The (Non) Impact of Age on Games Played Among NBA All-Stars
A quick analysis
Hey everyone! I wanted to try something different for this post. Generally, I publish long-form content that is well thought out and researched. Today, I’m sharing some ad hoc number crunching I did over the course of an evening that was done more out of curiosity.
published a post the other day about NBA star absences being way up this year, mainly due to injury. I have previously written about load management in the NBA and found that most injuries occur not because of a lack of load management. Instead, injuries happen when exertion of the body happens without a gradual process to work up to high amounts of load.
In other words, it’s a bit like being out of shape and being asked to participate in strenuous physical activity. It is why when runners train for a marathon, they gradually work up to running 26.2 miles instead of running 26.2 miles from day one.
In his post, Haberstroh states that injuries are up 52% year over year. He leaves the reason for all the injuries up in the air, citing a variety of potential reasons.
Aside from possible reasons cited in my post above, I wanted to see if one reason there have been more injuries is the aging of NBA stars. Anecdotally, this feels right. The older generation of stars, led by LeBron James, Kevin Durant, etc., are all getting up there in age. You would assume missing games due to injury is a logical result of aging bodies.
I’m not sure how Haberstroh defined stars, but I defined it by those who made an All-Star game in a particular year. I know this definition is probably a bit more stringent and by definition skews the data towards more games played because, well, you have to play to make an All-Star team. Nonetheless, this was the best way I knew how to define stars without getting too subjective.
I looked at the data over the past few seasons (2017-2024—I could have done more, but the data collection was very manual) and found that the mean age of NBA All-Stars has increased since COVID. The years before COVID saw an average age of NBA all-stars around 27 years old. Since COVID, the mean age is closer to 27.25-27.5. However, this increase is driven by LeBron James. When I removed LeBron from the mix, the mean age went back to about 27.
Instead of looking at why an All-Star missed a particular game between 2017-2024, I decided to use the number of games played as a proxy metric. Now, I know injuries aren't the only reason why a player doesn’t play a game, especially in today’s NBA, but I figured it would be more efficient to collect the data this way.
I found that the mean and median number of games played by All-Stars has decreased from the mid-70s (~90% of games) to the mid-60s (~79% of games) from 2017-2024. One logical answer to this is the increased prevalence of load management.
Next, I looked to see if age impacts the number of games played by All-Stars. I found that as NBA All-Stars get older, they tend to play fewer games, (slightly), but the relationship is extremely weak. In fact, age only explains 1.3% of the variation in games played, suggesting other factors are much more influential.
I didn’t find anything too groundbreaking here, but we have a strong signal that age doesn’t explain the number of games played among All-Stars. If someone wanted, they could expand on the dataset and look at additional variables and relationships. That’s all for today—have a great week and enjoy your sports viewing.
I wonder what would happen if you did this looking at games played the NEXT season for All-Stars from a given season? As you say, there's a sort of circular selection effect going on here by which a player won't make an All-Star roster if he doesn't actually play. But looking at the following season could help get around that.