Alvin Gentry and Ken Pomeroy ought to chat. Earlier this season, the Phoenix Suns head coach opined that opponent three-point percentage is "the worst stat in the NBA," a claim our Dan Feldman investigated. Not long after, Pomeroy published research showing that college teams have essentially no control over the percentage they allow from beyond the arc. Neither statement should come as a surprise to long-time Prospectus readers. In the course of researching the team component SCHOENE projection system, I discovered that there is little correlation between the three-point percentage teams allow in consecutive years, a finding that applied mostly to the outlying Los Angeles Lakers.
When the Lakers continued defying historical trends, however, I had no choice but to revisit my research. As it turned out, I slightly overstated the randomness in three-point defense. It is not, in fact, totally unpredictable. This year's version of SCHOENE incorporated both the three-point percentage teams allowed the previous season as well as their two-point defense, which is indicative of strong rotations that help cut down on threes as well. Still, the dominant factor in three-point projections is league average because they tend to regress so heavily to the mean.
None of this is to say that there is no skill to defending the three-point line. Part of the problem is the smaller sample sizes associated with three-pointers. Over the course of an 82-game season, the typical NBA team attempts about 1,500 three-pointers and more than 5,000 two-pointers. If the season was three times as long, opposing three-point percentages would tend to stabilize. (This also explains why opponent three-point shooting is even more variable in college, with a schedule less than half as long as the NBA's.)
The randomness of defending the three is exacerbated within a single game. Given the average of 18 three-pointers attempted per game, three shots are enough to make the difference between a good night defending the arc and a terrible one. So what can a team control? The number of threes its opponents attempt. As Feldman noted, defensive three attempts tend to be much more consistent from season to season, especially for teams like the San Antonio Spurs. That makes what the Miami Heat is doing this season all the more unusual. Zach Lowe of SI.com pointed out earlier this week that Miami ranks second in the league in the percentage of opponent possessions that end in three-pointers, trailing only the Denver Nuggets.
So far, the Heat's strategy hasn't exactly been a problem overall. Miami ranks fourth in the league in Defensive Rating, an improvement of one spot over last season, when opponents attempted threes at a rate higher than average but not at the very top of the league. Despite the similar results, the Heat's defensive style has been very different. Miami's increased pressure has forced enough turnovers--the third most in the league--to offset declines in shooting defense and defensive rebounding. Given the way those additional steals have translated into easy buckets in transition, the Heat's defense may have improved more this season than the numbers would indicate.
The issue here lies in the topic I discussed yesterday: consistency. Just like teams that attempt a lot of three-pointers, defenses that allow opponents to shoot more threes tend to be subject to wider swings from game to game. As Pomeroy put it, the three-point line is a lottery, and both shooting more threes and allowing more means buying additional tickets.
In Miami's 13 losses, opponents have shot a combined 39.9 percent from beyond the arc, as compared to 36.0 percent in Heat wins. In four of those games, including Miami's last two outings--losses at Oklahoma City--teams have made at least 50 percent of their threes. Now, these results are certainly not stunning. Teams tend to lose when their opponents shoot well from beyond the arc. But by allowing three-point attempts at a high rate, the Heat's defense amplifies the impact of hot-shooting games.
Whether playing the three-point lottery is a good thing or not depends on a team's overall quality, especially in a postseason context. A lesser team can ride a hot streak from beyond the arc to an unlikely upset; witness the 2007 Golden State Warriors, who averaged more than 10 triples a game in knocking of the top-seeded Dallas Mavericks. For a team like Miami that will enter every playoff matchup up to the Eastern Conference Finals as a heavy favorite, variance is generally to be avoided.
The NBA's seven-game series works to the Heat's advantage, surely, but a matchup against a three-happy team--like, say, the Orlando Magic, which attempted 70 triples in two home wins over Miami earlier this season--could be especially interesting.
Ultimately, the Heat is trying to win in a wildly different manner from the rest of the league's elite defenses. While there are four top-10 defenses among the 10 teams that have allowed opponents three-point attempts most frequently (Dallas, Indiana and Memphis are the others), the three teams that rank ahead of Miami in Defensive Rating (Boston, Chicago and Philadelphia) also happen to rank 1-2-3 in fewest threes allowed. The Heat has chosen an alternative path.
Kevin Pelton is an author of Basketball Prospectus.
You can contact Kevin by clicking here or click here to see Kevin's other articles.