Are baseball players getting bigger and slower?

Willie Wilson, where have you gone?  You were tall and skinny (6’3″ 195 lbs.) and you stole bases like they were going out of style.  And every team seemed to have a copy of you.  In fact, the St. Louis Cardinals of the 1980s basically had a whole team full of guys like you.  Stolen bases used to be plentiful in baseball and league leaders used to rack up 100 of them in a season.  And back then, 30 home runs might just win you the home run crown.
But, now chicks dig the long ball and so big bulky guys who can drive the ball populate the rosters.  Big bulky guys who can’t run.  But my can they hit!  Some commenters have blamed this as the reason that the stolen base seems to be a lost art nowadays, compared to the heyday of Willie Wilson/Vince Coleman/Tim Raines.  Teams favor big mashers.  But then, there were some big mashers in the 80s, right?  Has baseball really become a big man’s game in the last few decades?  Dan Fox did some work on a similar question in Baseball Prospectus (subscription needed, and speaking as someone who has no vested interest at all in BP, other than being a subscriber myself, it’s worth it) this past January.  He looked at triples and body-mass index (BMI) over the course of the 100+ years of major league baseball being played and found that as players have gotten bigger over time, triple rates went down.  But has there been a specific shift in player size since the 1980s?
Well, let’s pop open the Lahman master data base and see what we can find.  One of the 500,000,000 things to be found in the Lahman data set is a listing of all players who have ever played the game, most with height and weight information.  There’s only one height/weight listing per player.  So if you used to be a skinny-as-a-toothpick guy, but put on a lot of muscle mass quickly, you’ll forever be listed at one weight only.  (By the way, I’m not trying to imply anything with that link.  Really.  Seriously.  What?)   That’s a drawback here, but so it goes.
Let me introduce the concept of body-mass index more fully.  Some of you may be familiar with it, but for those who aren’t, it’s a measure of the ratio of weight to height.  Just knowing someone’s weight doesn’t tell you much.  (Think of a guy who is 5’3″ and 190 lbs. versus a guy who is 6’3″ and 190 lbs.)  In fact, in medical research, we find that it’s this ratio of weight to height that best predicts a lot of weight-influenced outcomes, such as risk for heart trouble.
The standard formula for BMI is weight (given in kilograms… I know, you think the metric system is evil, even though it makes a lot more sense.  And you have the same argument about the DH) divided by height squared (with height given in meters).  In general, it’s best to be between 19 and 25 on this scale for health reasons, although it’s been found that men can go up to 26 with relatively few bad effects.  BMIs from 26-30 are considered overweight and 30 and above is considered obese.  (PSA: Need to check for your own health?  Here’s a good and easy to use BMI calculator that uses inches and pounds.)
I took players who were active during each of the year from 1970-2005 and calculated the league average BMI per year.  Want to see a pretty picture?
Sure enough, the average BMI of players was lowest in the early to mid-80s and began rising throughout the course of the decade.  I’m not sure what to make of that sudden drop in 1994 (The strike apparently was so tough that it made some of the players have to cut back on food and subsequently, they must have lost some weight!).  The drop clearly didn’t last long.  In 2005, league wide BMI was back to all-time high levels. 
So, the intuition is right.  Players really were skinnier in the 80s.  Does it mean that the increasing BMI caused the drop in stolen bases?  It doesn’t necessarily mean that, but it’s a pretty good theory.  A few more pieces of circumstantial evidence point to it.  I chopped players up by their BMI into class intervals of 1 (that is, everyone from a BMI of 22 to 22.999 were grouped together) and selected for players who had at least 100 AB (to get rid of the pitchers and cup-of-coffee guys).  I ran a one-way ANOVA on season stolen base totals (inelegant and could be more precise, but you get the idea.)  Looking at the averages, we see that skinnier players steal more bases, and most of those differences are significant.
BMI               Avg. SB
0-21.999      14.27
22-22.999    13.16
23-23.999    8.85
24-24.999    7.32
25-25.999    6.75
26-inf            5.51
I could reproduce a similar chart going in the opposite direction with home runs.  So, on a very broad level, skinny guys steal bases and big guys hit home runs.  This doesn’t surprise anyone.  Sure, there are guys who can do both (and guys who can’t do either), and there are some little guys with big sticks, but most sluggers are rather big guys. 
Well, we’ve seen that the average player is getting bigger (and I don’t just mean taller and heavier), but are these bigger players getting more plate appearances?  Let me break out the fingerpaint and show you.  The following graph shows the percentage of league-wide plate appearances given to players broken down by BMI class in a given year.  The classes 1-6 correspond as they do in the table above.
Teams appear to always have been fond of big guys, but in the mid-to-late-80s there was a sharp increase in the amount of playing time given to especially big guys.  The peak for skinny guys, such as it was, came in the early 80s, which corresponds, not surprisingly, to the heyday of the Wilson/Raines/Henderson/Coleman era.
Again, in this data set, a player remains in whatever BMI bracket he’s listed in for his entire career, even if in reality he  put on or took off some weight from year-to-year.  But teams do draft, develop, and perhaps promote based on body type.  It looks like they’ve made a conscious (or perhaps not so conscious) decision to give more at-bats to big powerful guys.  It would be wrong to say that the general population of people interested in baseball is getting slower (a plausibe theory, but more evidence is needed).  In fact, there’s probably plenty of fast skinny guys around who want to play baseball.  
Here’s what I think is going on.  Teams control whom they draft, who they have on their rosters, and how they pass out at bats.  Whether it’s because home runs sell more tickets or GMs believe that home runs win games, it’s a real pattern that’s emerged over the last 15 years.  Perhaps a few other GMs will come along and build a team like the Angels are built now (with a bunch of speedsters and Vlad) and focus their development on skinny, fast guys.  But, for now, the way to get playing time as a hitter is to be a big guy, presumably one who hits home runs.
Which brings me to an interesting point.  Not to step on J.C.’s toes, but does anyone else see an interesting economic explanation on why hitters might use Vitamin S nowadays?  Perhaps MLB teams have created the very monster that they seek to slay.  It also speaks to the home run “explosion.”  Sure, the parks are smaller, the pitching has been expansioned to death, blah blah blah, but there’s a very real physical shift to consider as well.  More home runs are being hit because teams are more and more prioritizing hitters who are physically more likely to hit home runs.
Are players getting bigger and slower?  Well, the ones who are getting playing time sure are.


9 Responses to Are baseball players getting bigger and slower?

  1. Derek Nelson says:

    I can’t believe this hasn’t been commented on yet — this is a truly great study, kind of reaffirming what most had an inkling of.
    I can’t believe Willie stole 83 that year!

  2. John Beamer says:

    Interesting study. I’m not sure this has that much effect on the power surge we saw in the 1990s.
    First, the rise in BMI happened before 1993 and was static — in fact even dropped in some parts. Second, I’m not sure how significant the rise is — is a change of 0.2 a large change? Third, wouldn’t we expect BMI to naturally increase over time as humans evolve and become bigger, and stronger …
    Also it’s like a market. Everyone knows that home runs are worth more than stolen bases so it is natural for clubs to stock up with more bulky players which means that young hitters want to have more power so try to bulk up etc …

  3. Pizza Cutter says:

    Considering that almost all MLB players are within the 22-26 range, a change of 0.2 is about 5% of the range, which is pretty hefty. Put it to you this way, on a 6’0″ human, a BMI change of 0.2 is equal to around 3 pounds. Humans will “evolve” (usually this is technology, including medical techonolgy, evolving, not humans) to be bigger, but not that much bigger that quickly. My guess is that most of that 3 lbs is muscle mass. It’s a subtle effect, but it’s there. Teams are stocking up on bulkier guys at the expense of thin speedsters. It doesn’t explain the whole of the power surge, but I think it does explain some of it.

  4. Pizza Cutter says:

    Guy, good points. I might just turn my attention that now that finals are over! There might a time-lag factor between when the players come up and when they start hitting home runs. For example, a player gets a late season call up when he’s really just a AAA player who’s showing some promise and getting a cup-of-coffee call up. After a few years of development, he goes on to develop into a legit power hitter.
    Good ideas all around. I’ll take a look.

  5. Gregor says:

    always good to meet someone who likes Kurt Vonnegut as well – and who remembers the way he loves to end paragraphs. A pity that Kurt recently died. The world will miss him and his iron clad humour – but so it goes…

  6. Guy says:

    Interesting study. If you’re so inclined, you should look at this by position. My perception is that 2B and SSs especially got a lot bigger starting in the early 80s, and a lot of the increase may be found at those positions. But maybe the trend was everywhere on the field.
    It seems unlikely that BMI has had any impact on HR and SB rates, since the BMI change you find takes place mainly between 1980 and 1991, while changes in HRs and SBs all took place after that. For example, if you look at SB per 40 PA, it was .82 in 1980, essentially the same (.78) in 1991, but just .66 in 2001. HRs rates of course began to increase in 1993 (except for a one-year jump in 1987), and increased too quickly for BMI to be a big factor in any case. If BMI was a big part of the explanation, we should have seen an impact on HR and SBs by 1991-92, but we don’t.
    You could figure this out by comparing players with similar BMIs. Look at hitters in a given BMI range in 1990-1991 and then again in 200-2001, to see if their HR rates increased and/or SB rates declined. My guess is you will find the rate changes to be nearly as large when you control for BMI as when you don’t.
    In fact, players are probably getting faster, once you control for BMI. If you compare players in 1980 vs. 1990-1991, again controlling for BMI, I think you’ll find that players in each BMI range stole more bases in the later period.

  7. Walk On says:

    Fascinating article. What about other eras. The 50s was all about the longball – especially the Yankees. And the 60s was back to speedsters – the Cardinals and Dodgers for example. One would think player BMI was higher in the 50s than in the 60s.

  8. Robert says:

    Interesting results, but you didn’t include the control. The data are meaningless until you show the same graphs and statistics for the general population. You haven’t asked the question, “Are baseball players bmi’s trending any different than the average joe’s?” While you’re at it, you should compare to the bmi trends of other sports. If you can’t find a sport where the bmi trend differs from baseball I’d say that your data, although correct, is completely uninteresting because it is true of all sports.

  9. Randy says:

    Except the title question is specifically about baseball and how the change in body types has influenced the style of game we see in that particular sport.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: