27 Straight Missed Threes: When Moreyball Betrayed Itself

On May 28, 2026, at Toyota Center, the Houston Rockets led the Golden State Warriors by 11 points entering the fourth quarter of Game 7 of the Western Conference Finals. Mike D'Antoni's team had won 65 regular-season games, the best record in the league, led the NBA in three-pointers made, and had James Harden, who would be named MVP just weeks later. Over the final 20 minutes of their season, they missed 27 consecutive three-pointers. Not one shot touched the net from beyond the arc.

The Warriors won 101-92 and rolled to the title. Houston, a roster assembled by spreadsheet, collapsed on the very weapon the spreadsheet worshipped most. Chris Paul, the second playmaker, sat out with a hamstring injury — a detail many forget, yet the one piece that explains the whole story.

I rewatched that tape four times in a single week. Each time, the 7-of-44 line from three rose up like an indictment. Modern basketball had built a cathedral named “efficiency,” and that Game 7 was the day its doors slammed shut in front of everyone.

That game taught me something still true years later: basketball does not reward the team that shoots the most; it rewards the team that controls its own variance.

To understand why those 27 misses were no mere accident, you have to go back to the philosophy that ruled the NBA through the 2010s. Daryl Morey, then Houston's general manager, offered a formula simple to the point of cruelty: scrap the mid-range entirely, shoot only at the rim or beyond the arc. By points per shot, it is the optimal choice. A three is worth 1.5 times a two if the efficiency is equal.

In 2026-18, Houston launched 3,470 threes, a record at the time. They won 65 games. The league rushed to copy the formula. Pace rose, three-point volume rose, and faith in the data model peaked.

But the model had a hole no spreadsheet could draw: the playoffs do not play by regular-season rules.

When I went back through the champions from 2026 onward, a pattern emerged more clearly than any three-point chart. Nearly every champion ranked near the top in free-throw rate and defensive rating, not in three-point volume. Milwaukee won in 2026 with Giannis Antetokounmpo relentlessly attacking the rim. Denver won in 2026 around Nikola Jokić and an interior game. The teams most dependent on threes — Houston 2026, Utah from 2026 to 2026, Atlanta 2026 — all fell exactly when the pressure spiked.

The most telling number: in the final five minutes of playoff games, league-wide three-point efficiency drops roughly 3 to 5 percentage points versus the regular season. Defenses tighten, space vanishes, and open shots become contested ones. The three lives on space. The playoffs take space away.

From my experience watching games, the moment a team starts shooting threes out of frustration rather than opportunity is the moment the game has slipped beyond its control.

The data also reveals a paradox of variance. A team that takes 40 threes a game is not certain to beat a team that takes 30. But it is certain to swing wider. In a seven-game series, that swing is what kills you. Houston optimized for the average, but a title is not awarded to the average. It is awarded to whoever wins four games on four specific nights.

Numbers do not score, but numbers are quietly rewriting history. And the history it wrote for Houston 2026 was a marginal note, not a line in the record book. The “number saint” label I have carried through my career never came from worshipping numbers, but from knowing when numbers lie.

Now comes the part where I have to challenge myself. The “threes do not win titles” argument has an obvious hole.

The Golden State Warriors won in 2026, 2026, 2026 and 2026 playing a brand of basketball built on ball movement and perimeter shooting — they are the best three-point shooting team in history. The Boston Celtics won in 2026 on massive three-point volume. If threes cannot win, how do you explain those two teams?

The answer lies elsewhere: shot quality, not volume. The Warriors and Celtics never lived and died beyond the arc. They still attacked the rim, still had a fallback when the shot would not fall. Houston 2026 had no Plan B. When 27 shots missed, they had no other way to score. The problem was a single-source dependency, not the three-pointer itself. And I admit I conflated the two for years.

Ball control is an illusion; the score is the naked truth. But the score, in turn, is only an illusion if it comes from a single source that can be choked off.

So my prediction for the rest of this season is verifiable: the league leader in three-point attempts will not win the title. Watch which team tops the 3PA column, and brace yourself to see them exit before the Finals. Basketball does not hand the trophy to an algorithm. It hands it to the team that knows how to score another way when the algorithm fails.