100% certain he used OWP as a part of his argument, in fact, he started off his whole argument by observing that Seitzer had a higher OWP... He didn't explain OWP in that article, he merely used it the way someone here will use OPS+ of EQA, and assumed his readers would know what he meant.
no he didn't
No he didn't- if he HAD I'm sure the article would not have been written, it was his error in calculating OWP that lead to his erroneous belief that Seitzer's was higher, that was really the jumping off point.
Basically, let's say that Sean Forman messed up on data entry one day, and one team, got pegged with a .90 park factor when it should have been 1.05, and an unheralded ROY candidate on that team had a listed OPS+ of 130 when it should have been 115. And people would look at that BBREF page and likely take it at face value, and pretty soon you would have people arguing that he's better than some other rookie who only has a 127... and pretty soon the argument would diverge into a discussion of positional adjustments and defense...
and the next day Forman would fix the glitch and the next post in that thread would be WTF
McGwire in 1987 had an OPS+ of 164, an OWP of .727 (per BBREF) Seitzer had an OPS+ of 128 and an OWP of .648. James calculated the OWP for everyone back then, and when he did it for the 1988 Abstract, being a Royals fan he couldn't help but notice that Seitzer's was higher than McGwire's
Now he could have done 3 things:
1: say how about that and let it pass
2: Say, hmmm something is wrong here, McGwire has a higher OBP X SLG and plays in harder park to hit in.... (and James had written extensively by then about how difficult Oakland Coliseum was for hitters)
3: Say, hey! Thats' not fair, not only does Seitzer play a tougher defensive position, but he's better hitter too! I'm going to write about this!
He chose #3.
In the NBJHBA He wrote that he was wrong to believe that Seitzer would be the next George Brett...
If you cross reference his winshares tables you'd notice that he gives McGwire quite a few more winshares in 1987 than Seitzer, but I didn't notice him explicitly mentioning that he was wrong about 87Seitzer v. 87McGwire...
I have every Abstract from 1982 onward, and the later Bills James Baseball books.
What was impressive was he asked questions, someone would say that Nolan Ryan was worth $X to a team because attendance was up because fans came out to see him. James would take that quote and
would look up the paid attendance for every game started by Ryan and compare it to games when Ryan didn't pitch.
No one else was doing that back then (at least not to a wide audience).
What amazed me then (it no longer amazes me) what that some people got pissed off by stuff like that- basically they didn't WANT TO KNOW THAT SOMETHING THEY BELIEVED WAS UNTRUE
A lot of what James writes now saddens me because he's changed (haven't we all), he is no longer inquisitive as he once was, he's quicker to defend conventional wisdom now where before he was quicker to attack it. I get the impression he does not like where the sabr revolution has gone, worst of all I've read hints that he's not above behaving like Seymour Swioff in some ways.
He is still THE GREATEST figure in the history of sabrmetrics.
After seeing Wieters, can you really blame them?
Level vRHP vLHPMinors .961 1.146
Majors .761 .567
I can't find a source for college splits from '04-07, but my recollection is that Wieters was also significantly more successful against lefties at Georgia Tech.
So what gives? Are major league LHP qualitatively different from those in the minors and college in a way that major league RHP are not? I wish I had the minor league splits for other switch-hitters so I could see if there's a precedent for someone to struggle to adjust so much more on his strong side.
Wow, you have a good memory. You both do, actually, but Ron is almost quoting the article.
Here's what he wrote in the '88 Abstract:
McGwire had a higher OWP, but they do seem far too close given the RC/27.
SMALL SAMPLE SIZE ALERT
he has 121 PAs against LHPs
he could be a true talent .300/.400/.550 hitter against lefties and still put up a .200/.250/.300 line in 100 PAs
Shouldn't that result in, like, a shitload of regression to the mean?
(Maybe it did and Wieters' MiL numbers were just that good.)
I remember that
my memory is bad- I distinctly remember that James claimed Seitzer's OWP was higher...
But my memory concerning HOW James calculated OWP is accurate I am glad to say:
McGwire: 8.61 r/g, the A's scored 806 and gave up 789, that gives 4.92, pythag 8.61 against 4.92 and you get .754
Seitzer: 7.03 R/g, the Royals scored and allowed 715 and 691, that gives 4.34 runs a game, Pythag 7.03 against that and you get .724
If he did OWP correctly (IMHO), he would have taken league average R/g: 4.90, and adjusted for park
Oakland's was .93 so (4.90*.93+4.90)/2= 4.73
KC's was 102 so (4.90*1.02+4.90)/2= 4.95
McGwire's OWP = .768
Seitzer's OWP = .669
(using traditional Jamesian pythag- if you use 1.8 power you get lower numbers for each...)
Also, did he never adjust for park? If not, when did park factors become standard?
Still, McGwire was a slightly superior offensive player.
is in no way related to this:
McGwire: 8.61 r/g
Seitzer: 7.03 R/g
That's not "slightly" anything. That's at the least a marked, significant difference, even ignoring the basic (and well-known already at the time, of course) concept of decreasing marginal gains at the high end of the spectrum and how that makes gains at the right side of the parabola more valuable than equal gains in the middle. When you factor in that it's harder to get win 5 than win 1, at the levels we're looking at, calling 8.61 vs. 7.03 a slight advantage is just a misstatement of fact.
No, but he was basing it on a .754 v. .724 OWP (which was calculated badly...)
as far as I know he always did- he also knew or should have known it wasn't right, he defined OWP much as Forman does: the winning % a team of 9 of this guy would have assuming average pitching and defense- to do that you need to figure out a run scoring park factor for each team - adjust the league level of r/g and use that as [part of] your pythag denominator.
Back in 1980, computing power was FAR less than today, info on home road splits was far less readily available, and...
he got lazy.
He simply said ok, Brett plays fro the Royals, the Royals scored 675 and gave up 625, that's both home and road, that's both Royal batting and pitching and everyone else's batting and pitching, it's CLOSE ENOUGH, to show the run context Brett played in...
You know what? It a team's (reverse)ERA+* and OPS+ add up to 200 (100+100 or 95+105, or close to it)- Jame's "shortcut" works.... for about 1/2 the teams in any given year
But in 1987? The Royals team OPS+ was 93, and the team ERA+ was 118 (Reverse ERA+ approx 82)
That meant Royals games were very low scoring in 1987 BECAUSE they had terrific pitching and bad offense (not quite as extreme as the 2003 Dodgers...) so their rERA+ and OPS+ added up to 175, which is very low.
James never really fixed it- he moved on- eventually to winshares [I'm still amazed at how he decided to finally fox RC when he did winshares- he'd known for years what the main flaw in RC was- it didn't scale up correctly at the extremes- especially if a batter had a VERY high OBP and SLG, his RC would start producing 11 to 12.5 runs/g when 9-10 was more realistic. The problem arose from how he multiplied times on base against total bases- he refused to change that aspect because he insisted it was the RIGHT approach, ERP was better- hell Jame seven printed an article on it once, but ERP was essentially a linear weights formula- and he'd spent years claiming that linear weights modelled offense incorrectly... and then came baseruns- which in some ways even looks like RC... but nope... at some point he just decided he was not going to adopt someone else's idea.
He fixed RC, by swapping a player's statline in and out of a hypothetical team stat line and measured the difference- declaring that was the "correct" way to do it. Well personally, to me, that "fix" while it "works" is evidence of his latter thought processes in the face of better alternatives, besides it doesn't really fix the problem so much as dampening it down.
*reverse ERA+ is team ERA/league ERA so a 95 reverse ERA+ would be equivalent to a 105 ERA+
Especially when you know, as James did, that Oakland's stadium decreased offense, and KC's did the opposite.
You must be Registered and Logged In to post comments.
<< Back to main