Wednesday, August 13, 2008

Git 'er Dunn

A (mostly) imagined conversation with the little voices in my head:

Hello, my name is Pete. I’m a huge Reds fan. I love listening to Marty Brennaman and Jeff Brantley call the games. I also have a life-size poster of Chris Sabo on my wall. I see you’re a baseball fan, and you seem to be from Ohio.

Uh, yeah. My name is P and I blog about baseball once in a while. But I’m from the northern part of the state and I’m actually more of an Indians fan.

Well, your team hasn’t won the World Series since 1948. We had the Big Red Machine! And we’re the first professional baseball team!

Yes, the Indians do not exactly have a storied history, and the Reds of the 1970s were pretty good. However, the modern Reds don’t really have anything to do with the 1869 Red Stockings, except they play in the same city and kind of use the same name. But that’s a topic for another day…

Whatever. Hey, I see we got Micah Owings as the player to be named later for Adam Dunn! Awesome! He’s a better hitter then Adam Dunn.

Uh, you’re kidding about that last part, right?

Well, it may be a bit of hyperbole, but Adam Dunn isn’t that good of a hitter. All he does is walk and hit home runs. The rest of the time he strikes out.

I agree that Dunn is not one of the elite hitters in baseball. However, he’s been the Reds' best hitter by a wide margin over the last few seasons, and was again in 2008.

But he’s only a lifetime .247 hitter. And he strikes out a lot.

Yes, he does strike out a lot, and his batting average isn’t great. But he makes a lot less outs than the average player does; he’s got a lifetime .374 OBA.

You can talk sabermetrics all you want. The point of baseball is to score runs and stop the other team from scoring.

Indeed. That’s what sabermetrics is all about on the player evaluation level: trying to estimate how many runs a player was worth to his team in all aspects of the game.

Then why don’t you look at PRODUCTION? Runs scored and RBI. Look at Adam Dunn’s game last night (August 12) for the Diamondbacks. He went 1-3, and he hit a double and drew two walks. That makes his OBA .600 and his SLG .667 for a good OPS. But no production…zero runs scored, zero batted in. And he didn’t play well in the field, and he struck out two times, which is why he had no production. He’s why Moneyball is stupid.

Wow, Pete, I don’t know where to start. For one thing, I’m not saying he’s a great fielder. He’s pretty bad, and he’d probably be better off at first base or DH. And I’m not saying that I would sign him to the type of contract extension that he wants, nor is at least one other sabermetrically-inclined person. I'm not even saying the trade was bad. I’m just saying that he was by far your best offensive player, and arguably the most valuable player on your team this season.

For one thing, I don’t accept R and RBI as “production”. They are dependent on what Dunn’s teammates do before and after he bats, which is out of his hands. You can argue a number of different positions on this, but at the very least you have to admit that the most he can do is make good on the situations with which he is presented.

So from that perspective, let’s look at his game from last night, PA by PA:

#1: bats with runner at 1st, 2 out in 1st and fans. This is an irrelevant strikeout, since the result would have been exactly the same had he made another kind of out. This is a case in which an out is an out is an out...bad, of course, but not any more so because it was a K. Play reduced DBacks' estimated win probability by 2.3%

#2: leads off the 4th with a walk. And here's a case where a walk really is as good as a single--no runners on base to advance. +4.3% WPA.

#3: bats with 2 out, runner at first. Hits a double, advances runner to third. No RBI, but definitely productive...not his fault that the next batter failed to drive in a run. Increased WP by 4.8%.

#4: leads off the 7th with a walk. Again, walk is as good as a single in this case, +2.4% WPA.

#5: bats with 2 outs, runners at 1st and 2nd. Strikes out; again, two out K may as well have been a groundout. -2.6% WPA.

That’s a productive night at the plate.

That’s just sabermetric spin. He didn’t put any runs on the scoreboard, period. I’m not interested in potential, I’m interested in results.

If you’re only interested in the results, then why do you care about the two-out strikeouts? They have the same effect as any other out would have had--the inning is over, and no runs scored.

But if he would put the ball in play, something good might have happened. If he strikes out, nothing good can happen.

Fair enough, but we already account for the errors and other assorted events that occur on some “outs” in play by giving the other outs a higher linear weight value than a strikeout.

And if he drew a walk to lead off the inning, or hit a double to extend the inning and advance a baserunner, the next batter MIGHT have gotten a hit, and then he would have had “production”. Or maybe he wouldn’t have, since the accounting system of R and RBI ignores a lot of events that are crucial to a run scoring, but don’t actually result in the player scoring or driving someone else in.

But that didn’t happen. His teammates didn’t come through, and thus he produced nothing.

Those things like errors that could have happened on an out in play? Usually they don’t. Had he grounded out, you couldn’t complain about the strikeout, but the Diamondbacks would have been no better off.

But we don’t know what would have happened if he wouldn’t have struck out. Maybe there would have been an error.

I’ll give him credit for it when it actually does happen. In the mean time, back here in reality, he made an out, the inning is over, and we accounted for that. Can we move on?

No. The great hitters like Williams, Gwynn, and Boggs, they didn’t strike out much.

You keep making it sound like I think Dunn is the new Frank Robinson or something. Obviously Dunn is not in the class of those guys as hitters. And while Dunn may strike out more than just about any good player ever, don’t try to tell me that great hitters like Ruth, Mantle, and Jackson didn’t strike out a lot either.

Dunn isn’t as good as those guys. Dunn just doesn’t produce.

You’re right, Dunn is not as good as those guys. Say, do you know who leads the Reds in runs scored as of August 12?

Since you’re asking, it’s probably Dunn.

No, actually it’s Brandon Phillips with 67. Edwin Encarnacion is next with 61, and Dunn has scored 58.

However, Dunn leads in RBI 74-68 over Phillips.

Who leads in that new-fangled WPA stat you mentioned, that you claim actually looks at the situation which the batter is presented anyway?

Dunn, with +1.83. Jerry Hairston is second, and Phillips is down at +.41. You can see for yourself at Fangraphs.

Whatever. Dunn could be so much better if he changed his approach and tried to put the ball in play more often.

I suppose that’s possible. But now you’re evaluating him based on what you think he should be, rather than what he is.

Yeah, I am. If he’s the best player on your team, then you’re in trouble. Look at the Reds. They suck. Look at my namesake, Pete Rose. He got more out of his talent than anyone.

For one thing, you guys need to get over Pete Rose. He’s been retired for years, and banned from the game for almost as long. You need to move on, and stop idolizing every Chris Sabo and Ryan Freel that comes along and scraps his way into your heart. You don’t have to be Pete Rose to be a good ballplayer.

It is pretty common for a fanbase, or at least the less-analytical portion of a fanbase, to turn on their one or more of their best players when they are confronted with a losing or disappointing team. See Beltran, Carlos and the Mets. However, making a scapegoat doesn't solve any problems. It actually inhibits solving problems, as once the beast is slain (and he's in an Arizona uniform), the job appears to be done, when all you have actually done is shed one of the few bright spots on your team. Of course, it may still be a good move for your team in the long run, but if so, it's not because Dunn's performance is off your team, it would be because you have not ponied up for Dunn's future contract and/or because you received good talent in return.

Really, shouldn’t you point the finger at the guys who aren’t getting anything dunn?

That’s a lame pun.

Yeah, I know, I’m not very creative. That’s why I’m having a dialogue with you, Pete.


  1. For one thing, you guys need to get over Pete Rose. He’s been retired for years, and banned from the game for almost as long. You need to move on, and stop idolizing every Chris Sabo and Ryan Freel that comes along and scraps his way into your heart. You don’t have to be Pete Rose to be a good ballplayer.

    Damn right.

  2. I'd really like to see a super lwts evaluation of Dunn--to see how all of his good points and bad points balance out. My guess is that the good stuff (the walks, the homers) are largely canceled by the bad stuff (the low hit rate, the bad fielding, the slow baserunning). I wouldn't be surprised if he's merely an avg overall player.

  3. The problem with Adam Dunn has never been Adam Dunn, it's been the positions in the lineups that he's been used in, which he has no business being in.

    The truth is, Dunn should be batting in a slot where his "production" (high OBP, HR's) wont hurt the team by not having his come through in a "must run producing" situation.

    His contact rate is terrible, his BB's are great, and the occassional home run is nice, however, that doesnt mean he is a good enough hitter that he should be batting 3rd, 4th, or 5th, in ANY lineup, on ANY team, at any time.

    IMO, the best positions of the lineup for Dunn would be positions where he can use his high OBP and occassional HR to his advantage, and not have to worry about coming through in situations where he needs to make contact, or, i.e.-"produce".

    I think the #2,#6, and #7 spots in the lineup is where Dunn would fair best at (preferably #2).

    Not only would he feel less pressure to "HAVE TO PRODUCE", but his contact rate would go up and his strikeout rate would cut down, imo, due to him seeing more pitches in the #2 slot, along with more hittable pitches.

    Here are some stats (limited PA's) that show what kind of player Dunn could be, in the 2 slot in the lineup: (211 PA's in 2 slot)

    Dunn line while batting 2nd:

    54/176 hit/ab's
    29 runs
    13 hr's
    33 rbi's
    35 BB's
    5 SB's

    .307 BA / .436 OBP / .593 SLG

    1.029 OPS

    Here are the splits, if you want to see for yourself.

    See what I mean?

    Personally, I'd love to see if there was a stat that kept track of the number of runner's LOB, per individual player.

    Is there such a thing?

    Also, what does WPA stand for?

    Thanks for your time,


  4. David, I too would be interested in seeing Dunn's SLWTS workup, and he is one of the players for whom it could make a difference.

    Doing a poor-man's knockoff, though:

    B-R lists Dunn at +36, +13, and +30 Batting Runs for 2005-2007. I don't think that accounts for the difference between K and other outs, so I'll dock him another 2 runs to be liberal.

    His equivalent baserunning runs as figured by Baseball Prospectus are actually not bad. He's at -3, -1, +2 for those seasons. So putting those all together, he's about +24 runs on offense.

    Take away 10 runs for a positional adjustment and he's still +14 before considering fielding. And that's the hard part, ain't it?

    As a DH, you could take away another 10 runs for position, but then there's no fielding penalty and you've still got a +14 player. Would I give a 28 year old DH who can play LF without completely embarassing himself, with old player's skills a long-term contract? No. But I would love to have him on my team next year.


    I wouldn't really put a whole lot of stock in his splits batting second, as their only over ~200 PA. However, I don't disagree with the point that Dunn is a guy who could be used in the #2 spot due to his high OBA. Of course the conventional wisdom is that you would not want to put him there because he does strike out a lot and thus fail to advance runners.

    WPA is Win Probability Added. It measures the change in win probability as a result of each play. They track it at Fangraphs and if you go there they have some more you can read about it.

    I'm not aware of a source for LOB totals, but it is tracked in some box scores. However, I think that if you look into WPA, you'll see that it accounts for the situation a lot more comprehensively than something like a RBI percentage would.


I reserve the right to reject any comment for any reason.