Vietnamese Basketball: 96 VBA Games and the Paradox of Imports Holding All the Ball

Fourth quarter. A five-point game. Six minutes and forty-two seconds left on the clock. During that eighteen-second possession, the ball passed through four pairs of hands, but only two players touched it more than once, and both were imports. Two Vietnamese players stood in opposite corners, barely moving, never cutting, never receiving. I marked that possession in red in my tracking sheet, the way I have marked every possession across 96 VBA games I logged myself over the past two seasons.

When the spreadsheet finished running, the summary line made me reopen the video to double-check. Across 412 fourth-quarter possessions in games decided by fewer than eight points, only 14.8 percent ended with a shot by a domestic player. The rest belonged to imports, or to turnovers where nobody got a shot away at all. That is a count, not an opinion.

What bothers me is not the number. It is how the number gets explained. In the stands and on social media, the familiar story goes: imports are better, so they hold the ball more, and that is only logical. I spent two seasons testing that claim with possession data, and the results do not support it.

Context: a league too short to build a system

The VBA launched in 2026 and still runs a short-season model: a group stage lasting a few months, roughly twenty games per team, then the playoffs. That gives a club less than three months to assemble a roster, install a system and build chemistry. In most professional leagues, that window barely covers the preseason.

The roster structure is specific too. Each team registers a limited number of imports plus a group of overseas Vietnamese players, while the rulebook still requires a Vietnamese player on the floor for most of the game. The design protects domestic players. But it protects minutes, not possessions. A player can spend thirty minutes on the floor and touch the ball on three offensive trips. The rulebook cannot measure that, and neither can the official box score.

My method is unremarkable. Every possession is logged by the standard definition: an offensive sequence ending in a shot, a turnover or a foul that leads to free throws. I classify the finisher, the shot type, the game situation and the score margin. In total, 96 games and more than 9,400 possessions. For each team I calculate an index I call IDI, the share of possessions ended by imports.

I am not using this data to attack imports. They are doing exactly what they are paid to do. The question I want answered is whether handing the ball to two players actually produces more points.

Four layers of evidence

The first layer is pace. In my tracking sheet, league pace fell from 78.4 possessions per 48 minutes to 74.1. The share of possessions ending inside the first eight seconds dropped from 21 percent to 15 percent. Games became slower, fast breaks rarer, and most of the clock now runs with the ball in one player's hands on the wrong side of half court. Slower pace is not inherently bad. It becomes bad when it comes with fewer passes, and in my data average passes per possession fell from 2.7 to 2.3.

The second layer is who finishes. Last season, imports accounted for 54.3 percent of all possessions league-wide despite playing roughly 30 percent of available minutes. In fourth quarters of close games, that share jumped to 71.2 percent. Put simply, the later the game, the further the ball drifts from Vietnamese hands. There were games where, across the final eight minutes, the entire domestic group on the floor touched the ball fewer times than a single import.

The third layer is shot quality for domestic players. Of all possessions ended by a local player, only 9.2 percent were three-point attempts off a pass. Thirty-one percent were mid-range jumpers. Thirty-four percent came at the rim, but most of those followed offensive rebounds rather than organised actions. If you are a Vietnamese player in this league, chances are you touch the ball in the worst spot on the floor, in the smallest sliver of the 24-second clock.

That layer also reveals something about space. With two imports on the floor, the average distance between the two domestic players on the weak side was 4.8 metres. With only one import, it dropped to 4.1 metres. Defences read this quickly: they double the import, ignore the other two threats, and turn every possession into a one-on-two duel. Breaking that coverage requires a domestic player who can shoot from beyond the arc, and the data says they are rarely given the chance.

The fourth layer is correlation. I ranked the seven teams by IDI and compared it with results. To avoid inviting readers to pin labels on individual players, I refer to the teams as A through G.

| Team | IDI | Fourth-quarter point differential | Win rate | |---|---|---|---| | Team A | 61% | minus 2.8 | 41% | | Team B | 58% | minus 1.4 | 48% | | Team C | 55% | plus 0.6 | 52% | | Team D | 52% | plus 1.1 | 55% | | Team E | 49% | plus 2.3 | 61% | | Team F | 46% | plus 3.0 | 64% | | Team G | 43% | plus 2.6 | 62% |

The correlation between IDI and win rate is minus 0.06, essentially zero. The correlation between IDI and fourth-quarter point differential is minus 0.41, a signal but not a strong enough one to settle the argument. All three teams above 55 percent IDI were in the group that entered the post-season early. The lowest IDI in the sample belongs to the team that topped the standings in both seasons I tracked. That team plays the least replicable basketball in the league, and it also owns the deepest pool of domestic players capable of creating their own shot.

One more detail matters more than the table. Across the 96 games, I counted 63 occasions where a team changed how it used its imports immediately after falling behind by eight points or more: more possessions for the primary import, fewer intermediary passes, and an almost complete abandonment of feeding domestic bigs inside. The average result of those 63 stretches was minus 1.6 points over the following six minutes. Falling behind is precisely when dependency becomes visible.

I also measured volatility. For each import, I calculated the standard deviation of points per possession across games. The average came out at 0.34, well above the 0.21 recorded for the leading domestic group. Imports generate more points on average, but they swing far more wildly. In a twenty-game season, high variance means three or four losses nobody can explain.

Overseas Vietnamese players deserve separate treatment because they sit between two worlds. On paper they count as locals; in style, many were trained abroad and shoot better than the domestic baseline. In my tracking sheet this group ends a markedly higher share of possessions than locally developed players, but still fewer than imports. The league therefore has a middle tier. If regulation narrows that tier, the gap will most likely be filled by imports rather than by Vietnamese players.

Figures such as Dinh Thanh Tam or Nguyen Huynh Phu Vinh are notable exceptions, part of a small group of domestic players whose usage is high enough that defences treat them as genuine threats. But exceptions confirm the rule: most Vietnamese players in this league occupy a role without the ball, and that role does not produce national-team players.

The contrarian angle

A negative correlation is not proof of causation. I need to say that clearly before anyone uses the table above to demand fewer import contracts. A high IDI does not make a team lose. Causation most likely runs the other way: a weaker team, one with less practice time, or one that lost a starter to injury, funnels the ball to imports as an emergency measure. Dependency is a symptom, not a cause. Reading the data while ignoring that is the fastest way to turn analysis into sloganeering.

The real blind spot lies elsewhere. A player holding 30 percent of possessions is not harmful in itself. Plenty of superstars in bigger leagues do exactly that. Problems appear when a heavy load comes without creating space for others: no extra defender drawn, no open passing lane, no stretched shell. In my data, only 22 percent of import possessions led to an open shot for a teammate. Among the leading teams, that figure was 34 percent. That twelve-point gap is larger than any talent difference I managed to measure.

My sample is also smaller than it looks. The 412 fourth-quarter possessions amount to roughly eight full games. The standard error on a proportion like 14.8 percent at that sample size is about 1.7 percentage points, and the real error is larger because one person logged the data. I have no automated camera tracking, and no positional data accurate to the metre. Everything here should be read as a signal, not a verdict. If league pace rises again in three months and the domestic open three-point rate passes 15 percent, my model is wrong, and I will be the first to say so.

Numbers do not lie, but they do not tell stories either. The story here is not about whether imports are good or bad. It is about a league designed too short to build a system, and carrying too much pressure to allow anyone the luxury of being wrong. When a coach has twenty games to keep a job, giving the ball to the best player is not a tactical choice. It is survival.

Signals for the next round

Three indicators will hold my attention for the rest of the season. First, the domestic open three-point rate. If it clears 15 percent, the offence is genuinely changing rather than enjoying a lucky stretch. Second, assists from domestic players to imports. That number sits unusually low right now, which suggests locals have been pushed out of creation, not just out of scoring. Third, fourth-quarter pace. If it keeps falling, teams are using caution as a strategy, and I can say in advance that caution does not win playoffs.

Every coach talks about feel. I do not have feel. I have standard deviation. But standard deviation is only a tool for asking better questions. The right question here is not whether to cut the number of imports. The right question is this: in the next 24-second possession, beyond the two players paid to score, how many Vietnamese players genuinely get a touch in a spot that forces the defence to guard them?

Data is a monastery: the less noise, the more clearly you hear something trying to speak. The loudest noise in this room is 14.8 percent.