The Blank Cell on the Scorecard: The "Insufficient Data" Standard Vietnamese Golf Still Lacks

Seven twelve on a Tuesday evening. On the table: a spreadsheet with 42 rows and eleven columns. Three columns are completely blank — average driving distance, greens in regulation, strokes gained putting. The coach needs me to pick two names out of five for a regional qualifier. The deadline is eight the next morning.

I look at the three blank columns and I know exactly what is about to happen. Whoever talks loudest in the room will make the call.

Those blank columns did not come from a technical failure. They came from something simpler: in the domestic tour system, nobody records those things. No device, no staff, no process. When a data cell is empty, whatever slides in to occupy it is almost always a story — the best story, told by the most confident person in the room.

Data does not lie. But reputation whispers into the ear of anyone who will not read the table.

Three layers of measurement, and the layer we actually stand on

Golf is measured more densely than most individual sports, and misunderstood more than most of them too, simply because its three data layers get merged into one during conversation.

The first layer is the scorecard. Eighteen boxes, filled in pencil, dating back to the 19th century. This layer answers exactly one question: who took fewer strokes today.

The second layer is the raw statistics — fairways hit, greens in regulation, putts per round. They emerged in the second half of the 20th century and carry a fatal weakness: they measure circumstance, not skill. An approach from 60 yards and an approach from 210 yards are counted identically.

The third layer is Strokes Gained. Mark Broadie, a professor at Columbia Business School, systematised the method in "Every Shot Counts", published in 2026; the PGA Tour put Strokes Gained on its official statistics page in 2026. The principle is compact: every shot has an expected value based on distance to the hole and ball position; the difference between the expectation before and after the shot is the number of strokes that shot created or destroyed. Strokes Gained splits into four categories: off the tee, approach, around the green, putting.

The third layer requires infrastructure. The PGA Tour has run ShotLink since the early 2000s: a system that records every shot to the yard, combining measurement devices with volunteer crews stationed at every hole. The DP World Tour operates its own infrastructure with different coverage. Most regional tours in Asia stop at layer one, occasionally touching layer two with a pencil and an untrained recorder.

The Official World Golf Ranking, launched in 2026, sits above all three. It awards points by finishing position, multiplied by field strength, accumulated over a rolling 24-month window with a minimum divisor of 40 events. That structure makes the world ranking a relative measure, not an absolute league table.

The Vietnam Golf Association was founded in 2026. The country currently has fewer than 100 courses in operation. The domestic competitive system consists of one main tour, a handful of opens and a junior circuit. Layer one is complete. Layer two is inconsistent — some events record fairways, some do not, some record putts by asking the recorder to count them himself. Layer three is essentially empty.

Put another way: we are making decisions about people using data from one and a half layers.

A hot putting week is not a skill

Across 18 holes, a golfer takes roughly 28 to 32 putts. That is a small sample, and small samples in putting are worse than small samples anywhere else. Strokes Gained: Putting from a single round varies so widely that it carries almost no predictive value. Public research on the stability of Strokes Gained reaches the same conclusion everywhere: putting requires more rounds than any other category to stabilise, typically quoted in the dozens, while driving stabilises far faster, in roughly 15 to 20 rounds.

The practical consequence is concrete: a 23-putt round is an event, not a capability. It says the ball went in the hole that day. It does not say the player putts well.

I have watched this loop repeat many times. A good round appears on the results sheet at nine in the evening. By the next morning, that name is in three conversations. By the end of the week, it is on a selection list. Nobody comes back to check after ten more rounds, because checking requires ten more rounds to check against — and we are not recording them.

Based on my experience following rounds at junior and semi-professional events over the past seven years, I can put it plainly: most "new names" in Vietnamese golf are created by one round, and re-evaluated by another round. Between those two moments, no data layer is present to arbitrate.

A course record is not a unit of measurement

A 64 on a 6,200-yard course and a 64 on a 7,400-yard course are different things — different in length, rough height, green speed and prevailing wind. The USGA Course Rating and Slope system exists precisely because of that gap: two numbers describing course difficulty for a scratch golfer and for an average bogey golfer.

In Vietnam, course difficulty rating has not been applied consistently across all operating courses. As a result, any comparison of "course records" between two venues is comparing two different things under one name. Add tee boxes to that. A round from the white tees and a round from the black tees can differ by four or five expected strokes, while the results sheet still prints a single line.

When I built a comparison table for a junior selection group, I had to add a difficulty adjustment column that I estimated myself from length, green area and green speed measured with a device. That column exists in no official results sheet. It is the product of sitting down and asking: if there is no standard yardstick, do I have the right to compare? The answer is yes, but only if I state clearly what I am comparing.

A title is not a unit of measurement

In the world ranking system, an event's points depend on field strength, and a player's points depend on how many events he has played in 24 months against a divisor floor of 40. A win in a weak field can be worth fewer points than a top-10 in a major. A player who contests 12 events over two years has his total divided by 40, a structural disadvantage against someone who plays 25.

Whether that structure is right or wrong matters less than the fact that it gets read wrongly. In Vietnamese, both cases are called "vô địch" — champion. In the spreadsheet, they are quantities several multiples apart.

I wrote about Germany's collapse before the 2026 World Cup. Not because I was clever, only because I did not believe in the legend. That lesson transfers to golf almost intact: a title is an event with a context, not a permanent certificate of ability.

This leads to an uncomfortable consequence for domestic rankings. When someone asks me for "the current top 10 golfers in Vietnam", the honest answer is: I can rank by tournament results, but I must state clearly that the list blends fields of different strength, courses of different difficulty, tee boxes of different length and seasons of different conditions. After blending, what I have is a composite index — useful for reference, useless for conclusions.

The flashy drive and the silent approach

This is where Vietnamese golf data is most often misread, and also where it would be easiest to fix if layer three existed.

Take a hypothetical example to show the mechanism, not to describe any individual. Golfer A drives the ball long and beautifully and gets filmed a lot. He gains +0.35 strokes off the tee. But his approach play sits at −0.55, because the shorter distances left after those drives land him in the 90-to-130-yard range, where his distance control is weak. Putting adds +0.10. Net: −0.10 strokes per round against the baseline.

The camera shows the drive. The scorecard shows the loss. The viewer remembers the drive. The coach reads the scorecard and wonders why a player who looks that good does not win.

Data does not lie. But it only speaks when someone opens the table and reads the right column.

In a system with only layer one and half of layer two, a 300-yard drive becomes a golfer's entire capability file. The 130-yard approach is judged by feel. And feel, in sport, is easily shaped by the memory of the best shot rather than by the distribution of all shots.

Potential is paid for in advance, the bill arrives later

There is another blind spot in the youth development layer, and it has nothing to do with measuring devices.

Talent evaluation models pay for the slope of the development curve, not for its peak. A 15-year-old improving his scoring average quickly will be rated above a 17-year-old already stable at a good level, even if the older player has the more durable technical base. Slope is visible in data. The peak is not yet.

At many academies, this produces a compressed schedule: a 15-year-old may play more than 20 events in a year, on top of training. A body that has not matured gets pushed into an adult competitive rhythm, and what usually gets traded away is the quality of the technical foundation rather than short-term results. Short-term results may even improve, because playing a lot makes pressure familiar.

I have argued against that scheduling, and I argued with a simple comparison table: rounds played per year against injury rate and technical regression rate in the 15-to-18 age group. Small sample, I know. But even with a small sample, writing "we do not have enough data to conclude causation, but this correlation deserves a pause" is better than continuing to schedule because "that is what we did last year".

The easiest place to be wrong: correlation is not causation

There is an argument I hear constantly: players who win a lot are players who hit a lot of fairways. It sounds reasonable. When you split the data, the relationship is not linear at all.

Hitting a driver into the fairway and hitting an iron into the fairway are two different behaviours. A golfer who chooses an iron on a 380-yard par-4 will have a very high fairway percentage, but he enters the green with a longer approach and loses strokes in the approach category. Fairway percentage rises, score does not fall. Read only the fairway column and you will reach the wrong conclusion and give the wrong advice.

This is why I never convict or crown a golfer on a single metric. Any single yardstick can always be optimised by sacrificing something else.

And this is where I want to raise the most neglected standard in the entire analytical trade: when a data dimension is missing, the professional handling is to mark it "insufficient information, cannot assess", not to guess. Guessing feels useful because it produces a number to argue about. But the cost of guessing is not the wrong decision — the cost is that after the wrong decision, nobody learns anything, because no trace remains of where the guess was inserted.

I hate uncertainty. But I have learned that a blank cell correctly labelled is worth more than a cell filled in with the intuition of the most confident person in the room.

Signals for the next round

Those three blank columns on Tuesday night, I left them blank. I sent the sheet back to the coach with one note: the first two columns can be estimated from tournament results, the third cannot, and the selection should rest on the two columns that have data plus a live assessment session. The next morning he chose that way. We may still have chosen wrong. But next time, we will know where we went wrong.

I do not predict. I read the data and accept the consequences.

The signal I am watching next season is not who wins which title. The signal is: which domestic event will be the first to publish a results sheet carrying a line about what it could not measure. It sounds small. But a system willing to say "we do not know here" is the only kind of system capable of improving.

If next season a VGA Tour results sheet carried the line "insufficient data to assess", would you trust that event more, or less?