EconoBall: Doubles and the Aging Process

One of my larger goals with this series is to evaluate player performance over time to get a better idea of how individual players will evolve as they age.  The discussion two weeks ago regarding a player's prime was one example of this.

To this end, analysis of historical data suggests there are at least two ways to increase the number of doubles hit: one is to hit with more power and the second is to have more speed.  Backing up this interpretation, while these two variables both have an impact on the number of doubles hit, speed does not impact the number of home runs hit and power does not impact stolen bases.

This finding leads to an interesting question: do fast guys and power hitters see differences in how their doubles evolve over time?  To test this, I used the data set with every individual season since 1945 and broke the players up into power hitters and speed guys.  (I need to take this moment for a hat tip to my brother's wife, Tanya, who wrote the code that allowed me analyze power hitters and speed guys.  While I'm worrying about important things like who wins the World Series, she and her husband deal with minutiae like how the universe came into being and where it's going.)

First, I look at home runs between the two different groups.  These can be seen in the chart below:

hr p v s

The difference is pretty stark.  Power hitters gradually grow into their "peak years" and then slowly descend.  The speedsters, meanwhile, maintain a fairly steady (lower) level of production throughout their career.

Interestingly, the same is not true of doubles.  As the chart below demonstrates, what is remarkable is how close the two lines resemble one another.  The notable exception is during peak years when the power hitters tend to separate themselves.


doubles p v s


This result was surprising to me.  I was expecting to see a larger differential down the back side of a career as power hitters had balls that used to go for home runs go for doubles instead and speed guys could no longer leg out a double.  These results suggest that if power does have an impact, it's in the ability to produce at a higher level during the prime years.  It suggests, once again, that Theo knew what he was doing when he attempted to put together a team that will have the major pieces enter their prime at the same time.


Hey, I finally get to update one of these things!

John suggested that some of the doubles that speed players get in later years may be coming from hits that they can no longer leg into triples.  There's a reason he's in charge of this place.  The data suggests that he may be on to something.

For speed players, number of triples peaks at age 23 at roughly 5.7 triples a season (if that seems low, consider that Billy Hamilton, one of the fastest players of all time, has only 7 so far this season) and then tails off by almost 2 triples a year by age 30.  For Power Hitters, triples maxes out at age 22 at 3.8 per year and drops by only 1 per year by age 30.  This is a bit of an extreme assumption, but if we assume that all the fall-off in triples is due to decreased speed and those triples become doubles, we can then calculate an adjusted number of doubles hit, excluding those the player gets from triples.  When we do this, we do see the results closer to what we initially expected, with power hitters hitting more doubles throughout their later years.

doubles p v s adjusted

Filed under: Uncategorized


Leave a comment
  • Very interesting way of looking at the data.

    The future Cubs are going to be "loaded" with just unbelievable power hitters. Hopefully some of those guys will develop high(er) OBP during their peak years, to provide even more chances for scoring runs.

    I feel that if we have 4-5 guys with OBP over .360 every year, we will be doing quite well in the playoffs. I can see that happening, even with many of the power guys coming up in the next few years.

  • Interesting comparison. I would have thought speed players would be more affected as well, especially since their power remains steady while their speed presumably decreases over time.

    I wonder if maybe that some of the maturity in terms of build and approach compensate with some added strength, much as what has happened with Starlin Castro, albeit on an accelerated scale.

    We no longer see Castro stretch out singles into doubles but we have seen him hit more traditional doubles.

    Also, what of the effect of triples? Could it be that what used to be triples for aging speed players now are only doubles?

  • In reply to John Arguello:

    I think your last paragraph about triples becoming doubles for speed guys is highly significant. I found the data interesting and surprising.

  • In reply to John Arguello:

    John, I think Castro's decline in stretched out singles and doubles this year has more to do with his hamstring or leg problems, which I believe kept him out of most of spring training. I think he is under orders to lay low on the extra kick until possibly another year when a exercise program will strengthen his legs.

  • In reply to Leo the Lip:

    Could be right there. Didn't think of that. However, he does look bigger to me and he does seem to be running as hard as ever to 1B, just not as fast.

  • In reply to John Arguello:

    RE: "he does look bigger to me"

    Me too, especially the lower half (legs and backside)

  • In reply to Ghost Dawg:


  • fb_avatar
    In reply to Ghost Dawg:

    I guess I'm not observant enough of Castro's backside. LOL

  • Interesting results Mike.

    Have you thought about trying to do some comparisons - especially for the 'power' hitter category to see if there is some sort of inverse relationship between the absolute numbers of HRs hit Vs the numbers of 2B hit as they go from young -> prime -> aging? It seems to me that the absolute numbers of extra-base hits for many players seem to hold fairly constant but there is a shift from hitting 2B -> HR -> back to 2B as they age.

  • fb_avatar
    In reply to drkazmd65:

    That's what I thought as well. But it actually didn't work out that way. If you put the doubles line and home run lines in the same graph, the doubles line largely mirrors the home run line, without about 5 more per season at every age.

  • In reply to Mike Moody:

    Interesting,.... thanks Mike.

  • fb_avatar

    First off, how did you define "Power" and "Speed"? Just curious. It sounds like you put players into two distinct categories.

  • fb_avatar
    In reply to Joel Mayer:

    They're rough but I thought fair measures. A power hitter is defined as someone who finished in the top 20%of all home run hitters (min 400 PAs) in any season when he was 26-30. A speed guy is defined as someone who finished in the top 15% in either SB or IFH/GB (same stat as 2 weeks ago) in his young years -- 22 to 26.

    The real benefit to those measures, from the perspective of this article, is neither one is defined based on the number of doubles they hit. Where there is more work to do is to find ways to measure the tools objectively and see how they play out over time. That's the silver bullet. Right now my bullet is kinda iron-ish.

  • In reply to Mike Moody:

    I'm guessing the non-power and non-speed guys have the same type of curves though. I bet the "power/speed guys get more doubles" is one of those long standing fallacies that isn't supported by fact.

  • fb_avatar
    In reply to Mike Moody:

    That works for cut offs.

    Now that brings up another question. Did you have any players that actually fit the bill for both lists resulting in a "Double count" (and a player we might want to sign?!).

  • In reply to Joel Mayer:

    You mean like Alfonso Soriano right? ;)

  • fb_avatar
    In reply to Cubswin4harry:

    Not the 2013 version of Soriano, but yes, that kind of player.

  • fb_avatar

    Also, I think you are right about the logic of having guys hit their prime at around the same time.

    2011: (Inherited) Raw, high ceiling slugger
    2012: (relatively polished) HS hitter with good defense
    2013: (Polished) College Power Hitter
    2014: (Polished) College Power Hitter

    At least in the first round this is how the picks went. When the college players were drafted the HS guys had 2 years of experience in the minors. Now all their top-8 (virtually all lists agree on the ingredients, debate is over order) are between the ages of 20-22.

    I am curious if they go HS in the next draft to start another wave or if they go College bat again and fill the rest with VOLUME picks. Either way, it will be fun.

  • In reply to Joel Mayer:

    It's interesting that you brought up the high school drafting. Although it was a good idea in Theo's first draft I think they should just keep picking the best players available from now on. Either they play for the Cubs or are just trade pieces, the value is still there IMO. Since pitching is arguably the weakness in the minors I suspect they will address that again next year with the later picks and draft a hitter in the first round.

  • fb_avatar
    In reply to CubsBuck22:

    I think you are right about BPA but the timing happened to work out really well too.

  • Interesting analysis. I wonder if the declines would be somewhat more pronounced if you removed the main steroid years.

  • I'll re-read this article in a bit. I'm still dealing with the idea of Mike's brother's wife helping with this analysis.

Leave a comment