Vietnam's Golf Data Gap: From the Scorecard to Strokes Gained

A 22-row report, and one surviving word

The file ran to eight sections, complete with subheadings and tables. Twenty-two rows, six columns. Every content cell empty. The only surviving token was the domain label: golf.

I read it once and assumed a rendering fault. I read it a second time and understood it was an honest document in the most uncomfortable way. No source title, no outlet, no one-sentence summary, no entity list, no time-sensitivity assessment. All eight analytical sections were still printed in full, each cell carrying the same phrase: insufficient information.

A scorecard with no numbers cannot be read. A scorecard with plenty of numbers but the wrong substance is worse. For one hundred and fifty years, golf has been recorded through a single data type: strokes. People know a player shot 68, made five birdies, hit eleven greens in regulation. Nobody knows whether those birdies came from a 320-yard drive down the fairway or from three long putts. The scorecard tells you how many, never how.

The empty report did the one thing the Vietnamese sports data industry rarely does: it stopped. It did not fill the blank with adjectives. It did not write "surging form" to cover the hole. That is why I am using it as the starting point here.

Context: Vietnamese golf is playing with year-2026 data

The PGA Tour has run ShotLink since 2026. Every shot, on every hole, in every round, for every player, is captured with coordinates, remaining distance, lie and ball state. One season generates hundreds of thousands of shot-level data points. The DP World Tour has its own system, the LPGA has its own, even feeder tours record at shot level. Without that infrastructure, Strokes Gained is a beautiful idea on paper.

Vietnam is different. The VGA Tour, national amateur events, open championships and pro-ams mostly still record at scorecard level. You get total strokes, birdies, bogeys, putts if a scorer writes them by hand, and occasionally GIR. You do not get driving distance by hole, ball position after the approach, or putt length. Without those three inputs, every Strokes Gained calculation is technically impossible — not for lack of tools, but for lack of raw material.

That gap is not small. Vietnamese golf is growing fast in courses, rounds played and inbound golf tourism. Coastal courses in central Vietnam, mountain layouts in Da Lat and Sapa, urban courses in the south keep appearing. Data infrastructure has barely moved since the previous decade. We hold more raw data than ever — every phone films a swing, every watch measures heart rate — and still cannot build a baseline deep enough to say whether a 165-yard approach into the green at Hoiana Shores was good or bad.

That is the central paradox. We measure more about a golfer's body than about the golfer's shots.

What Strokes Gained actually measures

The principle is simple, and it is why this concept changed how professional teams make decisions.

Every ball position on a course carries an expected number of strokes remaining, averaged across the tour. From the tee on a 460-yard par-4, the expectation is roughly 4.08 strokes to finish the hole. Hit a 300-yard drive into the fairway and the new position expects about 2.92. The value of the shot is 4.08 minus 2.92 minus 1, or plus 0.16 strokes. A three-foot putt expects 1.04; holed, it is worth plus 0.04.

The key property: the calculation adds and offsets. Total Strokes Gained across a field in a round is always zero. Nobody gets extra credit for reputation. Nobody is favoured for leading the world ranking.

Four categories are separated out: off the tee, approach, around the green, putting. In Vietnam, their order of importance differs from the United States. On central coastal courses, wind is a permanent variable, not an exception. A player with weak approach play in calm conditions can look better than he is; a player with strong approach play into heavy wind is often undervalued because the scorecard spares nobody.

The 68 example: a scorecard lying politely

Take a composite round from my model, built on scorecard data from a regional amateur event in southern Vietnam. The field averaged 71.2 that day. Player X shot 68, four under.

The card reads like this: five birdies, one bogey, eleven greens in regulation, twenty-seven putts. Any bulletin could write two lines of praise from that. Reallocated through the model, the picture flips:

Putting plus 4.6. Around the green plus 0.9. Off the tee minus 0.2. Approach minus 2.1. Total plus 3.2.

Read down the column and this 68 was built by the putter and eroded by the irons. Player X scored through the least repeatable of the four skill categories. That is the most important technical fact the scorecard hides: across seasons, approach is the most stable Strokes Gained category; putting swings the widest. A 68 built on putting forecasts worse than a 71 built on approach.

The next round, Player X shot 74. The clubs did not break. The swing did not change. The putter merely regressed to average, and the minus 2.1 in approach surfaced.

With only a scorecard, you conclude the player "lost form". With shot-level data, you know he never had that form to lose.

OWGR: a ranking that pays for field strength, not for fame

The world ranking operates as a market. Points depend on finishing position, field strength and the number of events counted across a two-year cycle. The direct consequence: winning a weak-field event can earn fewer points than a top-10 at a Signature Event. The system does not care who you are.

The biggest test case is LIV Golf. Since its 2026 launch, LIV has received no OWGR points. In October 2026 the OWGR board rejected its application, citing event format and qualification allocation. LIV withdrew the application afterwards. The consequences fell directly on players who joined: they lost the ranking pathway into majors and must live on past-champion exemptions. The Masters invites champions for life. The PGA Championship carries long-term exemptions for past winners. The U.S. Open holds a place for ten years. The Open admits past champions until age sixty.

Jon Rahm won the 2026 Masters and joined LIV that December. By the data, he remains one of the best approach players in the world. By the system, his route into majors is far narrower than a third-place world ranking once gave him.

The lesson transfers directly to Vietnamese golf at a smaller scale. When you build a points system for the VGA Tour or an open championship, you must decide from the start whether it rewards field quality or sponsor prestige. Those two choices produce two different systems, and only one survives three seasons.

Ball rollback and an unanswered question: who loses more

In March 2026, the USGA and R&A announced new ball regulations requiring balls to be tested under revised conditions. The rule applies to elite play from January 2028 and to recreational golf from 2030.

Public debate centres on whether classic courses are neutralised. The data question is different and harder: rollback reduces distance proportionally, but is that proportion the same at every swing speed? If the loss is a fixed percentage, long hitters lose more absolute yards and the leading group compresses. If the loss only appears meaningfully at high swing speeds, long hitters lose relative advantage and the balance shifts entirely toward shorter, more accurate players.

Two schools of modelling produce opposite answers. I pick neither. This is a question shot-level tour data will answer in the first three seasons after 2028; current models are controlled guesses.

For Vietnamese golf, the relevant note: if the goal is to protect courses from being overpowered, adjusting course setup — rough thickness, green speed, tee positions — is cheaper and faster than changing the ball standard for an entire industry. A course in Vietnam can do that within a season. The global ball market takes five years.

Wind, terrain and an unadjusted baseline

Every Strokes Gained model rests on a baseline implicitly built in average conditions. On the PGA Tour, average conditions are a real, measurable, narrow concept. In Vietnam, wind is the default rather than the exception.

Hoiana Shores, BRG Da Nang, Vinpearl Nam Hoi An — three central coastal courses with entirely different wind regimes and one shared trait: the value of each shot depends on wind direction and speed at the moment of the strike. A 165-yard approach into a 6 metre-per-second headwind is not the same shot as 165 yards in calm air.

The practical consequence is concrete. A 74 on a windy coastal course can be a better round than a 71 on a sheltered inland course, and the scorecard will never tell you. Apply a PGA Tour baseline to a central Vietnam round and you are measuring your own house with someone else's ruler.

Here is the smallest change with the largest effect: record three variables. Driving distance by hole, estimated wind direction and strength, and putts grouped by distance band. Three variables. About five minutes per round. After roughly two hundred rounds you have enough to build a wind-adjusted baseline per course, and every argument about player form changes in kind.

Three times I learned this in another sport

In 2026, aged twenty, an international communication student in Binh Duong, I built an xG model in Excel and ran it across twenty-six V.League rounds. Result: Quang Nam won the title averaging 48 percent possession, lowest of the leading group, with a 17.5 percent conversion rate among the league's best. I wrote "The champion that did not need the ball" and was mocked. Three months later Quang Nam were champions.

The lesson was not that I guessed right. The lesson is that possession, like driving distance in golf, describes what happened rather than explaining why. A side holding 52 percent of the ball and losing 1-0 looks "in control" in every bulletin.

In 2026, working as a data assistant for a sports outlet at the Russia World Cup, I reviewed Germany's previous four matches after their 0-1 loss to Mexico. Mexico's PPDA was 8.7 — extremely aggressive pressing. Germany needed 11.3 passes per defensive action. Their midfield generated just 0.89 xG despite 61 percent possession. I wrote "The rusting machine" before the final group game. Germany lost to South Korea and went out. The piece reached forty-five thousand views.

The lesson there was structure. I did not write "Germany are playing badly". I identified which metric warned of failure before the match. In golf, the equivalent warning metrics are Strokes Gained approach and fairway hit rate on long holes. They forecast far better than putting.

In 2026, the pandemic closed stadiums. I joined Becamex Binh Duong as a data assistant and found the V.League home win rate fell from 49 percent in 2026 to 38 percent without crowds, across a screen of forty-two matches. The coaching staff wanted to keep identical home and away plans. I objected, presented the comparison table, and proposed proactive defending away from home. The team won four of the next five.

The empty stadiums of 2026 made me ask: does home advantage come from the ground or from the crowd? The data has an answer. The same question now waits for Vietnamese golf: does a player's course familiarity come from the turf, from green-reading experience, or from having family standing behind the ninth green? Those three sources imply three different training approaches. Today we merge all three into the word "familiar" and nobody checks.

Numbers do not lie. Reputation whispers into the ear of anyone who does not read the table.

The contrarian angle: four traps in Vietnamese golf coverage

First, the selection effect in scrambling. The scrambling leader is usually not a player with supernatural recovery skills. He is simply a player who misses more greens than necessary. Every green hit adds a chance to sit in the scrambling sample; every miss adds an opportunity to be penalised. Reading scrambling without checking GIR is reading the tip of the iceberg.

Second, GIR does not equal good approach play. Hitting the green from thirty metres on the far side of the pin differs entirely from hitting it from three metres. Same GIR column, two different shot qualities. GIR is a counting metric, not a quality metric. A count can never replace a measure of quality.

Third, putting is the most seductive and least predictive category. Every story about "a hot week on the greens" has a shelf life shorter than one season. Professional teams know this and still sign contracts on putting, because putting is the thing viewers see most clearly on television.

Fourth, and most important: a data gap is not a failure, it is a finding. The empty report I received was not wrong. It was honest. What is wrong is our reflex: meet a blank cell and fill it with an adjective. No Strokes Gained, so write "steady form". No putt data, so write "good feel on the greens". Every adjective inserted is a permanent loss of data, because nobody goes back to measure.

I hate uncertainty. But 2026 taught me that one unforeseen variable can outweigh every algorithm. An empty report is such a variable. It does not break the model; it shows which layer of the model is short of material.

A minimum gate before publication

When a golf data pipeline finds no information points in a source article, the correct behaviour is not to emit a professional-looking report with full headings and "insufficient information" throughout. The correct behaviour is to halt and raise an error at the ingestion layer.

I propose a minimum gate that any sports data team can apply, even a two-person team on Google Sheets:

One, count information points. Below three, halt the pipeline; do not publish.

Two, check title and source. If both are empty, treat the input as invalid.

Three, count extracted entities. If zero, reroute to full-text ingestion of the source.

Four, capture publication timestamp and source URL at ingestion time, because after re-ingestion you may not recover them.

For Vietnamese golf, the editorial equivalent is simpler: every analytical piece must contain at least one traceable metric, with context for course, wind and season stage. A piece with no metric is a commentary. Commentary is not bad, but it must be labelled as such.

I started a blog from a lecture hall, believing data would speak for itself. Eleven years later, I teach it to speak. The hardest part of that teaching is not building the model. The hardest part is knowing when to stay silent.

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

Looking ahead: three signals for the next cycle

First signal to track over the next twelve months: how many Vietnamese tournaments begin recording shot-level data, even for three holes per round and only two variables — driving distance and ball position after the approach. One event doing this for one season gives domestic analysis its first foundation.

Second signal: the appearance of a wind-adjusted baseline for at least one coastal course. When that happens, comparing form across venues becomes meaningful rather than a comparison of weather.

Third and most important: the number of times a Vietnamese golf outlet publicly states "we do not yet have enough data to conclude". That is not weakness. That is the maturity marker of an analytical culture.

The empty report sat on my desk for a week. I did not delete it. It is the best reference document I own on my own limits: a model is only as good as its inputs, and the only question worth asking at the end of any analysis cycle is not "what did we conclude", but "do we have enough data to conclude yet".

If the answer is no, what is the right next move?

Numbers do not lie. Writers can.