All scores →
Loading…
Loading…
Loading…

The three-point revolution in the numbers

By 4Basketball Staff · · 11 min read

The three-point shot went from a novelty to two of every five attempts. This article follows the numbers, the simple arithmetic that drove the change, what happened to accuracy and what it did to spacing, rebounding and scoring.

In the 1979-80 season, the first in which the NBA used a three-point line, the average team attempted fewer than three of them per game. Many teams treated the shot as a gimmick borrowed from the rival American Basketball Association, to be used on a desperate final possession or by a specialist who had no other talent. A little more than four decades later, the average team attempts more than ten times as many, and roughly two of every five shots in the league come from beyond the arc. Few changes in the history of any major sport have been so complete, or so thoroughly measured.

This article follows the revolution through the numbers: how fast the volume grew, why the arithmetic of the shot pushed teams that way, what happened to accuracy, which parts of the floor mattered and what the change did to the rest of the game. For the story of how the line came to the league and the personalities involved, see The history of the three-pointer: ABA origins and NBA adoption and The 2010s: the Warriors, LeBron and the three-point revolution. Here the figures lead and the stories follow.

How fast the volume grew

The best way to see the shift is to follow the number of three-point attempts per team per game. The values below are approximate, and they are intended to show the scale of change and not to serve as record-book entries.

Period Three-point attempts per team per game Share of all shots
1979-80 Fewer than 3 A few percent
2010-11 Under 20 About a fifth
Late 2010s More than 30 More than a third
Early 2020s Mid-30s Roughly two in five

The growth was not smooth. It came in a slow climb through the 1980s and into the 1990s, helped for three seasons in the mid-1990s by a shortened line, a change the league reversed afterward. It then stayed modest for years, because the shot was still seen as a complement to the real offense, not its foundation. The steep rise came after about 2012, and it was fast: the volume that took roughly three decades to reach 18 or 20 attempts per game was close to doubled in less than a decade.

Individual records tell the same story. Ray Allen made 269 threes in 2005-06, a record that stood until Stephen Curry made 272 in 2012-13. Curry then raised the mark to 402 in 2015-16, a number that would have looked like science fiction a generation earlier. At the team level, the Houston Rockets of the late 2010s took roughly 45 threes a game in their most extreme season, about two and a half times what the average team attempted a decade earlier.

The arithmetic behind it

The reason for the shift is not mysterious. It is a matter of multiplication. A shot's value is the number of points it is worth times the probability that it goes in. A two-point shot that goes in half the time yields 1.00 point per attempt. A three-point shot needs to go in only one time in three to match that: 33.3 percent times 3 is 1.00. Anything above 33.3 percent beats a 50 percent two.

Compare that with what the league actually shoots. In the era of the revolution, the league-wide three-point percentage has been in the neighborhood of 35 to 36 percent, which yields about 1.05 to 1.08 points per attempt. The average two-point attempt, which mixes close shots with long jumpers, went in somewhat above 50 percent in recent years, which yields roughly 1.0 to 1.1 points. Those numbers are close, and critics sometimes note that the average three is not wildly better than the average two. That is true, but it misses the point, because the average two is a blend of two very different shots. Shots at the rim are highly efficient. Long twos, the jump shots taken just inside the arc, are the least efficient shots in the game. The case against the long two, covered in Why teams stopped taking midrange shots: the efficiency argument, is the other face of the case for the three. If you remove the long twos from the mix, you are left with the rim and the arc, which are the two best places to shoot.

The shortest version of the logic is in Points per possession: why efficiency beats volume: a team should take the shots that produce the most points per attempt, and the best of them are layups, free throws and threes. A good shooter from three is a threat worth more than a good shooter from the midrange, which is why teams began to prize shooting as the central skill of every position.

There is also an effect that goes beyond the shot's own value. A good three-point shooter makes the defense guard him at 24 feet, and every foot a defender moves away from the basket opens space for others. A spread floor makes it harder to help on drives and easier to find openings for cutters. The volume of threes taken understates their influence, because the threat of the shot reshapes the whole possession. That is the theme of Pace and space: how the three-point shot reshaped spacing.

The geometry of the line

The three-point line is not at one distance. In the NBA it is 23 feet 9 inches from the center of the basket at the top of the arc and 22 feet in the corners, where the line runs straight along the sideline. The corner three is therefore the shortest, and it has long been the most accurate three-point attempt, with league-wide success rates a few percentage points above those from the top of the arc. A shot that is 1.75 feet closer is easier, and the numbers reward it. Teams have built offenses to generate corner threes, because they are the best of the long shots: worth three points and easier to make than the others.

This is a reminder that "threes" are not a single product. A wide-open corner shot, a catch-and-shoot from the wing and a heavily contested pull-up from 28 feet are all counted as three-point attempts, and their efficiencies are very different. Tracking data, described in Expected points and shot quality in the tracking era, separates them by how open the shooter was and what kind of shot it was. The overall figure for the league masks that variety.

College basketball changed its line twice in recent years. The men's three-point line was moved back to 20 feet 9 inches before the 2008-09 season, and then to 22 feet 1.75 inches, the international distance, in 2019-20. The effect of each move was a change in the mix and accuracy of long shots. Moving the line back thins out the cheap threes and rewards genuine shooters, which is a way of seeing how sensitive the shot's economics are to distance.

What happened to accuracy

The remarkable fact about the revolution is that accuracy did not collapse as volume soared. A natural expectation is that as a league takes more of a difficult shot, the average success rate falls, since the additional attempts must come from worse shooters and worse looks. Instead, the league-wide three-point percentage held in the mid-30s through the years in which volume roughly tripled.

Several things happened at once. Teams selected better shooters at every position, so the typical player taking a three was a more accomplished shooter than before. Players trained for the shot from a young age, and the quality of shooting form improved. Offenses generated better looks by spacing the floor and moving the ball. And the shots shifted toward catch-and-shoot attempts in good positions. A rise in difficult, self-created threes by the best shooters offset some of that, so the average stayed roughly flat, but the composition improved. The skill of becoming a better shooter, and the mechanics of doing it, is discussed in Three-point shooting: how to build range without ruining your form.

Efficiency overall rose. Team offensive ratings, which measure points per 100 possessions as explained in Offensive and defensive rating: how teams are measured per 100, climbed across the era. The league-average offense of the early 2020s was substantially more efficient than that of a decade earlier, and a good share of the improvement came from the same source: more of the shots were threes and shots at the rim, and fewer were long twos.

The rules that helped

The revolution was not driven by analytics alone. In 2004-05, the NBA changed how it officiated perimeter defense, limiting the hand-checking that had allowed defenders to impede ball handlers with their hands. The change made it easier for guards to get past their defenders and to create open shots, and teams that exploited it, such as the Phoenix Suns of that period, played a fast, three-heavy style that foreshadowed the later era. Their story is touched on in The 2000s: Shaq, Kobe, Duncan and the Spurs dynasty, which covers the decade that also saw the Suns' fast style emerge.

Other developments followed. Defensive three-second rules had changed in the early 2000s, and defenses became more flexible, which in turn pushed offenses to spread the floor. Analytics departments grew in the 2000s and 2010s, and the arithmetic above was no longer an insight but a standard presumption. The success of the Golden State Warriors in the mid-2010s provided a proof of concept that persuaded the rest of the league: a team built around shooting could win championships, not merely entertain.

Why it took so long

If the arithmetic is so simple, why did the league need three decades to act on it? Part of the answer is that the arithmetic was not simple in practice. The average three did not look overwhelmingly better than the average two, and the evidence that long twos were a poor shot required shot-location data that was not widely available until tracking systems arrived. Part is a shortage of shooters: in the 1980s and 1990s, few players at the bigger positions had ever practiced the shot, so a team that wanted to take many threes would have had to field a lineup that did not exist. Part is the nature of coaching, a profession in which a failed experiment is costly and imitation is safe. And part is that the game rewarded other strengths. A team with a dominant post scorer or a great midrange creator was better off using him.

The change needed a pipeline: players who had grown up shooting threes, front offices that could quantify shot value, a few early adopters who showed it worked, and a rules environment that made the perimeter easier to attack. When those pieces were in place, adoption was quick. That pattern, a slow start followed by a rush, is typical of innovations that depend on both knowledge and supply. It also suggests why the shot's share leveled off in the early 2020s: the big gains from converting the least efficient shots had been taken.

Who takes the shots has changed as well. For years, the three was the province of guards and a few specialist forwards. Today, wings, forwards and many centers are expected to take them, and a player who cannot is treated as a limitation in a way he would not have been a generation earlier. The change is a quiet one, but it explains why lineups look so different from those of the 1990s.

What it did to the rest of the game

When a shot grows from a rarity to two out of five attempts, it changes everything around it.

Spacing and driving. With shooters on the perimeter, defenders cannot sag toward the basket, and lanes open for drives. The pick-and-roll became more dangerous because the screener's defender had to worry about a pop-out shooter as well as a roll to the rim. The variations are catalogued in The pick-and-roll: every way to run it and every way to stop it.

Big men. Centers who could not shoot found their roles shrinking, and those who could became valuable. Positions blurred, with taller players expected to handle the ball and shoot, and the style described in Five-out basketball: playing without a traditional center became a viable approach for teams with the right personnel.

Rebounding and transition. Long shots produce long rebounds, which bounce farther from the basket and are harder for offensive players to recover. Teams sent fewer players to the offensive glass and more back on defense. The trade-off is discussed in Offensive rebounding versus getting back: the central trade-off. League-wide offensive rebounding rates have fallen over the long run, and the shape of the game has moved toward shooting and running back.

Defense. Defenses had to defend the entire floor. The old model of protecting the paint and conceding the perimeter no longer worked, so teams invested in closeouts, switching and perimeter defenders who could guard multiple positions. See Defending the three-point line: closeouts, rotations and trade-offs and Switch-everything defense and its weakness: mismatches.

Variance. A game decided by threes is a game with more variance. A team that takes 40 threes has a wider range of outcomes than one that takes 20, since a three-point percentage can swing from 28 to 44 percent on a given night. That helps underdogs and makes some nights look random. It also means that a shooting slump or surge can decide a playoff game, which explains some of the dramatic reversals in recent postseasons.

Does the revolution have a limit?

There are signs of a ceiling. The share of threes is high but not infinite, because a defense that is determined to take away the arc can force offenses to take the shots it prefers. As volume grows, the marginal three is a little worse than the average, and eventually the extra attempt is no better than a good two. Teams are therefore searching for the optimal mix, and the answer has more to do with who is on the floor than with a general rule. A team with elite shooters should shoot more; a team without them should not simply copy the pattern. The recovery of the well-taken midrange shot, among skilled scorers, is part of the same adjustment.

The numbers also remind us that analysis does not eliminate judgment. The arithmetic tells a team that threes are valuable, but not how many its players can take at what quality, not how its defense will hold up and not how its rivals will respond. The next stage of the story is about those questions.

What to take away

The three-point revolution is an arithmetic story with human consequences. A shot that needs only a 33 percent success rate to match a 50 percent two was undervalued for decades, and once teams measured it, the volume rose from under three attempts a game to the mid-30s. Accuracy held steady, spacing improved and offenses became more efficient, at the cost of changes to rebounding, positions and defense. If you want to follow the effect in current games, the scores page shows the results as they come in, and the teams section is the place to look up each club.

Search more on this

Keep reading

Plus-minus, on/off and why the raw number lies

Plus-minus records how a team fares while a player is on the floor. Learn how it, on/off splits and adjusted versions work, and why the raw figure often says more about teammates and luck than the player.

4Basketball Staff · Aug 18, 2026 · 11 min read

Home-court advantage in the numbers

Home teams win about three in five games, but the number hides a lot. This article converts the edge between wins and points, shows how noisy a single team's home record is, and works out what it is worth in a playoff series.

4Basketball Staff · Sep 11, 2026 · 10 min read

Net rating: the best single number for a team

Net rating is the gap between the points a team scores and allows per 100 possessions. Here is how it works, why it predicts so well, and where it can mislead you.

4Basketball Staff · Sep 12, 2026 · 11 min read