When the Spurs topped the Kings on Monday Night, one particular statistic caught my eye: the Spurs shot a blistering 62.5% from 2 and a dismal 26.1% from three, although they did shoot at a decent volume from deep with 23 attempts.
I like to see the Spurs win this way, because three-point shooting percentage is extremely variable, and has far more to do with chance than two-point shooting does. Consider that uncontested layups are made at well over a 90% rate, whereas contested turnaround baseline jumpers from 17 feet are connected on very infrequently.
On the other hand, while uncontested catch-and-shoot threes are made at a higher clip than contested or off-the-dribble shots, the difference is much, much smaller. Rather than being dazzled by a high three-point shooting percentage for a given game, or discouraged by a low one, perhaps a better measure of an offense's performance is the number of three pointers attempted.
An offense which generates a lot of looks from three may experience huge fluctuations in single-game three point shooting percentage, but over the course of a season those numbers will normalize. In other words, a three-point attempt is more or less a three-point attempt, so measuring the number of looks afforded in an individual game is a better proxy for three-point proficiency than is three-point percentage itself.
By contrast, a two-point attempt is most definitely not a two-point attempt. The single biggest factor in two-point shooting percentage is the kind of looks generated - a fast-break dunk is rarely missed, and those contested turnaround fadeaway jumpers are rarely made. So at the individual game level, 2-point shooting percentage is a fine proxy for offensive proficiency, since as we've seen with the 2013-2014 Spurs, good offense tends to generate two-point shots which are rarely missed, whereas even a humming offensive machine is greatly impacted by game-to-game three point shooting percentages.
Compounding this effect is the fact that teams generally shoot about 3 times as many twos as they do threes, so the sample size is much smaller for threes, resulting in even more variability. So three-point FG% is the single greatest chance-based factor in a game's outcome.
On the defensive side of things, you want to look at the inverse: if your opponent goes 10/33 from deep, rather than celebrating your team's amazing three-point defense, you should perhaps be concerned that the other team had 33 three point opportunities.
Of course, the degree to which three pointers are contested is a very important factor, as completely uncontested threes are death, but the statistics on contested vs uncontested shots that are available today do not break it down by the type of shot, so it's not easy to measure how well a team contested their opponent's three-point shots. Generally, if a team is failing to contest threes well, that will be captured by more attempts by the opponent, so I would still argue that three-point volume is a better measure of a team's defense than is three-point FG%.
Which brings us back to the Spurs' victory over the Kings. Although the Spurs gave up a solid 36.8% 3pt FG%, the Kings were only able to take 19 shots, which means the Spurs probably did a pretty decent job of contesting those threes. From inside the arc, the Spurs crushed the Kings, holding them to just 41.5% on their two-point attempts.
Here's a helpful chart:
|2pt FG%||3pt volume|
This kind of advantage is what you hope to see on a nightly basis, as it means that your offense is generating better looks than your opponent. Variance in highly chance-based factors like three-point shooting percentage can cost or win you games, and non-shooting factors like offensive rebounding percentage and turnovers can make for big advantages or disadvantages, but an edge in 2pt FG% and 3pt volume is a recipe for a lot of wins over the course of a season.
All of which to say that fans need not despair when the Spurs fail to rain death from distance. As long as the Spurs hit their 2s and shoot more threes than their opponent, their eventual success is assured. In theory, at least.