73 Points a Game: Is Panathinaikos the EuroLeague's Best Defense, or Its Slowest Team?

4 a.m. in Tokyo. The apartment is quiet enough that I can hear the fridge. I rewatch the Panathinaikos tape, frame by frame on the defensive end, because one line of data is being quoted across Europe: after six EuroLeague rounds, the Athens club has conceded 73.0 points per game, the lowest in the competition. Fenerbahçe sit at 74.5. Red Star at 76.2. The label “best defense” was attached almost immediately.

I am not in a hurry to believe it. At 16 I built a hand-made spreadsheet tracking Rui Hachimura at Japan's U18 championship, and the first lesson from that file still holds: a metric is only worth something when you know exactly what it measures. Points conceded per game measures pace. It does not measure defensive quality. Data does not lie, but the people reading it do.

What the table is actually comparing

Six games, 438 points. Fenerbahçe 447, Red Star 457 — gaps of 9 and 19 points across six games, or 1.5 and 3.2 per game. At the same stage last season, Panathinaikos had conceded 524, or 87.3 per game. In that stretch they held an opponent under 80 exactly once (79 against Bayern) and gave up 103 to Barcelona at OAKA.

The roster names worth noting: Mathias Lessort, Kendrick Nunn, Kostas Sloukas.

One detail has to be verified before any analysis. The source dataset attaches this team to Željko Obradović, while in recent seasons the bench in Athens has belonged to Ergin Ataman. I have not been able to confirm which is correct, and I will not pretend otherwise. If this is a coaching-change season, the entire tactical read has to be rewritten. If it is a source error, the comparison still stands but the story changes. The source also labels the seasons “2026-2026 and 2026-2027”, which cannot be consecutive — an editing error somewhere.

Pace: the 4.5 to 5.5 point leak

Points conceded per game equals defensive rating divided by 100, multiplied by possessions per game. A team that plays six fewer possessions concedes roughly 4.5 to 5.5 fewer points without defending a single possession better. That is multiplication, not magic.

The source provides no defensive rating per 100 possessions, no pace figure, no list of the six opponents, no home/away split. Those four gaps are enough to void the “best in the league” label before anyone opens the film.

Six games is not a ranking

With a per-game standard deviation of 10 to 13 points — typical variance in European basketball — six games produce a standard error of roughly 4.5 to 5.3 points. The 95% confidence interval around 73.0 runs from 64 to 82. Fenerbahçe's 74.5 sits comfortably inside it. A 1.5-point gap gives a t of roughly 0.24. There is no statistical separation at all.

Put differently, the table is ranking noise. One 10-point blowout, or one bad three-point shooting night from an opponent, flips the order. I have seen this repeatedly in Japan's B.League: after six rounds, the leader in points allowed is usually not the best defensive team but the slowest team with the softest schedule.

The comparison that survives

The claim that holds up is the year-over-year one: 87.3 down to 73.0, a drop of 14.3 points per game. A t of roughly 2.26, about 10 degrees of freedom, a p of roughly 0.05. Borderline. Still a real signal rather than noise.

But this comparison does not control for opponents. Different Round 1-6 schedules, different home/away splits, different opponent strength. If this is a coaching-change season, the comparison also conflates a system change with a form change. My confidence here is medium, not high.

The mechanism: FIBA rules and a wall named Lessort

FIBA basketball has no defensive three-second rule. A defender may stand in the paint indefinitely, which legally enables a deep drop coverage that the NBA structurally forbids. With a rim protector of Lessort's caliber, a club can run a scheme NBA coaches can only dream about. This is the most plausible mechanical source of the improvement.

The other lever is tempo suppression. Cutting opponent possessions from about 72 to 67, at constant defensive rating, saves 4 to 5 points per game. Roughly one third of the 14.3-point drop comes from style, not from defensive skill. The remaining 8 to 10 points is likely real improvement.

The system has a single point of failure: Lessort's body. Drop coverage, rim protection, defensive rebounding — every mechanism routes through one frame. FIBA disqualifies at five fouls, not six, in a 40-minute game rather than 48. A high-motor, contact-seeking big who picks up two early fouls inside ten minutes collapses the whole structure.

The weakness is in the backcourt, not the frontcourt

Nunn and Sloukas are the two offensive engines, and both are targets on defense. Sloukas is on the far side of the age curve, where lateral mobility degrades first — a pattern familiar enough that it needs no argument. Nunn is returning from injury, and “his return came earlier and he feels good” is a player's self-report, the least reliable category of medical signal there is.

The source contains zero individual statistics: no points, rebounds or assists, no true shooting, no usage rate, no plus-minus. Everything above is role-archetype inference, and I am downgrading confidence to medium for exactly that reason.

The contrarian read: the label is meaningless, the trend is real

One detail the ranking hides. Last season their best defensive game at this stage was 79 points. This season the average is 73. The old floor sits above the new mean. The entire distribution has shifted down, not just the central value. That is structural change, not a hot streak.

And the 103-point Barcelona loss at OAKA is the blueprint. Any team that can force tempo and generate early-clock threes breaks this scheme. Between Rounds 15 and 25, once staffs have the film, they will copy that blueprint. Defense travels to the playoffs. A table position does not.

I have made the opposite mistake before. In 2026 I predicted Japan would reach the Olympic quarterfinals in Tokyo purely because Hachimura and Watanabe were NBA players, and I ignored their 118.4 defensive rating. They lost all three group games, including 77-97 to Argentina. I wrote a 1,500-word self-audit. Since then I do not rank a team on player reputation. A giant's failure is a gift to the observer.

What to watch

Three things will decide how long the “best defense” label survives. Lessort's foul count in the first ten minutes of each game. The number of possessions where Nunn is dragged into a direct defensive assignment. And the number of three-point attempts opponents generate inside the first eight seconds of a possession.

Japan taught me that the treasure is always there, you just need the patience to dig. Here, the treasure is the pace and defensive-rating data nobody bothers to publish. Empires are not built overnight, but data can build them in a season. A “best in the league” label can be erased in one game.