Advanced Stats 101: Player Shooting

In this space last week, we explored some basics regarding APBRmetrics and their application when considering team shooting. Let us now consider the same three metrics used to evaluate team performances and examine how they pertain to individual shooters.

Prominent APBRmetricians such as Bob Bellotti and Kevin Broom are among the growing number of statisticians who support the use and development of new, intricate basketball statistics. As discussed previously, these dynamic metrics provide more precise insight that extends beyond what traditional measures consider. One segment of the sport that lends itself particularly well to this analysis is player shooting. Exploring the significance of variance between new and old measures shows us how appealing it is to shift the standard paradigm for player stats.

First, let’s review the league leaders for this season in Traditional Field Goal Percentage:

TRADITIONAL FIELD GOAL PERCENTAGE
RNK NAME FG%
1 A. Biedrins, GSW 0.626
2 T. Chandler, NOR 0.623
3 D. Howard, ORL 0.599
4 S. O’Neal, MIA/PHO 0.593
5 A. Stoudemire, PHO 0.590
6 J. Childress, ATL 0.571
7 R. Brewer, UTH 0.558
8 D. Lee, NYK 0.552
9 C. Boozer, UTH 0.547
10 K. Garnett, BOS 0.539

This list is a good reflection of what one expects to see with FG% leaders, as the strong majority of the players in the top ten play close to the basket. It is self-evident that layups and dunks are the easiest shots to make on the court, for their proximity to the basket makes them less challenging. Bigger players then should, and do, dominate FG%.

If you look at the Top 20 in the league in Traditional FG%, you see that only three are true guards (7. Ronnie Brewer, 14. Monta Ellis, 18. Jose Calderon). Yet no one would suggest that Shaquille O’Neal is a better shooter than Steve Nash, even though Shaq is twenty spots ahead of him in league rankings.

Knowing then that this standardized stat favours big men, let us look at the league leaders using Adjusted Field Goal Percentage [AFG%: = [(PTS - FTM)/FGA]/2].

ADJUSTED FIELD GOAL PERCENTAGE
RNK NAME AFG%
1 A. Biedrins, GSW 0.626
2 T. Chandler, NOR 0.623
3 D. Howard, ORL 0.599
4 S. Nash, PHO 0.597
5 M. Miller, MEM 0.596
6 S. O’Neal, MIA/PHO 0.593
7 A. Stoudemire, PHO 0.592
8 J. Childress, ATL 0.590
9 J. Smith, DEN 0.578
10 J. Calderon, TOR 0.575

The intention of this adjustment is largely to evaluate the impact of three-point shooting and the considerable reorganization of the list suggests it does just that. While the top three performers remain unchanged, Nash catapults up the list from 25th to fourth as a result of considering the relative difficulty of threes.

Mike Miller, another player revered mainly for his outside shooting touch, jumps from 27th on the conventional list to fifth.

Perhaps no two players better display the degree this stat adjusts customary measures than J.R. Smith and Anthony Parker; the Nuggets’ Smith flies up from 66th in the league in FG% all the way to ninth overall using AFG%, while Toronto’s Parker moves from 50th to 12th.

In general this adjustment succeeds in propelling guards and outside shooters up the standings. In this way it gives a more accurate and true-to-life reading on a player’s shooting proficiency.

Having seen what that adjustment did to alter the statistical shooting landscape, it would prove beneficial to also give thought to True Shooting Percentage. As detailed last week, it is my opinion that TS% is the most accurate and important measure in use today with respect to gauging shooting effectiveness. TS% [(Total points x 50) divided by [(FGA + (FTA x 0.44)] accounts for free throws and three-pointers to give a more complete picture of a player’s all-around shooting ability.

Consider now then the league leaders using TS%:

TRUE SHOOTING PERCENTAGE
RNK Player TS%
1 Andrew Bynum, LAL 0.659
2 Amare Stoudemire, PHO 0.656
3 Brent Barry, SAS 0.655
4 Josh Childress, ATL 0.647
5 Erick Dampier, DAL 0.645
6 Steve Nash, PHO 0.641
7 Carl Landry, HOU 0.641
8 Andris Biedrins, GSW 0.637
9 Kendrick Perkins, BOS 0.632
10 Tyson Chandler, NOR 0.632

Bynum’s appearance atop the list is somewhat misleading: his injury caused him to miss too many games to be considered eligible for the league’s standard FG% list, otherwise his .636 FG% mark would have led the NBA.

Bynum aside, what is interesting is how far Biedrins, Chandler and Howard fall. AFG% did not account for free throws, meaning that these post players were not penalized for their awful foul shooting (.62, .59, and .59 respectfully). When FT% is accounted for, each drops noticeably, with Howard falling all the way to 14th.

No player is hit harder by this new measure than Shaquille O’Neal, the posterboy for awful foul shooting (O’Neal drops all the way to 36th with TS%).

Meanwhile, Brent Barry represents the other side of the coin as his .481 FG% would place him 45th in the league normally (he did not have enough attempts to qualify), but as a result of his .429 3P% and other-worldly .950 FT% he would place third in TS%.

This trend is expressed elsewhere in the Top 20 with the bump received by Manu Ginobili, Sasha Vujacic and Kevin Martin.

Allowing the assertion that True Shooting Percentage is the best overall measure of shooting proficiency, have a look at the Active Career Leaders in TS%:

CAREER ACTIVE LEADER
Rank Player TS%
1 Brent Barry 0.6074
2 Steve Nash 0.6028
3 Amare Stoudemire 0.6025
4 Dwight Howard 0.5962
5 Yao Ming 0.592

At first glace, Barry’s topping this list seems odd. This is telling of how traditional measures are incomplete or lacking. When you consider how amazing he has been from the free-throw line throughout his career and his long distance touch, it suddenly seems obvious he would be so high, yet classic stats and subjective analysis don’t let us appreciate that.

Barry has never had the notoriety of a pure shooter in the way that Ray Allen has. Observers have long canonized Allen’s textbook mechanics and stroke: his jumper is a thing of absolute beauty and his name is the most common reply to the question, “Who has the best jump shot in basketball?”

But, if we remove the aesthetic component of his form and look exclusively at percentages, does Allen fall short of Barry, Nash and others?

His traditional field goal percentage of .445 places him only 90th in the league, suggesting that perhaps his days as a premier gunner have past him by. But a closer look shows his merit: Allen’s Adjusted Field Goal Percentage is a lofty .537, good for 28th overall. This shows how this APBRmetric equates for the impact of strong three-point shooting as of greater difficulty than, say, Tyson Chandler’s dunks.

Allen’s sterling free-throw percentage (.907) bolsters his True Shooting Percentage to an impressive .584, good for 45th in the league. Allen shot only .398 from three last season, placing him a surprisingly slow 35th in the league. That said, he still made the fifth most threes leaguewide, so his lower percentage may be a product of the higher volume of shots he is expected to take compared to players like Jason Kapono.

Still though, an observer can’t help but notice that his career .571 TS% mark puts him only 19th on the active list, behind the likes of Mike Miller and Wally Szczerbiak.

Therein lies the beauty of such advanced stats: his fame and shooting form have made Allen the best shooter in the game by subjective measures, but objectively this can be contested.

In conclusion, APBRmetrics do a great deal to open our eyes to unconsidered possibilities with respect to shooter comparisons. Marquee names like Michael Redd gain credence via their reputation as pure shooters, yet at times this is misleading (Redd’s .559 TS% is good for only good for 92nd overall and is comparable to that of Chris Quinn). Notable shooters like Ben Gordon have a higher reputation than a player like Mike Miller, but advancing stats show us that maybe they shouldn’t.

Comments (4)

  1. IMHO no one who considers himself to be a basketball expert would use FG% to illustrate an individual’s overall ability to shoot the ball in the first place. ‘New Age Stats’ like TS% attempt to quantify what actual basketball experts have known and been able to judge correctly for a long time by simply watching the game closely and attending accurately what specific players actually do well (or not) when they’re on the court. Even so, a simple stat like TS% is also flawed as it fails to account for innumerable other variables as well which are important to an accurate assessment of which players are the best shooters overall within a specific set of individuals. Human observation by a skilled practicioner remains the best judge of basketball ability not statistical analysis.

Leave a Reply