The (Tentative) Rethinking Basketball Rookie Ranking Framework

This is my explanation of how I created my first rookie rankings this season. Those rankings can be found here.

I return to a January 2008 article by Bradford Doolittle at Basketball Prospectus, which provides some great insight into analyzing rookies in commenting on Kevin Durant and the 2008 NBA Rookie of the Year race:

Any successful team must have players that can create offense. A team of the five most efficient players in the game (call them the Fabricio Obertos) would have a tough time scoring because they rely on these high-usage types to create opportunities for them.

This explains statistically the difference between a guy like Durant and someone like J.J. Redick. We can watch those two play and the difference is obvious. Redick is undersized and slow and even though the Better Basketball folks consider him perhaps the greatest shooter of all time, it doesn't really matter because he can't create his own offense. Meanwhile, the long, athletic Durant can pretty much pull up and take a jump shot any time he wants, which is something he chooses to do a little too often at this early juncture of his career. Durant's usage rate (8.2) dwarfs Redick's (-2.9).

We don't need the numbers to differentiate between Kevin Durant and J.J. Redick, but for a lot of players, the differences are more subtle and usage rate can help sort out the prospects from the suspects. For the most part, the ability to get your own shot (or get to the foul line) is either something you have or you don't. Usage rates certainly fluctuate from season to season but, generally speaking, usage rate tracks an innate skill.

I remember the first game that Michael Jordan played after his first retirement, which was a Sunday afternoon game at Indiana. Jordan went 7-for-28 from the field and the color analyst, who I want to say was Doug Collins, said something like, "You have no idea how much skill is involved in being able to get off 28 shots in an NBA game."

Just to elaborate and make the point more explicitly, a rookie’s ability to produce points might not be quite as important as the means by which the player creates those points. So even if we use points per game as our primary point of reference, paying attention to what the player actually does once they have the ball in their hands may be useful. In some ways it comes down to a matter of creativity – the ability to perceive multiple outcomes to a given situation and having the ability to bring one of them to fruition.

As Doolittle’s article suggests, it might be interesting to think about how a rookie balances creating opportunities with their scoring efficiency, so as not to just valorize ball hogs. Doolittle does that by comparing the usage rates and efficiency rates of NBA rookies. I’m going to do something similar.

Usage rate is the estimated percentage of plays a player "uses" while on the court (click here for description). It gives us an idea of how often a player attempts to create a scoring opportunity while on the court. Just for some perspective, here are the top ten players in usage percentage as of Friday, July 10 this WNBA season, rookies italicized.

Sophia Young 28.43%
Betty Lennox 28.42%
Becky Hammon 28.41%
Angel McCoughtry 27.87%
Deanna Nolan 27.72%
Nicole Powell 27.63%
Kristi Toliver 27.24%
Charde Houston 27.05%
Seimone Augustus 26.70%
Chamique Holdsclaw 26.66%

For the sake of comparison, last year’s final numbers are available at the Lynx statistics site.



I think (hope) we can all agree that this list represents the players who most frequently produce their own offense. But that is not to say that Angel McCoughtry should win rookie of the year – at some point, we need to take into account whether the player is just chucking everything they touch or actually taking good shots. Doolittle uses his own creation of "efficiency rating" and I’m going to use something similar that Bob Chaiken created called "points per zero point possession". Since that label is just confusing, let’s just call it "Chaiken Efficiency Ratio".

Conceptually, Chaiken Efficiency Ratio (description here) is the ratio of points a player is individually responsible for to the possessions that a player is individually responsible for ending without points. It’s a proxy for scoring decision making – if a high usage player is able to create points more often than they waste a scoring opportunity, we can say they are making good scoring decisions.

So here are the top ten as of Friday in Chaiken Efficiency Ratio:

Shameka Christon 3.27
Seimone Augustus 3.23
Lauren Jackson 3.07
Diana Taurasi 2.67
DeWanna Bonner 2.59
Sancho Lyttle 2.58
Cappie Pondexter 2.53
Ruth Riley 2.52
Vanessa Hayden 2.48
Le’Coe Willingham 2.44

So Seimone Augustus becomes an example of what looking at these two metrics together tells us. Prior to her very unfortunate injury, Augustus not only demonstrated the ability to create shots, but she was creating good scoring opportunities for herself rather than wasting a scoring opportunity with a missed shot or turnover. You can put the ball in Augustus’ hands and count on something good happening.

Doolittle also considers a metric named "Wins Added" to see how much (or little) a player has done to help their team win. This is extremely useful to differentiate players who are extremely effective in limited minutes vs. the stars who are consistently able to impact the game in bigger minutes. After all, that’s the greatest thing about sports -- you play to win the game. You don’t play to just play it… and run around jacking up shots. A player who is taking a lot of shots and not contributing to wins is actually harmful.

But instead of using Wins Added, I’m going to go back to using David Sparks' Boxscores, which I used last year (described here). Boxscores measures player value in terms of how their contribution to the team’s overall production relates to the team’s success (wins). It essentially asks the question, what proportion of the team’s victories can be attributed to the player’s statistical production? As a metric that measures the relative contribution of a player to their team, it seems like a good fit to measure a player’s impact. The number is essentially the proportion of the team’s wins that the player is individually responsible for.

Here are the players with the top ten Boxscores this season:

Lauren Jackson 2.26
Diana Taurasi 2.11
Tamika Catchings 2.00
Nicky Anosike 1.99
Cappie Pondexter 1.96
Katie Douglas 1.57
Jia Perkins 1.52
Tammy Sutton-Brown 1.36
Charde Houston 1.32
Sancho Lyttle 1.31

So Doolittle’s framework of statistical analysis might lead us to make this statement about the standard by which we evaluate a rookie’s impact: the best rookies can create their own scoring opportunities – and do so efficiently – while contributing to a team’s wins. Of course, different players will balance those things differently...but the ideal in this framework would include creativity, efficiency, and winning.

However, that does not address the third point I made in my post last week – that the best rookies demonstrate the ability to do something well. Scoring ability is just one skill of many and given that people are likely to focus narrowly on points per game as their criteria for evaluating rookies, a more nuanced understanding of scoring ability is useful. But what other skills might be valuable?

Yes, there is even an interesting statistical answer to that.

A recently published study by Miguel A. Gomez and a group of colleagues from Spain found that the most powerful variables discriminating between starters and non-starters in the WNBA are successful 2-point field goals, successful free throws, and assists – in particular, in the WNBA successful 2-point field goals and assists are strongly associated with winning teams, while successful free throws are strongly associated with starters and non-starters. So rookies who do those things well are particularly valuable to a team’s success.

This idea essentially fits with a lot of the other ways I already use statistics for analyzing point guards and team dynamics. However, I’m going to deviate from Gomez and colleagues’ framework slightly and look at made free throw rate instead of free three percentage as a secondary proxy with which to analyze a player’s ability to make offensive moves that draw fouls.

That free throw rate statistic has to be used with the caveat that the number could theoretically be skewed by players earning three free throws when "idiots" foul someone in the act of shooting a three or when a lead ball handler earns free throws during the bonus. Nevertheless, when looking at who has the highest free throw rates in the league, the players who are most adept at driving and getting to the basket do indeed float to the top.

And last, I’m going to add rebounding as another essential skill that is useful to WNBA teams based on Dean Oliver’s Four Factors, rebounding being the only one of the four factors not discussed thus far. Offensive rebounding in particular is important because it extends possessions and puts additional pressure on the defense, but even more simply, the fact that they are more rare makes players who can do that well more valuable.

Find the first full rookie rankings here.

X
Log In Sign Up

forgot?
Log In Sign Up

Forgot password?

We'll email you a reset link.

If you signed up using a 3rd party account like Facebook or Twitter, please login with it instead.

Forgot password?

Try another email?

Almost done,

Join Swish Appeal

You must be a member of Swish Appeal to participate.

We have our own Community Guidelines at Swish Appeal. You should read them.

Join Swish Appeal

You must be a member of Swish Appeal to participate.

We have our own Community Guidelines at Swish Appeal. You should read them.

Spinner.vc97ec6e

Authenticating

Great!

Choose an available username to complete sign up.

In order to provide our users with a better overall experience, we ask for more information from Facebook when using it to login so that we can learn more about our audience and provide you with the best possible experience. We do not store specific user data and the sharing of it is not required to login with Facebook.

tracking_pixel_9347_tracker