Jump to content


Yo! You're not logged in. Why am I seeing this ad?

Photo

Peak offensive performance age has changed


This topic has been archived. This means that you cannot reply to this topic.
13 replies to this topic

#1 OttoC


  • SoSH Member


  • 7,388 posts

Posted 07 September 2005 - 11:59 PM

Ken Cherven presented a paper at this year's SABR convention which questioned the widely-held belief that the peak age for batters is 27 years of age. After examining several offensive metrics for players with at least 5 seasons of 300+ plate appearances, he concluded that the average age for peak OPS is now nearly 30.

Among other things he noted were:

a. the percentage of players in the 2000s whose peak OPS performance came after age 28 rose substantially over that of the 1970s

b. two-thirds of the hitters in the 1970s had their first 300 plate appearances by age 23 while less than 50% of them did so in the 2000s

c. more than 60% of "regular" playing time was logged by 28-and-under players in the 1970s while it was less than 50% of that age group in the 2000s.

He attributes this change to:

a. more players joining the pro ranks after college
b. advances in conditioning
c. an increased emphasis of power over speed
d. the use of DH in the AL, which favors older hitters.

See Research Spotlight - Age vs. Offensive Productivity (Power Point presentation) at
Visual Baseball

#2 singaporesoxfan

  • 3,760 posts

Posted 08 September 2005 - 02:26 AM

Among other things he noted were:

a. the percentage of players in the 2000s whose peak OPS performance came after age 28 rose substantially over that of the 1970s

b. two-thirds of the hitters in the 1970s had their first 300 plate appearances by age 23 while less than 50% of them did so in the 2000s

c. more than 60% of "regular" playing time was logged by 28-and-under players in the 1970s while it was less than 50% of that age group in the 2000s.

<{POST_SNAPBACK}>


It would be interesting to compare if young pitchers were also increasingly being "crowded out" by older pitchers. Points a and b of Cherven's hypothesis (more players joining after college, conditioning) would apply, but not points c and d (power over speed, use of DH).

#3 AZBlue

  • 1,547 posts

Posted 08 September 2005 - 05:24 AM

Could the dramatic increase in salary levels keep older players on rosters longer, leaving fewer openings for younger players?

#4 bowiac


  • I've been living a lie.


  • 9,746 posts

Posted 08 September 2005 - 07:23 AM

Cherven's study isn't the first one to indicate this.

J.C. Bradbury did a study a bit back which indicated the same thing. Bradbury is in my opinion, probably the best publically available saber guy out there right now. He's an economist, and has been using statistical techniques significantly beyond where traditional saber stuff has gone.

Bradbury followed up his research here.

And here.

I theorized earlier that research like this is perhaps why the Red Sox had been signing and looking to sign so many guys past the traditional age 27 peak.

Nonetheless, Cherven's work is a good addition and corroboration of what Bradbury noticed.

Edited by bowiac, 08 September 2005 - 07:23 AM.


#5 OttoC


  • SoSH Member


  • 7,388 posts

Posted 08 September 2005 - 07:32 AM

Could the dramatic increase in salary levels keep older players on rosters longer, leaving fewer openings for younger players?

<{POST_SNAPBACK}>

I think this could be a factor, but I'm not sure you have the right emphasis. What is the role of free agancy in this?

From an owner's point-of-view, replacing high-salaried older players with lower-salaried, younger players would be desirable, but they are finding that in order to retain the less-costly players they have to offer long-term contracts to compensate for loss of free agency rights.

They also find they have to offer long-term contracts to sign experienced players as free agents or to retain older productive players.

On a side, but related note, it strikes me that the role of the farm system has changed dramatically. When owners still had the upper-hand, they relied primarily on their farm system for replacements, and if they were unable to develop a replacement for a specific position they would use the trade route, often trading an older player for whom they did have a replacement.

Today, the farm system seems to be more of a star search. Prospects become currency for trades. Years ago, the Red Sox might have opted to go with Kelly Shoppach and trade 32-year-old Jason Varitek for pitching; today, they dangle Shoppach as bait...better to go with the proven than take a chance with the unknown.

Of course, many years ago there about half as many major-league teams to stock as there are now and there was much less competition from other sports for athletes, so clubs might have several prospects for a given position.

#6 Smiling Joe Hesketh


  • now batting steve sal hiney. the leftfielder, hiney


  • 25,760 posts

Posted 08 September 2005 - 07:40 AM

It seems to me that the idea of baseball players peaking at age 27 was introduced by Bill James (or at least discussed at length by him) in the early 1980s.

At that time, weight training and regimented exercise programs by players were still the exceptions rather than the norms for players. It was still much more common for players to come into spring training out of shape and use ST as their time for getting into game shape as opposed to coming into ST having trained all winter long. There was still a bias against muscular players back then; James mentions Sparky Anderson complaining that Lance Parrish looked like a Soviet weightlifter after adding 10 pounds of muscle through weight-training.

James mentions Brian Downing as one of the first players to use weight training full time, and that Downing transformed himself for a soft catcher with a flabby face into a dead ringer for Christopher Reeve playing Superman. As a result, Downing's career was lengthened significantly.

What I'm trying to say is that the age 27 peak study was undertaken at a time when players had not yet fully embraced full-time training regimens, and that it makes sense that the peak would now be later (because careers are being extended) due to a more fully realized exercize program.

#7 OttoC


  • SoSH Member


  • 7,388 posts

Posted 08 September 2005 - 08:08 AM

What I'm trying to say is that the age 27 peak study was undertaken at a time when players had not yet fully embraced full-time training regimens, and that it makes sense that the peak would now be later (because careers are being extended) due to a more fully realized exercize program.

<{POST_SNAPBACK}>

Yes, I think the important thing here is to realize that the peak age is not what is was 30 years ago. For some people, the 27-year-old peak age is written in stone.

#8 DamonasaNomad

  • 2,613 posts

Posted 08 September 2005 - 08:27 AM

It seems to me that the idea of baseball players peaking at age 27 was introduced by Bill James (or at least discussed at length by him) in the early 1980s.

<{POST_SNAPBACK}>

IIRC, James also speculated that rising salaries could drive up career length, and thus peak age, providing a huge incentive to veterans to both stay in shape and to keep playing.

There's also improved medical procedures, especially shoulder/arm surgery. I wonder what effect it would have on the numbers if one assumed that players who've continued to play after "heroic" surgery had been forced out of baseball instead?

I also notice that no one in this thread has yet mentioned the S word. Barry's late 30s performance alone must be driving peak age up a fortnight or so.

#9 smastroyin


  • simpering whimperer


  • 17,032 posts

Posted 08 September 2005 - 08:35 AM

Yes, I think the important thing here is to realize that the peak age is not what is was 30 years ago. For some people, the 27-year-old peak age is written in stone.

<{POST_SNAPBACK}>


This was always a misinterpretation, anyway. The fact of the study was that the most players peaked at age 27. However there was a fairly even distribution of guys who had peaks from 25-31. 27 just happened to have the highest n.

The curve itself seems to have moved up. However people who held to the dogma that all players or even most players peak at age 27 were always wrong anyway.

Steve

#10 Tudor Fever

  • 3,412 posts

Posted 08 September 2005 - 08:36 AM

Another contributing factor might be that it used to be a lot easier to lie about your age than it is now. It wasn't just the Latin players, either. Jim Bouton talks about this in Ball 4, IIRC about Roland Sheldon in particular.

#11 philly sox fan


  • SoSH Member


  • 9,748 posts

Posted 08 September 2005 - 08:48 AM

This was always a misinterpretation, anyway.  The fact of the study was that the most players peaked at age 27.  However there was a fairly even distribution of guys who had peaks from 25-31.  27 just happened to have the highest n.

The curve itself seems to have moved up.  However people who held to the dogma that all players or even most players peak at age 27 were always wrong anyway.

Steve

<{POST_SNAPBACK}>


That's the big thing for me. I still have an image of the original James bar chart in my head. The age 27 peak was a little higher than the peaks surrounding it, but there was a realtively even distrbution throughout the mid-20 to very early 30s.

The age 27 advantage was so small that I believe it was true that more players peaked at age 31/32 combined (the wrong side of 30!) than at age 27 itself.

The importance of single peak year performances in these kinds of studies has always been overplayed imo.

#12 DieHard3


  • SoSH Member


  • PipPipPip
  • 8,864 posts

Posted 08 September 2005 - 08:53 AM

I noted this two years ago on SOSH, just using a quick and dirty sample of players from Win Shares. Glad to see that more rigorous research backs up that study I did.

#13 Arock78

  • 1,953 posts

Posted 08 September 2005 - 04:19 PM

Going a little bit out on a limb here, but isn't it also possible that since hitters have increasingly been moving away from a see-and-hit approach at the plate and toward a more patient approach, that hitters are using skills that are less likely to wane anyway? It always seemed to me that a hitter who goes solely on his physical abilities will start to decline earlier than a hitter who makes opposing pitchers throw them balls he can handle.

I remember Nomar's abilities were explained in this way. He was able to hold off on swinging until later than other hitters because he had phenomenal physical reflexes and great bat speed. I wondered what would happen when he got a little older and didn't have the same physical abilities. And I was concerned for him that when those physical abilities started to wane, his performance would suffer greatly.

#14 Frisbetarian


  • ♫ ♫ ♫ ♫ ♫ ♫


  • 4,681 posts

Posted 08 September 2005 - 05:56 PM

First, I'd like to thank Bowiac for the J. C. Bradbury links. I'd never read him before, and he is excellent.

The most interesting detail in his three part article on peak offensive performance was that the peak age of superstars (OPS+>120) was 31-32, much older than other players. Certainly, advanced training habits, better nutrition, and medical progress have contributed to an older peak age, but I wonder whether current offensive era which yields more players with very high OPS+ has somewhat skewed the "average" for all players?

Yo la tengo,
Fris