NBA Study: What is the right way to analyze rookies?

Key Highlights:

• Since young players are inherently inconsistent, the best way to analyze their rookie seasons is to look at their flashes
• According to Game Score, Victor Wembanyama, Chet Holmgren, Brandon Miller, Keyonte George, and GG Jackson have had the top 5 best flashes among rookies this year
• The best way to use this study is to compare players that share similar roles/usage to each other

They say you shouldn’t judge a book by its cover. In a lot of ways, the same is true of rookie seasons. A player’s inaugural season is hardly ever a summary of their entire career. Rather, it is merely the beginning of their NBA story.

But what people often don’t mention is that you can glean some insight about the contents of a novel based on its cover. The same is also true for players and their rookie seasons. But what is it exactly that you can extract?

In this edition of “NBA Study”, we look at one (not the only) way to analyze a player’s rookie season.

Our Process

In his book (one that you should read), “Basketball Analytics: Spatial Tracking,” Dr. Stephen Shea wrote this in his chapter about evaluating prospects:

“College players are inconsistent. Their production can vary wildly from one game to the next. This is especially true for freshmen, and it is often after the freshman season that the most elite prospects enter the draft…The moral is that consistency is not something we should expect from even the very best of prospects. It is something that players gain with experience; it is something they can be taught.”

Young players are inherently inconsistent. But that’s okay because consistency can be learned. So, what we should focus on instead are a player’s flashes, since those flashes are what the player will be regularly if/when they master the art of consistency.

How To Measure Consistency

It’s been about two years since I read Shea’s book. And in that time, the best method I’ve been able to come up with for quantifying flashes is Basketball Reference’s Game Score metric. According to their website, Game Score is a metric “created by John Hollinger to give a rough measure of a player’s productivity for a single game. The scale is similar to that of points scored (40 is an outstanding performance, 10 is an average performance, etc.).”

What we will do is take the average of each rookie’s ten best Game Score games and compare that number to their peers. Our threshold to qualify for this study is at least 30 games played. That means we’ll be looking at the ten best Game Score games average of 34 rookies.

To be fully transparent, I’ve only run this test for a couple of seasons. So, the true validity of this study is still yet to be fully proven. However, last year’s version of this study featured Jalen Williams and Paolo Banchero as the two rookies with the best flashes (according to Game Score). And considering one of them is an All-Star (Banchero) and the other is playing at a borderline All-Star level (Williams), that is a pretty good sign.

Anyway, here are the results for this year’s rookies (as of March 26, 2024):

2023-24 Rookies Best 10 Game Score (GS) Games Average*

*Data Provided by Basketball Reference

It should be no surprise here that Victor Wembanyama (who is having one of the best rookie seasons of the last three decades) is first, with the other unicorn, Chet Holmgren, trailing behind him. It also makes sense that two-time Rookie of the Month winners Brandon Miller and Jaime Jaquez Jr. are in the top 6 on this list.

How To Interpret These Results

Two problems I’ve noticed with using Game Score is that it biases toward players who play a lot of minutes and carry a large offensive role on their team (since the metric uses box score data). So, the best way to use these results is to compare players who play similar minutes/positions.

For instance, one thing I found incredibly fascinating is that between Keyonte George and Scoot Henderson (two on-ball guards getting a bounty of repetitions on tanking teams), George is the one with the higher Game Score average (19.7).

Another data point that stands out to me is that Bilal Coulibaly’s Game Score average (12.7) is lower than other defense-first wings and forwards like GG Jackson (19.7), Amen (19.1) and Ausar Thompson (18), and Toumani Camara (13.4).

The Bottom Line

As I mentioned in the introduction, this is just one way to evaluate a player’s rookie season. You still need to consider factors like injuries, team context, and how certain players may be affected by what the metric is measuring.

As readers of this website know, I am fairly high on Camara. But he only ranks 21st out of the 34 rookies on this list. If I had to guess, it’s because Game Score doesn’t pick up on all his contributions to the defensive end of the floor (95th percentile in Defensive Estimated Plus-Minus).

This doesn’t mean that you shouldn’t take away anything from these results. Dr. Shea is a smart dude, and I wouldn’t write this article if I didn’t think it had some value. It just isn’t an end all be all.

It’s another tool in our rookie interpretation toolbox.

If you enjoyed this edition of “NBA Study,” check out our other articles, like this one on which coaches are the best at calling plays.