The Empty Column: LIV Golf and the Valuation Problem of an Unmeasurable Cohort

On October 10, 2026, I sat at my desk in my Nagoya apartment, opened my model file and ran a simple query: how many golfers under contract with LIV Golf were still inside the world top 100. I recorded twelve names, and next to them the timestamp of the query — a habit I have kept across seventeen years in this trade, because a quantity without a timestamp means nothing. That number twelve will change depending on the day you run it, and I always state that.

What stopped me was not the twelve. It was the three columns immediately to the right of those names in my file: Strokes Gained — Off the Tee, Strokes Gained — Approach, Strokes Gained — Putting. For those twelve players, all three columns were empty. Not zero. Empty. Those two states are worlds apart: zero is the result of a measurement that was performed, while a blank cell is a place where no measurement ever existed.

An entire professional men's golf analytics industry is operating on a spreadsheet with a human-shaped hole in it, and that hole widens with every season.

The 54-hole format and what it costs

LIV Golf launched in 2026 with three structural features: 54 holes instead of 72, a shotgun start instead of split waves, no cut, and a team leaderboard running alongside the individual one. On October 10, 2026, the governing board of the world ranking announced it had rejected LIV's application for ranking recognition. The stated reasons centred on exactly those three features. In March 2026, LIV withdrew its application.

World ranking points are not a trophy. They are infrastructure. They determine major championship exemptions, national team eligibility, negotiating leverage in sponsorship contracts, and the way a sponsor prices a golfer in a transfer deal. When a cohort of players is pushed out of that infrastructure, the first thing that disappears is not the chance to compete. The first thing that disappears is the ability to measure.

I follow professional men's golf through a data lens, and I have learned something uncomfortable: when a group of players is carved out of a shared measurement system, the market does not stop valuing them. It simply stops valuing them with numbers, and starts valuing them with memory. Memory is the worst valuation instrument I know, because it comes without an error bar.

Three pillars and the measuring machine

Strokes Gained is, at heart, a subtraction. Every shot is placed next to the average expectation for a shot of the same type, from the same distance, from the same lie. The gap between what happened and what was expected is the value of the shot. To perform that subtraction you need shot-level data: ball position, distance, lie, and the ball's position after the strike.

The PGA Tour operates its own tracking system and collects data on every hole of every round. LIV Golf publishes aggregate statistics — average driving distance, greens in regulation, putts per round — but does not publish shot-level data in a reusable format. That is the crux most public debate skips over.

The three Strokes Gained metrics are the currency of modern professional golf. Without them you cannot model a player's course fit, cannot separate putting skill from luck, and cannot forecast performance at a course the player has never played.

I once thought I could patch this hole with aggregate data. I tried, and I failed in a way I will describe below.

Without a cut, the average lies

This is the part of the analysis I consider the most important in the whole story, and it is almost entirely absent from mainstream commentary.

At a 72-hole event with a cut, roughly half the field is eliminated after round two. The final two rounds contain only the best players of that week. As a result, the scoring average of rounds three and four at a cut event is usually better than the average of rounds one and two. That is a selection effect, and every serious model must adjust for it.

In the 54-hole, no-cut format, all 54 players play all three rounds. The final round still contains the players at the bottom of the leaderboard. That means the final-round scoring average at a LIV event is dragged down by the lower half of the field in a way a cut event's final round is not.

The consequence is very concrete: if you place a LIV event's scoring average beside a PGA Tour event's scoring average and conclude that one cohort is weaker, you have compared two distributions truncated at different points. That comparison is methodologically meaningless, however convincing the output looks.

There is a further layer, and it is behavioural. A player who knows he will play all three rounds makes different risk choices from a player sitting on the cut line at the 17th hole of round two. A tee shot on the closing hole by a man who cannot be cut is a behavioural variable, and it appears in no statistical table.

The shotgun start erases the correction tool too

A shotgun start means every group tees off at the same moment, from different holes. For fairness, it is a good design: nobody plays in stronger wind than anyone else. For analysis, it erases an important correction tool.

At traditional split-wave events, the morning and afternoon waves face systematically different conditions. Serious analysts always account for this. When everyone starts at once, there are no waves to correct for, and weather variance is absorbed directly into individual scores in ways you cannot isolate. The noise floor of any model built on LIV data is therefore higher than that of a model built on split-wave data.

The team leaderboard compounds this. A player on a team leading by five on the closing hole will play differently from a free agent. The individual score you read is the output of a utility function with two objectives, and you only see one of them.

Proxy indices and the half-life

In 2026, when stadiums closed because of the pandemic, the club I worked for went two months without a match. With no match data, the coaching staff still needed a form forecast. I proposed using GPS training data and the precedent of historically disrupted seasons. They objected at first. I persisted because I had one concrete precedent to cross-check against, and in the end the model was used.

The lesson I took from that season is a formula: when the primary data source disappears, build a proxy index, and publish its error bar on day one. What you must never do is present the proxy as if it were the original measurement.

Applied to men's golf: the proxy index for a LIV player is built from his record before he left. The problem is that this index decays. The Strokes Gained profile of a player aged 28 to 35 drifts at a rate I estimate carries a half-life of roughly 18 to 24 months. That means after two years, half the signal in the old profile no longer describes the man.

The consequence is concrete and uncomfortable. A player who left the old system in 2026 will, by the 2026 season, have a proxy index that has largely decayed into noise. A player who signed in late 2026 still carries usable signal. Two players can be equals, yet the market will price the second higher than the first — not because he is better, but because the data about him is fresher.

That is where I believe most current analysis goes wrong. The market value of a LIV golfer today reflects the shelf life of the data about him far more than it reflects his ability. Elimination is the key to the transfer market, and here the elimination is removing the wrong people.

In December 2026, when the biggest deal in the sport was announced, I asked myself a question that I later realised was the wrong one. I asked: is this man still one of the best golfers in the world? That question cannot be answered with the data available. Data is never wrong; I simply asked the wrong question. The right question is: how wide is my error bar, and how much wider does it get after each season? That one I can answer.

The four-major test

The 2026 and 2026 major seasons once pushed back against my concern. In May 2026, a LIV player won the second men's major of the year. In June 2026, another LIV player won the third men's major of the year. Both had left the old system within two years of those wins. If my half-life model is right, that is exactly what should happen: the signal was still fresh, and it surfaced as results.

The 2026 season closed with four major champions — April at Augusta, May at Quail Hollow, June at Oakmont, July at Royal Portrush — and none of them on a LIV roster. I do not present this fact as proof. Four tournaments is far too small a sample to prove anything, and I would be the first to reject a conclusion drawn from it.

But I logged it in my model journal under the label "consistent with forecast, insufficient to confirm". What did NOT happen often tells the truth more clearly than what did, and what did not happen in 2026 was a major win by a player who had been outside the system for three years or more.

I should add something about myself. In 2026 I publicly admitted I had been wrong in analysing a major match: I read the pressing metrics and concluded the side I was tracking controlled the game, but I had ignored the opposition's running distance after the 70th minute. The result turned, and I sat down with the full video to find what I had missed. The variable I lacked was real-time physical load. I tell this story because it explains why I never conclude on form without physical data — and why I will not conclude on the LIV cohort without shot-level data. Gegenpressing does not break the data; it breaks my assumptions. So does a data gap.

The harmed party is not the only responsible party

The story currently told is tidy: a ranking body pushed golfers to the margins, and they paid the price. I think that telling ignores half the picture.

The real constraint here is not the ranking board's decision. The real constraint sits in LIV's own design. A ranking system awards points based on field strength, and field strength is measured by open access. When a tour operates with 54 invitation-only places, no open qualifying, no pathway upward from regional tours, its ranking value is capped by that structure. The same arithmetic governs every sport, not just golf.

I hold a fairly hard professional belief about this: a closed ecosystem can buy established stars, but it does not manufacture new ones. It spends from an existing fund of fame rather than creating a new one. In men's golf, that fund starts running dry after about four seasons.

I want to state the limits of this argument clearly, to avoid fooling myself. Two format features — the number of rounds and the cut — are fixable. An open qualifying route could be established. Shot-level data could be published in an open format. There is no technical obstacle to any of the three. So when someone says LIV had no other option, I do not believe it. I believe in cultivated probability, and here the probability was never cultivated.

The notable point is that LIV does not need ranking points commercially. Its business model rests on media contracts, sponsorship and the pulling power of big names. Ranking points are a cost, not a revenue line. So this gap has the shape of a product choice. The problem is that the people paying for that choice are not the organisers but the golfers — men priced by public memory rather than by their own data.

What to watch in the next cycle

Four signals I will log for the coming season.

The composition of the four major championship fields, and the pathway by which each LIV player receives his place. An exemption, a regional pathway, or a legacy amateur ranking. Whichever pathway appears most often will show whether the system is opening or closing.

The proportion of LIV players in the two squads at the Ryder Cup in September 2026 at Adare Manor, Ireland. This is the cleanest public test available to me: a national team is the only place where selection is decided by humans rather than by a ranking algorithm.

The progress of negotiations between the major tour bodies and the investment fund. If a merged entity emerges and adopts a single shot-level data standard, the biggest winner will not be the players. The biggest winner will be the valuation market.

And finally, a design change I consider more important than every legal dispute combined: whether an open qualifying route into a LIV event exists. That change alone would shift the ranking arithmetic more than any argument.

A gap in the table can speak, if we are willing to listen. What I hear in that empty cell is not an accusation. It is a question about design, and it will be answered not by a press release, but by an entry list on some Thursday morning.

I will keep that empty cell in my spreadsheet for one more season. Not because I like gaps. But because a table with a hole in it is still more honest than a table filled in with the writer's guesswork.