J.J. Hardy for Carlos Gomez

I agree with Craig Wright, who maintains that if an MLB consultant gets 60% of his calls right, he leads the field.  

Making difficult projections and getting them right is more like completing a pass in the NFL (60%), more like throwing a strike to Manny Ramirez in the NL (60%), than it is like making a free throw in the NBA (90%, if you're Ray Allen). 

Projecting J.J. Hardy, Matt Tuiasosopo, and Brandon Morrow, that's an inexact science (in fact, it's not a science at all).  It's not like Larry Bird stepping up to toss in a 15-foot set shot.   It's more like trying to complete a down-and-out against Marcus Trufant.  (Okay, bad example.)

Craig doesn't view 60% as making you legit.  He views 60% as making you the best.

.....

Bear in mind, we are not talking about "projecting" whether Felix Hernandez will be good.  We're talking about challenging calls, one on which good analysts disagree, questions such as "will J.J. Hardy probably recover to become a minor star, or will he probably not?"

At SSI, we don't view ourselves (or other websites, or even Bill James His Ownself) as having 975 out of 1,000 light bulbs on.  It's more like 200-300, and others' light bulbs are different.  Hence, it's not often that we consider our take on Aroldis Chapman or Jarrod Washburn or Michael Saunders or Edwin Jackson or Matt Tuiasosopo to be "correct" and somebody else's "incorrect."

But when we do complete a pass downfield, once in a while we don't mind pumping our fist a little.  :- )   You surely can find incompletions and interceptions if you scroll through the POTD's, bro'.

This is one reason we don't go back retroactively and claim, "My opinions that Zito and Schmidt were good signings, those were based on the best information available at the time.  Everything I say is correct but circumstances change."  My opinion on signing Barry Zito wasn't correct at the time; it was based on having a light bulb off, that needed to be turned on.

Internet analysts don't need to go back and defend a statement that "the Tigers were idiotic to trade Matt Joyce for Edwin Jackson," because 99% accuracy isn't the minimum acceptable standard.  67% isn't even the minimum standard. 

.............

Looking around the internet, we find shock-and-awe that J.J. Hardy was traded for a career bench player.  This fringe player being the best offer that Milwaukee got for J.J. Hardy, who just one year ago was one of the most valuable properties in baseball.

It's simply relevant to the discussion:  the comments this morning run along the lines of, "What, why wouldn't Milwaukee want Michael Saunders instead of Carlos Gomez?"  :- ) And the befuddlement is obvious and sincere.

What happened, was that cyber-Seattle sized up Hardy in a completely different way than the industry's GM's (and SSI) did.   That's cool.  Like we say, 60%, kiddies.

Milwaukee would (almost certainly) choose Michael Saunders, who is a center fielder loaded with potential, over Carlos Gomez, who is a center fielder exposed as a lifetime fringe player. 

Jack Zduriencik evidently didn't want J.J. Hardy.   Nobody did, very much. 

Zduriencik could easily have declined Jack Wilson's option, brought in Hardy, and saved millions, because Hardy makes a lot less than Wilson (who himself OPS+'ed only 51 for the Mariners).  He didn't.  He doesn't want J.J. Hardy.

..............

Carlos Gomez is only 24, and won't cost much in salary, but then again neither will any other young backup player who has little future in major-league baseball.

Gomez is, in practical terms, one of the fastest players in organized baseball.  Beautiful!  The problemo is that he is a banjo hitter who ran a 25:142 EYE (!!) in 2008.  That's the definition of a fringe 4th OF.

In 2009, Gomez' EYE was 22:72 but in this case, that's not "growth."  That's adjusting your game to embarrass yourself a bit less.  Gomez hit .229/.287/.337 in 350 AB's, and that's in the National League.

Gomez came up a hyped prospect, because of the Coleman-esque speed, but even his minors line was only .278/.339/.399.

Gomez is only 24, so it's not like he's not going to stick around baseball.  But he'll be attempting to put together an Endy Chavez career, and every org has a couple of real fast CF's who can't hit.

........

Could the industry be wrong about Hardy?  Sure.  But my judgment is that its winter assessment of Hardy is solid.

Minnesota wound up being the team that took a flier.   Hardy's liable to go through serious hitting travails anyway, but in Safeco ?! you were just asking for a Ronny Cedeno situation.  ::shudder:: Likely that the M's dodged a real bullet here. 

And commendable that Capt. Jack -- who could have had Hardy real cheap, but passed -- isn't married to his old Brewers stars.

Good show,

Dr D

Comments

1
Taro's picture

This is a pretty excellent buy-low move by Minnesotta.
If Hardy pans out, you look like a genius. If not, you didn't give up much.
I love the move for the Twins.

2
Anonymous's picture

that this was a deal made for fringe players, which is exactly what both are. Kudos to Doc for calling Hardy what he was, and to the rest of us for flailing about when the mid-season move of Bedard/Washburn/Morrow?  was bandied about as if it came from a fact-based perspective.  Hilarious over-rating of marginal players spawned from the replacement level based analysis that pains me.  When will we question the source?

3

So let's see...USSM Mariner proposed we trade Bedard AND Washburn AND one more player for Hardy in July.  With a straight face.  And they are still taken seriously for some reason.  Yes, I remember that post...and I remember loudly making fun of it at MarinerCentral.

4
Taro's picture

J.J. Hardy isn't a fringe player though.
Even in a scenario where his bat doesn't rebound hes a league-average player. Hes roughly comparable to a player like Jose Lopez.
If his bat rebounds, its going to be a hilarious steal.

5

I don't think he's good enough defensively to be a league average SS if he doesn't regain his bat.  And I don't think he's going to regain his bat.

7
Anonymous's picture

Hardy would need to hit about .270/.330/.400 if you are banking a defensive bonus of +10 runs.  In that event, he'd be +2 WAR and comparable to Lopez past tense.
Last year, he actually hit was was it, 250/300/350 or something and was nosediving with a bullet.
..........
You could very well be right, champ, seriously.  Maybe Hardy will hit a halfway respectable 85 OPS+ and because of his glove, be an average player.
But not a single GM believed that enough to offer Milw anything more than a bench player like Gomez, and there are serious needs for SS's out there.  A +2 WAR SS making $5M would get nice offers -- as cyber-Seattle expected that Milwaukee would get. 
Jose Lopez is going to bring back a little more than a 4th OF, right?
...........
If Hardy continues to struggle at the 250/300/350 level, or even worse in the AL, he's going to be out of a job period.
............
The fact that you like Hardy's chances to rebound to +2 WAR is worthy of respect.  I have serious doubts about that, even more serious doubts that he could do it in Safeco, and FWIW *every* GM agrees with me.  :- )  Jack Zduriencik evidently does, as well. 
The truth will be seen in Hardy's actual performances.
- Jeff aka jemanji aka Dr. Detecto

8
Taro's picture

".270/.330/.400 if you are banking a defensive bonus of +10 runs"
You are describing a 3-3.3 WARish player Doc (depending on park).
Hardy doesn't have to rebound to be a league-average player. He already was in '09. If he rebounds, hes an impact player.
I love the deal for the Twins. Theres a real good chance they just did a one for one where they got the player with the higher downside AND the higher upside.
You might like Hardy less than most, but he certainly isn't a fringe MLB player.

9

I see what you are getting at, now that I flip over to Fangraphs.  They've got him, mistakenly, at +1.4 WAR based on about 70% of a season's time.  Presumably that is where you're getting your +2.0 full season.
A .229/.302/.357 shortstop, with an above-average mitt, is going to be lucky to hold a starting job AS a replacement-level player, much less get paid $8M as a guy 2 wins better than replacement.
..............
I don't know whether the "bleed" here is in fangraphs' Replacement, or Positional, or simply that a bunch of SS's had lousy years with the bat, or what.
But stop and take a breath for a second.  If J.J. Hardy were a +2.0 WAR shortstop making $5M, don't you think the Brewers would have gotten serious offers on him?
................
Whatever Fangraphs' replacement numbers said in 2009, GMs know that 230/300/350 with a good mitt can be picked up off the scrap heap in AAA.  That's not +2.0 WAR; it's 0.0 WAR, whatever Fangraphs says.
We run into James' Johnny Bench rule here:  if a defensive system calls Johnny Bench a bad catcher, the defensive system is wrong.
If a numerical system calls J.J. Hardy's 2009 season +2 wins, the numerical system got it wrong.
..................
I don't like Hardy less than most.  I like him the same as all 30 GM's.  Let's not forget that both of our theories just received the experimental feedback of a MLB-GM bidding war.

10

Taro m'man, would you offer 3/$20M to a young FA shortstop hitting 230/300/350 with a +10 runs glove?
+2 wins, three-year term, critical defensive position, 3/$20 would be a no-brainer.
............
To answer my own question :- ) 230/300/350 would have ranked as the 5th-worst offensive player in the major leagues in 2009.   Hardy created 3.5 runs per 27 outs in 2009.  He's a league-average player already?!
David Eckstein was better than that.  Christian Guzman OPS'ed 700.
How bad would Hardy have to hit, to be a RLP?  -20 runs worse than a 650 OPS, I guess.  Boy, the state of AAA baseball these days...
.............
I suspect the replacement/positional values are funky this year for SS's.  In any case I expect to be able to develop a SS who can do +10 defense, 3.5 runs a game, for free.  
 

12
Taro's picture

UZR had him higher this year which is why he rated as a 2 WAR player over regular time. Regardless, his down season was a 1-2 WAR depending on how high you rate him defensively (if +10 seems is closer to true talent level).
You'd need to actually calculate it out though. You were saying a .330 OBP/.400 SLG +10 run defender would be league-average and this is just not true; thats a comfortably above-average player. Theres the positional adjustment that you need to take into consideration as well. SS is very high up the pyramid.
Hardy aint fringe. IMO this is taking the argument to too much of an extreme.  Hes a guy who was a minor star in his prime that had an awful season under his standards - that was still roughly league-average. Saying you don't think Hardy will rebound is one thing, saying Hardy is a RL scrub is another.
If Hardy doesn't rebound, then hey, you didn't give up much and its still an upgrade over what you already have. If he does even a little then its robbery. Good deal for the Twins IMO. Very little downside and very high upside.

13

Fangraphs uses a system similiar to Tango's which is:
-offense compared to league average
-defense compared to positional average
-separate positional adjustment (and I'm pretty sure it's the same across years)
-fixed replacement-level adjustment based on playing time
If you go to a player page you can see the value for all four of those components. The biggest issue that people need to keep in mind is that thier replacement-level is lower than some others, including Tango's. That means an average player is probably more like 3 wins, maybe more, using their numbers rather than 2.5. That still means Hardy was a useful player even at his worst. Plus, there are very few players you should give up on after just one season and I don't see why JJ is one of them. Therefor, I agree with Taro that he is a great player to snatch up for cheap.

14

...the error is in the positional adjustment and the "static replacement level adjustment"...
I think the problem is that we use other marginal players at each position to represent that position.  But the players selected from the minor leagues to play short are selected for specific skills that are often different than the skills selected for in full time starting options.  When you lose your starting shortstop, 9 times out of 10, the guy replacing him is chosen because he is seen as reliable on DEFENSE exclusively.  Further complicating matters is the fact that the 4th outfielder is chosen with different criteria than the 5th infielder for different reasons.  The way we've gone about defining positional replacement levels is inherently flawed because it's inherently empirical and has no logical basis.
I also think the way to find a player's replacmeent level is not to make some linear adjustment based on playing time but to directly calculate what a replacement level player would produce in a given amount of playing time using pythagorean logic.

15

Gotta agree with Matt.  Each year I am less enthused with UZR.  First off, it's based on a foundation that I believe has flaws.  (I think the official UZR scorers in "some cities" skew results -- and if the basic ZR is tainted, then everything behind it just goes completely bonkers).  Of course, I think in "most" cities the assessments are reasonable -- but on occasion the team UZR and DER results end up so vastly different from each other that it defies all reason.
But, the positional adjustments are the arena where I think the system completely and utterly falls apart and I don't like the underlying oversimplification that essentially says - if you strap the tools of ignorance on Jason Giambi, he gets 2-WAR BEFORE HE MAKES A PLAY.  (I know Tango's system actually gives catchers the highest positional adjustment, which is 2, IIRC).  I think SS is 1.5.  Moreover, the positional adjustments have ZERO to do with actually playing defense.  They were generated based on salary differentials.  I suspect when AROD moved from SS to 3B, every player at BOTH positions had their positional adjustments ... adjusted.
This is, (IMO), a case where the original ideal was to try and get a read on relative PAYROLL value between players - so one might clear up some of the fog in judging 1B for SS trades and such.  In that regard, this approach may well be excellent.  But, because that was the goal, the actual attention to defensive performance has been shoved to the back of the bus, rendering (IMO), UZR essentially useless for actually discussing defensive prowess, and in many cases, ultimately detrimental.
Honestly, if you go to Fangraphs and pull up ALL fielders, (auto-sorted by raw UZR), many things leap out as questionable to downright bizarre.
Gutierrez posted a 29.1 UZR -- (total runs saved), beating the next best glove in all of baseball, (Longoria - 18.5), by more than 10 runs.  That's a HUGE margin. 
Another oddity is that 4 of the top 7 run savers by UZR were third basemen.  4-3B, 2-LF, and 1-CF made up the top 7 run-saving gloves.  Does it really make sense that there would be a trio of LFs who saved more total runs than the 2nd CF?!?  Crawford, DeJesus and Juan Rivera, (plus RF Nelson Cruz), all ranked as saving more runs than the #2 CF (Upton). 
Jack Wilson, with a 14.0 UZR, saved more runs than any other SS, (while only playing 105 games?!?).  The #2 SS?  Cesar Izturies, (who played 112).  Then Everett (116).  You have to get all the way down to Rafael Furcal with a UZR of 8.0, before you get a SS who played the whole season, (149 games).  The fact that Furcal was tied with Ibanez with a URZ of 8.0 just strikes me as inexplicable.  UZR says that Ibanez saved as many runs, (in his 129 games), as the BEST full-time SS in baseball. 
According to UZR, Wilson was the best SS in baseball - (14.0 UZR and a 20.4 UZR/150), while YuBet was the worst, (by far), -20.5 raw -23.9 UZR/150.  Does ANYONE believe that EVERY other qualifying SS in baseball fits between those two in defensive runs saved?  Yes, the Ms DER improved a tad after Yuni left.  Is the difference between Wilson and Yubet REALLY more than 40 runs?  Per UZR/150, the diff was 43.9.  Can it REALLY be that big? 
I don't honestly know.  But, I see soooo many things about UZR when I scan the lists that just don't "feel" right -- not just a little off -- but drastically off what the eyes (and other stats) suggest -- that I cannot accept UZR as a valid stat for judging defensive prowess.  At this point ... I'd feel about as comfortable judging a fielder based off of UZR (alone), as I would judging a pitcher based off "runners picked off" alone.
 

16
Anonymous's picture

was all that I was referring to when I was reminded of a trade scenario that was discussed for a week near the deadline.  When the leading proponent for a stat "system(s)" uses the tool exclusively for defining value so vastly over-shoots the reality of return, it is probably a good time to question the validity of the process.  
I've seen statements claiming that the big-league clubs are actively using all or parts of these valuation models in personnel decisions.  This is an example which shows that we are a bit too confident in our conclusions as of now.  From what I can see (excellent post above, Sandy) UZR is assisting in overstating defense and where I can see the utility in the system for comparing different positions to some extent, the over all result is skewing the relative importance of defensive players.
Systems are broken IMO that show any left fielder to be more important defensively than the CF next to him.  

17

I think the only correct way to rate defense is to begin with a top-down model.  That's the only way you're going to get the scales and relative movements of defensive excellence right.  All of the personal defensive metrics like UZR are open to a ton of biases and subjective elements that RUIN THEM...a top-down team defensive context ties everything to the real world results...runs allowed.

18

I think a left fielder can be (and occasionally is) more important defensively than his CF counterpart in extreme cases where the left fielder is outstanding on defense and the center fielder is only average or worse...
However...any system that has 4 corner outfielders ahead of the second best CFer is broken...on that I agree.

19
Taro's picture

UZR is far from perfect (as any defensive stats), but I do think its just visually harder to see how a +9 can go to -9 or whatever in a single season as opposed to offense. Since we usually visualize offense as a rate stat and defense ONLY in relation to league-average, we often don't even think about offense in related to average.
I mean just looking at Hardy's offense (in relation to average) how would you explain it?
2005: -6.5
2006: -3.9
2007: 3.5
2008: 13.7
2009: -13.2
Thats how he performed offensively, but how do you make sense of this data? How can a guy be a 13.7 run offensive player one season and -13.2 the next? Are the offensive formulas flawed? I think players are lot more volatile season to season then we realize.
Also the position adjustments account for the average difference in defensive runs per position (which is why total raw UZR may be misleading).

Add comment

Filtered HTML

  • Web page addresses and e-mail addresses turn into links automatically.
  • Allowed HTML tags: <a> <em> <strong> <cite> <blockquote> <code> <ul> <ol> <li> <dl> <dt> <dd><p><br>
  • Lines and paragraphs break automatically.

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.

shout_filter

  • Allowed HTML tags: <a> <em> <strong> <cite> <blockquote> <code> <ul> <ol> <li> <dl> <dt> <dd>
  • Lines and paragraphs break automatically.
  • Web page addresses and e-mail addresses turn into links automatically.