May 5, 2013; Kansas City, MO, USA; Kansas City Royals left fielder Alex Gordon (4) drives in the winning run against the Chicago White Sox during the tenth inning at Kauffman Stadium. The Royals won 6-5. Mandatory Credit: Peter G. Aiken-USA TODAY Sports

Dustin Ackley and Alex Gordon


Dustin Ackley‘s struggles to this point of his career have obviously led to a lot of disappointment, anger, and probably some prayers for a break-out. But it has also caused people to search high and low for some kind of sign that suggests he is likely to reach the potential that most people assigned to him. One way that people do this is to look for a comparisons. They try to find another player who had similar expectations, but took awhile to reach them.

And the prime candidate seems to be Alex Gordon. Gordon won multiple awards in college, leading him to be selected 2nd overall in 2005. He then hit well in the minors, and was even named Baseball America’s Minor League Player of the Year. But, it all went down from there.

In his first MLB season, Gordon posted a lowly .316 wOBA and 87 wRC+, far below expectations. He had another disappointing season in 2008, and then saw his playing time decrease over the next few years. He went from the next big thing, to the next big bust. From the next George Brett, to “Who is Alex Gordon?”. But again things changed, this time for the better. He had a breakout year in 2011, posting a .382 wOBA and 141 wRC+, coupled with great defense in left field, which then led to a 6.7 WAR season.

So it is easy to see where the comparisons come from. They are similar players who both failed to reach expectations at the beginning of their careers. But let’s look at how the two actually compare through 1161 plate appearances. (Note: I used the triple-slash instead of wOBA and wRC+ because they are easier to calculate by myself. Also, the Swing and Contact rates may not be 100% accurate, but close enough.)

Player AVG OBP SLG OPS O-Swing% CONT% K% BB% ISO
Gordon 0.254 0.326 0.422 0.748 24.36% 75.56% 24.60% 9.20% 0.168
Ackley 0.243 0.311 0.354 0.665 24.90% 85.50% 19% 9% 0.111

Unfortunately, the comparison is not as close as people want it to be at this point. Gordon has the edge in everything except Contact% and K rate, two pretty minor stats. You would hope that your K rate is low and contact rate is high, but it isn’t necessary to find success.

The biggest difference though is the slugging percentage. Ackley’s power-outage this year is something I touched on before, and is really holding him back. His BB% is also way down this year as well, so that needs to change too if he wants to be more than a singles only hitter with an empty .250 average. (Hint: That isn’t very valuable.)

So at this stage of the game, the comparison looks pretty baseless. Sure, they were both highly regarded prospects who struggled/are struggling. But Ackley has out-struggled Gordon, and it is not even close. Plus, Gordon was a 2+ fWAR player even with an 87 wRC+ because of his awesome defense. And while Ackley has looked pretty solid out there himself, I don’t think he can hit like he is and still be considered an average starter (~2 WAR for those who don’t know).

Remember when I said people are constantly looking for glimmers of hope? Well, I am one of those people. Because of that, I ran another regression designed to “predict wOBA”. I ran a regression to project BABIP a few weeks ago, so refer to that if you need any more information or details on what a regression is.

Here are the results:

SUMMARY OUTPUT
Regression Statistics
Multiple R 0.856931
R Square 0.734331
Adjusted R Square 0.732239
Standard Error 0.014751
Observations 385
ANOVA
df SS MS F Significance F
Regression 3 0.229148 0.076383 351.0382 2.9E-109
Residual 381 0.082902 0.000218
Total 384 0.312051
Coefficients Standard Error t Stat P-value Lower 95% Upper 95% Lower 95.0% Upper 95.0%
Intercept 0.084866 0.022317 3.802791 0.000167 0.040987 0.128745 0.040987 0.128745
ISO 0.58324 0.019743 29.54124 1.4E-100 0.544421 0.622059 0.544421 0.622059
O-Swing% -0.11071 0.013724 -8.06652 9.51E-15 -0.13769 -0.08372 -0.13769 -0.08372
Z-Contact% 0.208877 0.022089 9.456279 3.35E-19 0.165446 0.252308 0.165446 0.252308

As you can see, I used ISO, O-Swing and Z-Contact as the “predictors”, because they represent the three phases of hitting — those being power, patience and contact — very well. And that turned out to be true, as there is a very high r^2, at .73.

So, using those numbers, we can try to predict what a player’s wOBA should be, to about 73% accuracy. When we do so for Gordon, using his numbers in his first 1161 plate appearances, we get the following equation:

0.084 + 0.58 * .168 – 0.11 * .24 + .21 * .84 = .331

So according to this formula, Gordon’s wOBA “should” have been around .331. And that is right between the .316 and .343 he put up in 2007 and 2008 respectively, which I think helps the credibility of this formula. He has since surpassed that obviously, but at that point in his career, that’s about the kind of hitter he was. It also kind of gives us something to compare Ackley too, because we know that it can be pretty accurate.

Now for Ackley:

.084 + 0.58 * 0.11 – 0.11 * 0.25 + .21 * 0.90 = .309

This is telling us that Ackley’s career wOBA should be around .309. His actual career wOBA? .293, a decent clip lower than the prediction. But you know what is kind of cool (for my self-esteem, not Ackley), is that if we factor in the standard error, which in this case is about 0.15, we get either .294, or .324. Ackley’s actual career wOBA is .293, .001 away from the negative standard error. That may just be a coincidence. I don’t have a good enough grasp of statistical analysis of this nature. But it certainly looks like it makes sense. We don’t want to focus on the negative though. This is supposed to give us hope, not reassure us that Ackley is bound to be bad.

If we look at the positive deviation, as I said before, we get a .324 wOBA. I think most of us would definitely take that kind of offense from Ackley, especially at his position, along with his above average defense. And, in yet another scary coincidence/vote of confidence for the accuracy of this whole thing, if take the standard deviation of Gordon’s .331 wOBA, you get .346, .003 away from the wOBA he posted the following year. Man I am good/lucky.

Anyway, I really do not know if there is anything conclusive from this last part. I mainly did this to fulfill my own curiosity, and to use meaningful math. As I said, I do not know if the things I found are a sign of luck or accuracy. I am leaning towards accuracy, but I really do not know. Maybe someone out there does. But either way, just take it for what it’s worth. Or completely ignore that part if you don’t believe math. Do whatever you want. Be your own person. Don’t do drugs. Sorry, off topic.

The first part though I think is very important. There really aren’t many similarities between Ackley and Gordon. Their levels of suck are vastly different. Ackley wants to be Alex Gordon of now when he grows up, but would not complain if he became 2007-08 Gordon either. That is how different these two guys are/were, and how ridiculous it is to compare them just because they both struggled in some way early in their careers. I mean, if you are comparing them just to reassure yourself that it gets better, then fine. But I would be very careful about assuming that Ackley is bound to become an All-Star all the sudden because Gordon is.

The harsh truth is, Ackley lacks the power that Gordon has. He led the league in doubles last year, and had 23 homers the year before. Ackley is going to have to make a massive improvement if he wants to get anywhere close to that kind of production. Could it happen? Sure. Is it likely? Unfortunately, no.

 

Next Mariners Game View full schedule »
Tuesday, Sep 22 Sep7:05at Oakland AthleticsBuy Tickets

Tags: Alex Gordon Dustin Ackley Seattle Mariners

  • Aidan Flynn

    Really nice job JJ, awesome regression analysis. I honestly wish I knew how to do that haha

    • JJ Allen Keller

      Hey, thanks man. I appreciate the kind words. It really isn’t as hard as it seems. I had a lot of help from an old SM writer on my first one, but that one I did by myself. If you want me to help me out, you can find me on twitter or something and ill give you my email and I can walk you through it.
      There is a lot of info on the regression that we don’t really need for basic baseball purposes, so it looks a lot more daunting than it is.