The 31 Empty Fields: In Cricket's Transfer Window, the Slot Sets the Price, Not the Strike Rate

A January morning in a small London office. I opened the template: 42 columns, each with a name, a unit, a source. I loaded the three names circulating hardest in the market that week. Thirty-one of the 42 fields came back empty. There was a name. There was one number. Nothing else. How many days would he be available this season? Does his central contract carry a franchise-release clause? How many times in the last three years has his no-objection certificate been withheld? No answers, because no feed records them.

The first thing the template does is tell you what it cannot see. When I built the original 42-field match template in London in 2026, I thought the job was collecting numbers. Six months later I understood the job was identifying empty boxes.

That same week another document landed. 19 December 2026, Dubai, the IPL 2026 auction. Mitchell Starc, ₹24.75 crore — the highest price in auction history. He had not bowled a ball in the IPL since 2026. In the same room Pat Cummins went for ₹20.50 crore. Neither man's T20 economy is a world-leading benchmark. The market paid anyway, and paid heavily.

This piece is an attempt to draw a line between those two documents: 31 empty fields on one side, ₹24.75 crore on the other. They tell the same story from opposite ends.

In cricket, "transfer window" is a borrowed phrase, and the borrowing is poor. Football has Bosman: a contract expires, the market sets the price. Cricket's market is a set of fragments, each with a different pricing instrument.

December brings the IPL auction: ten teams, a fixed purse, a set-based order, a rule permitting eight overseas players in a squad and four in an XI. January brings ILT20 and SA20 alongside the back end of the Big Bash and the Bangladesh Premier League. February and March bring the PSL. August brings The Hundred draft. English county deals are done in autumn, with overseas numbers governed by each county's own arithmetic.

One variable never holds still across those fragments: the player's calendar. And the calendar is not his. It belongs to his board. The bilateral dates already printed in the ICC Future Tours Programme cannot be moved by any franchise auction. So in cricket's transfer market the scarce resource is not talent. It is days.

This is where my two markets — Dhaka and London — separate. In Dhaka the question is usually whether the board will release him. In London the question is whether the county will, and in which week. Same player, same strike rate, two prices — because two permission structures.

Of those 42 columns, here is what I could actually measure: age, role, 24-month T20 strike rate, bowling economy, share of overs bowled at the death, batting position, domestic league appearances, central-contract status, yes or no. And one column that ended up deciding everything: days available in the season.

That single column took four days to add, because the data did not exist. Franchise player lists carry a name, an age, a country, a base price, sometimes one aggregate statistic. They do not carry NOC history, injury timelines, or clashes with bilateral fixtures.

So I rebuilt the template.

The first version added a column called squad gap. A gap is not a position. A gap is that specific deficiency an overseas slot is spent to fill. Ten franchises mean ten different gaps. Chennai's gap is not Kolkata's gap; Mumbai's problem is not Hyderabad's problem.

This is where Starc's ₹24.75 crore becomes legible. That price is not the price of Starc's T20 record. It is the price of Kolkata's gap. A side without a left-arm new-ball option and without an experienced death bowler wants one overseas slot to do two jobs. One slot, two innings phases. That doubling is what pushes the number.

Cummins's ₹20.50 crore follows the same logic for a different reason. Hyderabad was buying pace, but it was also buying a decision-maker. The auction sheet has no captaincy column. Neither did my first version.

The second example is cleaner. 23 December 2026, Kochi, the IPL 2026 auction. Sam Curran, ₹18.5 crore. Left-arm pace plus lower-order batting — two roles in one overseas slot. I call that slot efficiency. A player who fills two gaps with one slot is priced as the sum of two gaps divided by one. The arithmetic of the market is asymmetrical right there.

So the first conclusion went down without my hand shaking: an auction price is not a valuation of a player; it is a valuation of a team's gap. The same cricketer is worth ₹20 crore to one buyer and ₹2 crore to another, because the gap differs and the purse differs.

Writing that sentence cost me three rebuilds. I rebuilt the slot-value model three times before the window closed, and stopped at the third, because the third added a limit: I am inferring each squad's gap, not measuring it. Teams do not publish their intentions; only auction outcomes surface.

Version one tested the relationship between price and previous-season performance. Weak. Version two added availability — days genuinely available as a share of league days. The relationship tightened, but only downwards: low availability depresses price, high availability does not raise it. Starc is the proof. He offered no full-season guarantee and still fetched the record, because the gap was acute.

Version three set a threshold and stopped. The rule: below one-third of season days, a player drops out of the filter unless his role is scarce. And I had locked the definition of scarce in advance — left-arm new ball, death economy under seven, batting in the bottom three.

The spreadsheet is a monastery; every cell is a vow of consistency. If a cell fills from a different formula one day, the whole monastery is in question.

Now the real problem. Where did the availability column come from? News. Announcements. Board press releases. Which means the most important input in my model is my least reliable instrument. No official feed records NOC refusal rates. None records the release history of Associate players, or the franchise-obligation record of women cricketers.

I cannot close that gap. I can only mark it. So every output carries a confidence column — high, medium, low. The explanation of Starc's price is high, because auction outcomes are public. The explanation of his available days is low, because it rests on reporting.

From more than twenty years of watching matches, the lesson I keep is not about statistics. It is about gaps: the information nobody writes down is usually the most valuable information there is.

I do not trust a metric until it has survived a boring afternoon. A boring afternoon here means an ordinary Tuesday league match with no star, a thin crowd, one camera. Who bowls the death overs in that match, who is hidden in the field — those details set the price later. And those are precisely the scorecards that are least complete.

The transfer market does not lie, but it does negotiate with the truth. It states a price. It conceals the reason for the price.

Now the counter-argument. My slot-value model can explain past auctions. It cannot predict the next one. It should not try. Cause and correlate blur here.

Lower availability, lower price — I can measure that. But is availability a cause or a symptom? I think a symptom. A player his board releases reluctantly is usually a player with heavy bilateral obligations, which is to say an established international. Why is an established international cheaper? Because the franchise knows it will not have him all season. The real cause of the discount is not availability; it is the franchise's risk aversion. Availability is only its indicator.

Second weakness, and larger: this market is not efficient. Ten buyers, a hard purse, a clock — that is not a market. It is an auction with a sample size of ten. One team's mistake shifts the entire price record, because that single transaction becomes the largest data point in the set. Nobody paid ₹24.75 crore the following year. That does not prove the market learned. It proves the sample is small.

Third weakness, and this one is against myself: eleven of my filled fields came almost entirely from press reports, and press reports love big names. A player who produced evenly across 30 matches without ever making a headline left his box empty in my template. My index survives on survivorship bias.

An open confession. Of the three files I loaded that January morning, the most complete profile belonged to the least famous player — 36 of 42 boxes filled. He plays for a smaller nation, his board's rules are simpler, his calendar is less fogged. The star had nine of 42. We know more about famous players and less about their contracts.

So the honest version of this piece is this: I have built a model that explains the last three auctions, and I know exactly why using it to forecast the next one would be wrong.

In the coming window I will watch three things. The volume of NOC filings — a census of which boards release players quickly would move the availability column from reporting to feed. The dates of release clauses, which now sit inside franchise contracts and tell you who enters the market next. And the list of players who go unsold with top-quartile strike rates. Every one of them has an empty box behind him, and filling those boxes is how we learn what the market is actually buying.

The question stays open: if days set the price, what happens when the calendar gets longer? In a year with more franchise cricket, every player's available days fall. Do prices fall with them — or rise, because scarcity bites harder? My model cannot answer that. It can say one thing: the column that is still empty will be the most expensive column of the next season.