Last week, in a piece for Hardwood Paroxysm, I explored the idea that new developments in nutrition and physical training have made it possible for older players to remain more productive, and for a longer period of time, than was previously possible. The numbers showed that over the past 30 years the peak in both the number of players over-35 in the NBA, and their overall productivity, came not in the past few seasons but during a stretch from 1997 to 2003. The degree to which players like Tim Duncan, Steve Nash and Kevin Garnetthave been able to stay among the league’s elite is not entirely breaking new ground. Instead it’s a journey along the trail broken by the likes of Karl Malone, John Stockton, David Robinson, Kareem Abdul-Jabar, Robert Parish, Reggie Miller and Hakeem Olajuwon.
The analysis I did for the Hardwood Paroxysm piece only looked at the over-35 segment of the NBA population. I thought it might be interesting to look at how NBA players of other ages have faired over the past three decades.
I broke the NBA population into 4 age groups – 21 and under, 22 to 28, 29 to 34, 35 and older. Analyzing the demographics by season was incredibly labor intensive so instead of looking at each of the last 30 seasons, I looked at representative samples, each five years apart. For each season I looked three different areas by age group – the percentage of total players in the league, the percentage of minutes played, and as a measure of productivity, the percentage of total Win Shares.
This first graph shows how the league broke down in terms of the age of the players.
There have been fluctuations from season to season, but over the past 30 years the number of players in the 22-28 age group has continually shrunk. As that age group has decreased the other three have grown, particularly the oldest and youngest age groups. However, just looking at the percentage of roster spots held by each age group doesn’t account for specific roles reserved for players of a certain age. Plenty of teams acquire veterans, happy to stash them at the end of the bench for the benefit of leadership in the locker room. You’ll also see plenty of teams with very young players on the roster, ones who’s development fits snugly into future plans but has very little to do with what happens during that season.
A more telling way to slice these demographics is to look at how league minutes were divided up among these age groups, shown in the graph below.
Although not quite as drastic, the pattern here is roughly the same as what we saw when looking at the overall makeup of the league. In the first graph we saw big growth in the oldest and youngest segments of the population. That holds true here, although with one important difference. The growth among players, 21 and under, in both graphs is almost identical. But for the over-35 group, the percentage of minutes played was generally a little less than the percentage of roster spots they held.
Essentially, there are more younger players in the league than there were 30 years ago and their minutes played have grown at essentially the same rate. But for the over-35s the growth in the percentage of roster spots they’ve held has generally outpaced the growth in their minutes played. This would seem to indicate a boom in openings for that “veteran locker-room leader” role.
The final slice is to look at how productive each age group has been. The graph below shows what percentage of the league’s total Win Shares was produced by each age group.
Here the pattern reverses and the 21 and under age group lags slightly. While the percentage of league minutes they’ve played has generally increased, their production hasn’t increased at the same rate. Although their share of the league’s roster spots and minutes played has shrunk, the production from the 22-28 age group continues to outpace the other age groups by a wide margin. Although plenty of attention is paid to the growth in minutes and production by both the oldest and youngest players in the NBA, in each of these seasons roughly 85% of the minutes played and 90% of total Win Share production came from players aged 22-34.
There have certainly been some demographic changes around the edges to the NBA over the past 30 years. However, attention paid to these changes obscures the fact that the bulk of minutes played and overall production is coming from the same age groups that it did three decades ago. As usual the exceptions overshadow the rules.