Why Starc's Price Halved: The Ledger of Cricket's Transfer Market

On the second day of the auction hall in Jeddah late last November, I had a spreadsheet open on my laptop screen. The left column held IPL 2026 auction prices, the right column held 2026. In Mitchell Starc's row, two numbers sat side by side: 24.75 crore rupees and 11.75 crore rupees. A 52.5 percent fall in a single year.

In that same year, the bowler's T20 indicators did not fall 52 percent. His death-over economy did not rise, his rate of losing wickets in the powerplay did not rise, his boundary-conceding tendency did not jump. So what did the number actually measure?

After three days of picking through the gap between those two columns, I found something the auction sheet never prints. A price is not a valuation of a cricketer; a price is the measure of a buyer's fear. The franchise paying the money is not buying the cricketer's past — it is buying its own uncertainty about the coming season. The greater the uncertainty, the higher the price, because fear has no ceiling.

The paddle goes down, the money goes up. But the arithmetic running behind the paddle never appears on camera.

Context

This piece is written from the middle of a transfer window. In cricket, a transfer window no longer means a few fixed days in the year, the way it does in football. The IPL mega auction in November, ILT20, SA20 and the BPL in January, the PSL in February, The Hundred in April, MLC in July — all twelve months are now trading season. The player's agent, the franchise's strategy department, the auction analyst, the fantasy platform and the bookmaker all look at the same data and reach different conclusions.

The scale of the money needs stating. The value of the IPL's media rights for 2026 to 2027 is roughly 48,390 crore rupees. A large share of that flows to the franchises, and the franchises pour it into player wages. The salary cap per IPL team is now around 146 crore rupees. The entire player-wage pool of the BPL's seven teams is a small fraction of that.

In 2026, the first model I built for the BPL from a small office in Motijheel stood witness to the shift from paper scouting to digital tracking. That year Abahani Limited Dhaka carried 2.4 xG per match but scored only 1.8 goals. I showed the gap to the coaching staff, they ignored it at first, and then after losing the Federation Cup semi-final 0-2 despite 2.7 xG, they called back.

At the 2026 World Cup in Russia I calculated PPDA for all 64 matches, from Dhaka, working through the night. France's 8.4 PPDA was the lowest among the semi-finalists — a deep defensive block. Their transition output was 1.8 xG per match, the highest in the tournament. Before the final I wrote that France would beat Croatia. They did. Then I spent 72 hours re-checking every number before publishing.

PPDA is not a metric; it is a confession of how a team wants to suffer. An auction price is the same. The price does not say how good the cricketer is; the price says which kind of suffering a franchise is willing to tolerate.

The arithmetic of the field and the arithmetic of the market never match. On the field, one ball costs one run. In the market, the price of one ball depends on who can bowl it, how many others can bowl it, and how many of those are asleep in a hotel tonight.

One thing needs clearing up first. I am not calling any auction stupid. I am saying that what an auction reveals and what it hides are two different things. The spreadsheet was never the enemy; my blind trust in it was.

Core analysis

The first thing that caught my eye was that the relationship between price and performance sits close to zero. For the first forty cricketers sold at the 2026 mega auction I built a composite index — strike rate, dot-ball percentage, death-over economy, powerplay wicket rate and an adjustment for opposition quality across the last two T20 seasons. Then I looked at the gap between that index and the price ranking. Rishabh Pant's 27 crore and Shreyas Iyer's 26.75 crore were not the top two on my index. Many of those who ranked high on the index went for comparatively less, or sat unsold altogether.

The sample is small, I admit it. Forty names cannot settle a market, and one auction's data cannot predict the next auction. But the pattern that emerged did not require me to force a search — the pattern found me inside the data. In my reading, price forms on three layers.

Layer one — role scarcity. A franchise does not buy the best cricketer; it buys the rarest cricketer. A T20 innings has four jobs very few people do well: taking the new ball in the powerplay, left-arm spin through the middle, yorkers at the death, and fast runs at five to seven. A cricketer who can attach his name to one of those four commands a price above his cricketing quality.

Starc's fall is legible here. As a left-arm quick he was scarce in 2026-24, because he could take the new ball and bowl the death overs. By the 2026 market the supply of left-arm pace had risen, scarcity fell, and so did the price. The cricketer did not change; the supply did. This is not a story of demand, it is a story of supply.

Layer two — the recency premium. The market has amnesia. A good tournament inside the last three months inflates a price enormously; the same performance two years ago is worth nearly nothing. Immediately after the 2026 auction, players who had done well in the franchise leagues of that moment saw their prices leap. That is not market value, that is tape recall. The shorter a memory, the more volatile the price.

Layer three — ceiling and floor. Every franchise has a budget ceiling and a minimum-spend floor. The curious part is that the floor calculation is entirely separate from the logic of value. For a team that has already spent heavily, the marginal price of the next cricketer is close to zero. For a team that has spent nothing, the marginal price is astronomical. The same cricketer, on the same day, costs two different teams two different amounts.

Retention versus auction. The retention process before the auction is a separate market, and there the price is set by relationships, not performance. When a franchise declines to retain a player, that is a signal in the market — either a cost calculation or a dressing-room calculation. Decisions made outside the auction hall never appear in any index.

The uncapped market is the most opaque. For cricketers playing the domestic circuit outside the IPL in Bangladesh, data is not systematically stored anywhere. Chasing ball-by-ball data for a Dhaka Premier Division League match once cost me three weeks, and at the end I had a paper scoresheet and two handwritten notes. When Fortune Barishal or Comilla Victorians sign a domestic cricketer, the decision rests on video from a limited number of matches and the word of two or three people. There is no index here, only estimation.

The BPL's structural problem. The BPL auction is category-based. Players are sorted into A, B, C and D tiers, and franchises price them by tier. One side of this structure is good — it controls cost and keeps the league alive. The other side is bad — it erases variation in performance. A domestic bowler who has been excellent at the death for two straight seasons and a bowler who has been excellent through the middle can land in the same tier. So the BPL market never prices role scarcity, only names.

My index says the highest-priced cricketers in the BPL are almost all batters, and a large share of them are regular national-team members. Yet the league is won mainly through death bowling and middle-over spin. Fortune Barishal's titles show exactly this pattern. The market and the match requirement are looking in two different directions here.

There is a human cost too. A domestic pacer taking new-ball wickets for four seasons without a national call-up stays trapped inside the category structure. His work gains value on the field and none in the market. That gap is the biggest loss in Bangladesh's fast-bowling pipeline.

Nobody calculates the agent layer. What a cricketer earns also depends on how well his agent maintains relationships with franchises. Sometimes three players under one agent are sold together — what the market would call a package. That package never shows up in any index. This is where the question of data provenance cuts sharpest: who built the number I am handed, and why did they build it — without that question, the analysis is only half done.

The women's league is a separate market. In the WPL auction the price range is far smaller than the IPL's, but the logic of the decision is nearly identical. Batters are priced high, bowlers low. The one difference is that in women's cricket the ball-by-ball data history is shorter, so model uncertainty is larger. Making a big claim on a small sample is easy; the responsible work is keeping the claim small.

A parallel market looking at blockchain. Over the last few seasons, fan tokens and cricketer-card platforms have built a separate place where prices are set. Here the price does not come from auction logic but from fan imagination and trading liquidity. This market's data is visible on-chain, so it looks easily verifiable. Visible data and true data are not the same. In my model this market showed no predictive power at all, only a spread born of thin liquidity. I build models the way monks copy manuscripts: slowly, and with fear of error. On-chain data cannot be used without that fear.

The domestic pipeline story. Bangladesh's domestic cricket produces a particular type of cricketer — good middle-over bowlers, good new-ball bowlers, but few T20 finishers. The fifty-over culture still dominates the domestic league. So to meet finisher demand at the BPL auction, franchises have to look abroad. That is not a player's failure, it is a structural output of the pipeline.

The contrarian angle

This is where I have to be most careful. On hearing that the relationship between price and performance is near zero, many people jump up and say the franchises are burning money stupidly. I do not believe it.

First, a zero relationship does not mean the price is false; a zero relationship means my index is incomplete. A franchise holds information I do not — dressing-room chemistry, the true state of an injury, the specific role that cricketer plays in the coach's plan. In 2026, when stadiums were empty, I combed through 312 matches and found home advantage had dropped by 0.34 goals per match, and the primary driver was referee bias, not crowd support. That was the first time my own playing experience and the data contradicted each other. Week after week I watched my own tapes from the 1990s to reconcile the two. The lesson was singular: a variable I do not measure does not cease to exist. When the stadiums emptied, the home advantage did not vanish — it relocated. The same holds for the market: the cause I do not measure is still priced in.

Second, an auction is a strategic game. The price one franchise offers may exist to block a rival. There the price is politics, not cricket. Every transfer fee is a story the market tells to hide its own uncertainty. The bigger the story, the greater the uncertainty.

Third, "small clubs buy good cricketers cheap" is also a story. We remember the ones who succeed; we forget the ones who went cheap and failed. Survivorship bias here is enormous. By my count, cheap signings do not succeed at a higher rate; their stories are simply more attractive.

Fourth, the belief that a big price produces failure is unproven. Research on the link between transfer fees and performance in football is mixed on any negative effect. In cricket the research is thinner still, and in the Bangladeshi context it is almost nonexistent. Showing confidence where no answer exists is easy; that is not analysis, it is performance.

Takeaway

At the next transfer window I will watch one thing: whether the price of role scarcity rises further. If left-arm death bowlers and number-five finishers leap again at the next auction, I will read it as the market moving toward cricketing quality. If the stars of the most recent tournament once more occupy the top rows, I will read it as a market still suffering amnesia — and I know no cure for that disease.

The second number I will watch is the unsold list. In a market where the unsold count rises, the logic of price-setting is weakening. And in a market where the logic is weak, every paddle that drops carries an estimate, not a truth.

The rest is arithmetic, time, and a little fear.