Auction Price vs Load Ledger: Who Is Being Mispriced in the T20 Market

Auction Price vs Load Ledger: Who Is Being Mispriced in the T20 Market

December 2026, the IPL auction hall. Mitchell Starc's name was read out and the bid climbed to INR 24.75 crore — the highest in IPL history at that moment. My laptop was open on a spreadsheet: three seasons of T20 death-over economy, powerplay wicket rate, high-intensity over counts, and injury absences, stitched into a rough valuation model. Starc's death-over numbers were not in the league's top ten. Several bowlers ahead of him statistically were priced at a quarter of his fee. I built the Croatia xG model long before I learned to grieve a missed chance — that night I understood the auction does not price bowling; it prices a story.

The story is called recency. Whatever happened in the last six months carries four to five times the weight. A final spell, a series-winning innings, a viral delivery — these enter the model as an opaque, heavily weighted column named 'clutch,' for which nobody writes a definition. And an opaque column inflates forecast error.

Not one market, but four

We are in a transfer window now, but cricket has no single transfer window. IPL retention and releases, the ILT20 draft, SA20's central contract structure, the BPL's direct bidding, the PSL, the Lanka Premier League — each runs on a different currency, a different tax regime, a different NOC rule. When a franchise tries to tie a player down across multiple years, its real rival is not the league next door; it is the player's own body.

In franchise football the architecture is explicit: release clauses, buy-outs, wage bills, amortisation, all written down. In cricket it remains blurry. ILT20 and SA20 are moving toward central contracts; the IPL stays auction-driven. So the same player carries four different prices, and information asymmetry matters more than supply and demand. The player Bengali media calls out of form is 'undervalued in the powerplay phase' to a hyperspecialised data firm in Australia. The second buyer sets the price.

What gets paid for versus what should be

For bowlers I use a simple valuation frame: value = (powerplay wickets per over) × 1.4 + (inverse of death economy) × 1.8 + (sustainable high-intensity over ceiling) − (annual depreciation above age 30). The 1.4 and 1.8 are phase-impact weights I derive from tournament-level regression, not from auction prices. On this frame Starc is not a bad bowler — he is good, especially with the new ball. But the expectation created by INR 24.75 crore does not get met within twelve to fourteen months, because his death-over economy is historically mid-tier and the age-depreciation curve has already begun.

The auction buys recency and sells consistency — that mismatch is market inefficiency. Rishabh Pant's INR 27 crore in the 2026 auction (Lucknow Super Giants) sits in the same logic. The supply of international-grade wicketkeeper-batters is six or seven against ten franchises. In scarcity the price detonates. But a buyer looking only at shortage, not at leverage, pays double for the same asset.

One more thing stands out: nobody prices workload. I tracked Pedri across Euro 2026 and the Tokyo Olympics; he played 73 matches in the 2026-21 season, completed 92.3% of his passes at the Euros, and his high-intensity distance fell 11% in Tokyo's extra time. In franchise cricket the drop is sharper, because the calendar is denser and the travel longer. How many overs a fast bowler has delivered in the previous 18 months appears in no auction column.

You cannot simply drag football's xG model into cricket — I have to concede my own limits here. Bowling 'chance quality' is not measurable by strike rate or xG, because delivery quality and batter decision-making are entangled. What is measurable: which phase, how many overs, what line-and-length pattern produced what outcome, and how repeatable it is. My ledger stays inside those three. I do not overclaim.

Mistaking correlation for cause is the biggest error

The oldest auction proverb: whoever spends most wins the trophy. This is not causation; it is the correlation between annual revenue, liquidity, and the talent market. The franchise with more money already has better scouting data, better support staff, and stronger retention leverage. Spending is a symptom, not the source.

My old natural experiment on silence and venue is relevant here, conditionally. In the 2026 Bundesliga Project Restart, I found home win rate fell from 43.3% to 33.3% behind closed doors, and my regression model put the away-side gain at 0.18 xG per match. Cricket had a version of this test in ILT20 and SA20, at neutral or semi-neutral venues where 'home' is largely nominal. Empty stadiums taught me that silence is a variable, not an absence. But the limits must be stated: cricket home advantage depends chiefly on pitch preparation, not the crowd. If the surface is shaved the same way, spin economy shifts more than the return of spectators does. I avoid generalising — I do not write one sport's law from another's data.

Still, one signal is clear: in consistent environments where venue privilege is thin, squad depth returns more than one or two big auction names. That, to me, is the real purchasable asset.

What my model is dropping

First, reading players as assets makes us forget who owns the asset. If a franchise extracts 22 fast-bowling spells in seven months, it keeps the balance sheet clean by parking the depreciation risk on the body's owner. What American capital markets call an externality, cricket has not begun to price.

Second, the gap between a batter's available matches and selected matches. A presence percentage looks satisfying in a table, but without a selection-policy model we miss that a player features more for smaller sides and less in big matches. Selection is portfolio construction. A coach who always trusts five senior bowlers in big games never grows a young quick, because the high-leverage overs move to experienced shoulders and their capacity burns faster.

I have long argued that former stars' academies are mostly branding exercises. An academy means buildings, certificates, photo sessions. What is chronically underfunded is coach education: how to write a load calendar for a 14-year-old, how to set age-banded over limits for a fast bowler. The market will buy the photograph and will not fund the spreadsheet. Some quiet resentment accumulates there.

What I will watch next window

When I write retention numbers, I now keep three columns apart: money, overs, and rest. I do not tell a player he is wrong when he wants out — I check what the model omitted. The spreadsheet was my cloister; the auction hall was the first pilgrimage, where men raise hands and the accounting comes later.

The cheapest thing in the market is a rumour; the most expensive is verified workload. Whoever learns that difference will make calls in this transfer window they will not have to withdraw next season.

One question to leave open: if your team's most expensive player is also its least accounted-for player, who is at fault — the model or the table?