VBA 2026 Pace Rose 6.8 Percent While Shooting Efficiency Fell: Re-reading 92 Games Through a Data Log

It was 11:47 p.m., third quarter, Cantho Catfish on the road in the eastern part of the country. Over 2 minutes and 38 seconds, the Catfish ran seven straight possessions and all seven ended in a three-point attempt. Two went in. Five missed. The coach called timeout, clapped, and the scoreboard reset to zero. In column 41 of my spreadsheet I typed a short line: seven possessions, 0.86 points per possession, 14 percent below their own half-court average.

I did not save that moment because it was beautiful. I saved it because the same structure repeated 63 times during the 2026 season, across seven teams and twelve head coaches. When something repeats 63 times, it stops being inspiration. It becomes a pattern.

By the end of the season I had 92 manually coded games, 11,408 possessions, 4,217 three-point attempts classified across seven variables, and one very simple question: if the league is running faster, why is efficiency falling?

Numbers do not lie, but they do not tell stories either. The rest of this piece is an attempt to tell that story without inventing a single thing.

What I measure, and how I measure it

My spreadsheet started in 2026, when I was a third-year student in Da Nang and stayed behind after every game to count possessions by hand. Back then I had one column: shot attempts by a foreign forward. Six years later the spreadsheet has become a small system, but the principle has not changed: raw sources must be public, the method must be repeatable, and no number may come from memory.

In 2026 I tracked all 92 games, regular season and playoffs, logging every possession along four axes: when it ended, how it ended, the distance of the nearest defender when the shooter received the ball, and the number of passes that led to the shot. Every game was reviewed twice — once at real speed to catch rhythm, once at 0.5x to confirm defensive distance. Ambiguous situations were discarded, not guessed. In total I discarded 611 possessions, about 5.4 percent of the sample. I would rather lose sample than lose consistency.

Four indicators anchor my read of this season: pace (possessions per 40 minutes), eFG% (shooting efficiency weighted for threes), TOV% (turnovers per possession), and ORB% (offensive rebounding rate). I do not use scoring average as an input, because scoring average is an outcome, not a cause. Reading it is like reading a final transcript and inferring how someone studied.

Among players, I separate local-born Vietnamese players from overseas Vietnamese players, known locally as Viet kieu. The reason is not identity but regulation: the league's registration rule creates two entirely different labour markets. If you do not separate them, every analysis of roster construction will be wrong.

Data is a monastery: the less noise, the more clearly you hear something trying to speak.

Layer one: pace and its trap

League-average pace in 2026 was 79.4 possessions per 40 minutes. In 2026 it was 74.3. A 6.8 percent jump in a single year is the largest I have recorded since I began tracking this league. For comparison, between 2026 and 2026 league pace moved inside a band of just 1.9 possessions.

The first thing people assume when pace rises is that scoring will rise. Average scoring in 2026 was 79.5 points per game, up from 78.9. A gap of 0.6 points, or 0.76 percent. A league running 6.8 percent faster scored only 0.76 percent more. Those two numbers do not fit together, and that gap is the subject of this article.

I split possessions into three groups by end time: under 8 seconds (early offence), 8 to 18 seconds (mid offence), and over 18 seconds (late offence). In 2026, early offence accounted for 21.3 percent of possessions and produced 1.02 points per possession. In 2026, early offence rose to 28.7 percent but produced only 0.91 points per possession.

Read the other direction: late offence produced 1.08 points per possession in 2026 and 1.06 in 2026. Essentially unchanged.

In other words, the entire increase in pace flowed into the least efficient category. No team ran faster because it found a better way to score. Teams ran faster because they decided earlier, and deciding earlier is not the same as deciding better.

From watching games courtside, I can point to a specific cause: the 24-second clock in this league is operated under far more psychological pressure than in leagues with deeper benches. When a team has only seven or eight players capable of carrying real rotation minutes, coaches tend to take the first acceptable shot to save energy rather than wait for the best one. Rising pace here is a symptom of fragility, not of progress.

Layer two: the three-pointer and fake space

League-wide three-point rate (3PA/FGA) rose from 33.8 percent in 2026 to 39.6 percent in 2026, the highest I have recorded in this league's history. One team reached 44.1 percent, meaning nearly half its shots came from beyond the arc.

But league eFG% fell, from 49.1 percent to 48.3 percent. Three-point rate up 5.8 percentage points, shooting efficiency down 0.8. The two lines crossed in the opposite direction from conventional logic, which is when I go back to the tape.

Classifying 4,217 three-point attempts by nearest-defender distance sharpened the picture considerably. The share of threes considered open — nearest defender more than 1.5 metres away — fell from 41.2 percent to 36.7 percent. The share of contested threes rose from 58.8 percent to 63.3 percent. Plainly: teams took more threes, but under harder conditions.

This is the point most coverage skips. A broadcast will say Team X took 18 threes, which sounds modern. But if 13 of those 18 were contested attempts early in the clock, that is bad basketball wrapped in modern language.

I checked the correlation between 3PA/FGA and final league position across the seven teams. The coefficient came out at minus 0.11 — effectively zero. Shooting more threes neither wins nor loses you games. It is a tactical choice, and that choice only has value when it comes from shot quality, not shot quantity.

By contrast, the correlation between TOV% and final position was minus 0.72, very strong. Teams that turned the ball over less finished higher. The correlation between ORB% and final position was plus 0.58. Teams that rebounded their own misses better finished higher. Two old indicators, dismissed as mundane, predicted standings far better than the flashiest metric of the decade.

I have a professional habit: for every variable I ask whether it changes how I would predict. If not, it is only pretty noise. On this season's data, 3PA/FGA is pretty noise.

Layer three: the Viet kieu quota and a distorted roster market

This is the layer I consider most important, and the least discussed.

League registration rules allow teams to use players of Vietnamese descent, commonly called Viet kieu, through a mechanism that does not count against the import quota. In theory this is sensible policy: it opens a path for the children of overseas Vietnamese to come back and contribute while raising the league's quality. In practice it creates a two-tier labour market with radically different price levels, and that gap shapes roster building in ways no box score ever shows.

When I separated the share of points contributed by Viet kieu players across the four semifinalists, the figures ranged from 22.4 percent to 38.4 percent. The champion sat at the top of that range. The seventh-placed team sat at the bottom. Across all seven teams, the correlation between Viet kieu scoring share and final position was plus 0.64.

This is where I have to be careful, because correlation is not causation. There are at least three explanations for the same number, and they lead to three entirely different governance conclusions.

Explanation one: Viet kieu players are better physically and technically, so teams that use more of them are stronger. Conclusion: go find more Viet kieu players.

Explanation two: stronger teams have bigger budgets, so they attract the best Viet kieu talent; roster quality is the cause and the quota is merely the effect. Conclusion: do not mistake a symptom for a cause.

Explanation three: the Viet kieu quota lets a team save money at the import slot, and the savings get spent on bench depth. Conclusion: real value lies in budget allocation, not in a passport.

I lean towards the third explanation, but I do not have enough salary data to prove it. My spreadsheet has a column marked unverified, and I leave it that way. Inventing a salary figure would be far easier than admitting I lack data, but easy is not what I need.

What I can state with certainty: roster construction in this league is shaped more by registration rules than by tactical philosophy. When you hear a coach talk about building a modern playing style, ask how many Viet kieu slots he holds. The answer usually explains that style more precisely than any coaching manual.

Every coach talks about feel. I do not have feel. I have standard deviation.

Layer four: defensive friction and the price of substitutions

If pace rose and efficiency fell, the defence has to absorb the difference. The question is where.

I built a private indicator called the Friction Index, measuring the average number of opponent passes before a defence forces a state change — a turnover, a contested shot, or a shot taken with under four seconds on the clock. The best team recorded 2.41 passes per state change. The worst recorded 3.86. A gap of 1.45 passes sounds small, but it multiplies into roughly 7.3 possessions per game, and at 1.02 points per possession that is about 7.4 points per game between the best and worst defensive teams.

The more interesting finding sits elsewhere. Splitting the Friction Index by half, the best team recorded 2.38 in the first half and 2.52 in the fourth quarter. The worst recorded 3.29 in the first half and 4.71 in the fourth. In other words, the defensive gap between the two teams nearly doubled in the final ten minutes.

This is where bench depth, not tactics, decides outcomes. In the last ten minutes a defensive system does not collapse because it is wrong. It collapses because the legs executing it are gone.

I measured rest days between games for each team. The league average was 2.8 days, down from 3.4 in 2026 — a 17.6 percent increase in density. I also measured average minutes for the top five players by minutes on each roster: 34.7 minutes per game, up 3.2 minutes from last season.

Combine the two: teams playing a 17.6 percent denser schedule with their core players on the floor 3.2 minutes longer per game. No defensive system survives that addition without breaking in the fourth quarter.

Monthly three-point percentage matches the story exactly. Opening month: 37.1 percent. Second month: 35.8 percent. Third month: 33.2 percent. Playoffs: 31.6 percent. The curve declines steadily and does not distinguish strong teams from weak ones.

Here I have to say plainly something I have held for years: load management in Vietnamese basketball is romanticised. It is called science, athlete care, long-term vision. But look at the schedule and you see something else. The calendar thickens because there are more games, more sponsors, more ticketed matchdays. A player is rested in one game so he can play the next one with the bigger crowd. That is not load management. That is risk allocation by revenue.

I do not write this to attack anyone. I write it because when I open the log, I see teams paying the price in the fourth quarter, and that price is recorded in points, not in press releases.

The counter-intuitive angle: the league is not attacking more, it is defending less

This is the conclusion I reached after reading all 11,408 possessions, and it runs against how most viewers describe this season.

Pace rose 6.8 percent. The common interpretation is that teams are playing more offensively, more modernly, faster. My data says otherwise. Possessions rose, but points per possession fell. If teams were genuinely attacking better, both numbers would rise together. When only one rises, it signals lost control of the ball, not elevated attacking quality.

I tested this hypothesis three independent ways.

First, half-court pass completion. 2026: 61.3 percent. 2026: 58.7 percent. Less accurate passing means the ball moves less coherently, which means shots come from worse situations.

Second, assist rate on made field goals. 2026: 52.4 percent. 2026: 49.1 percent. More shots, but fewer of them created by teammates. That is the technical definition of individualised basketball.

Third, average time for a possession to cross half-court. 2026: 4.6 seconds. 2026: 3.9 seconds. The ball crosses half-court 0.7 seconds faster, yet the number of passes before the first shot fell. Teams push the ball up faster only to do less with it in the opponent's half.

Stitched together, the picture is not of a more attacking league. It is of a less defending league. Defences retreat later, rotate slower, contact less. Pace rose because defence loosened, not because offence sharpened.

And here is the most interesting detail, and the hardest to explain. The 2026 champion played at 75.1 possessions per 40 minutes — the second-slowest mark in the seven-team league. The team with the highest pace finished sixth.

I spent two weeks looking for an alternative explanation, because I did not want to believe the simple conclusion that slow wins. I checked whether the champion faced an easier schedule. It did not. I checked whether they shot threes better. Not especially — third in eFG%. I checked whether they defended better. They did, emphatically: best Friction Index in the league, second-lowest TOV%.

After two weeks, I accepted the conclusion the data offered, even though it is far less attractive than a story about a pace revolution. The champion won because it protected the ball and protected the rim, in a league where twelve other coaches were racing to run faster.

People watch the decisive shot to remember a game. I watch the possessions that end inside eight seconds to understand how the game failed to happen.

Two diseases of reading Vietnamese basketball data

Before the final section, I want to name two traps I nearly fell into, and that I see many people in this field falling into.

The first trap is choosing a rare number to manufacture drama. My spreadsheet holds some beautiful lines: a player going 5-of-6 from three in one game, a team scoring 14 points in 90 seconds, a quarter with 11 lead changes. Those lines give me the feeling of being clever when I write about them. But they are small samples, and small samples predict nothing. Before using a number, I ask whether it changes how I would bet on the next game. If not, it stays in the file.

The second trap is talking about Vietnamese basketball as a flat map. I have worked in the country for six years, I know the seven team names, I know which arena has the loudest crowd. That feeling makes it easy to believe you understand everything. But there is a boundary I force myself to draw every time I write: which part is a number I measured, and which part is inference from experience. A pace of 79.4 is a measurement. A coach choosing early shots because his bench is thin is an inference. I present those two things differently, in two different kinds of language, and I mark clearly which is which.

In 2026 the world mourned Germany after their World Cup group stage exit. I quietly re-read the log file of my model, where pressure and distance-run indicators had signalled the outcome three weeks earlier. The lesson I carried into basketball was not that data is always right. It was that when data and public emotion disagree, I am responsible for recording both and letting time adjudicate.

The strongest lineup is never the five prettiest names. It is five equations in harmony. In this league, I am still looking for the third equation.

Signals for the next cycle

If next season repeats this season's structure, here are four things I will track before they become headlines.

Schedule density. If average rest between games falls below 2.8 days, I expect playoff three-point percentage to drop below 30 percent. That is the threshold where I believe a shooter can no longer recover between games.

The Viet kieu market. If the league tightens or relaxes the quota mechanism, I will re-measure the correlation within three months. A rule change here has more predictive power than any import signing.

TOV%. I will track turnover rate in the first 15 possessions of every game. If a team keeps turnover rate under 12 percent in that window, I treat them as a semifinal candidate regardless of how many threes they take.

Pace. If league-average pace passes 82 possessions per 40 minutes while average scoring stays below 80, I will read that as a league eroding its own quality rather than evolving.

Models never tremble. They only record, and wait for a reader.

When a young coach tells me he trusts the feel of a game more than any spreadsheet, I smile. I touch the future with a keyboard. But I also write down what he said, in column 42, next to the date, because I want to know whether that feel still holds twenty games from now.

Data does not play basketball. But it decides who plays. And in a league whose schedule thickened 17.6 percent in a single year, the final bill is always paid by the players' legs — the one component in this entire system that has never been priced correctly.