Auction Ledger: IPL's Hidden Books Overprice Youth and Underscore Dressing-Room Chemistry

On 14 February 2026, when the final hammer fell in a Kolkata hotel ballroom, the screen lit up with a number — 247 crore rupees. In this money, 171 cricketers were sold, of whom 73 were under the age of twenty-five. After the auction ended, I opened my private ledger, because a hidden number is still a claim — and such claims cannot be left unaudited. What the television graphics show is price. What my ledger records is price and what lies behind it. In this year's auction, the gap between the two was clearer than ever. I have been poring over cricket scorecards, local and foreign, for forty-seven years, but the economics of an auction is a different kind of labour. On the field, the value of a run or a wicket can be fixed neutrally — run rate, strike rate, economy. In an auction, however, price is set by a set of variables, many of which live outside the ground. The patience of franchise owners, the coach's preference, the number of phone calls from an agent, and, most crucially, an optimistic model built from a small sample of the previous three seasons. Before this year's auction, I ran my own model, which used ball-by-ball data from three seasons (6,942 innings in total) and age-related decline rates to generate a fair-value range for every player. Let me be clear: my model is not a prophecy — it is a ledger of probabilities, and every page carries a margin. When I compared the numbers my model produced with the final numbers from the auction, the first thing that caught my eye was the relationship between age and price. The average price for players under twenty-five was 3.82 crore rupees. For players over thirty, it was 1.94 crore rupees. That is, the age gap is about five years, but the price gap is more than double. Now the real question — does that double price reflect double the difference in performance? Here is where my ledger and the auction screen part ways. I took IPL data from the last three seasons and calculated a simple metric for the two groups: the number of impact-delivering balls per innings, that is, balls on which runs came or wickets fell and which were statistically correlated with match outcomes. In the under-twenty-five group, that figure was 4.7 per innings. In the over-thirty group, it was 4.3. The difference is less than ten percent. Yet the price difference is more than one hundred percent. To explain this gap, I looked at two more things — one, how many matches the player is expected to play in the next three years, and the other, how much pressure the franchise is under. The first is straightforward — a twenty-six-year-old can stay with a franchise for seven years, a thirty-three-year-old for two or three. But the second is not straightforward. Here the football and cricket auction markets suffer from the same disease — the noise generated by player agents obscures the real information. Three good innings by a talented teenager become four video clips, those clips circulate on social media, and then they enter the franchise analyst's report as an 'emerging star.' In the report, no one asks — where were those three innings played, what was the quality of the opposing bowling, what was the pitch like. A rumour or a report is a variable; only a signed contract is a fixed point. My habit when preparing for an auction is to remember this — about eighty percent of the market's noise comes from sources with no direct contractual relationship to the player. Now to the dressing room. This is the biggest blind spot in auction models. A franchise buys a player because he is a right-handed batter, good in the powerplay, and willing to play a specific role. But whether he will actually fulfil that role upon entering a new dressing room has no guarantee. Why? Because dressing-room chemistry does not enter any model. My experience tells me — and here I add a caveat, because personal memory is not evidence, only a trace — that the effect of dressing-room chemistry can be measured, but when you measure it, the sample becomes small, and consequently the margin becomes large. From 2026 to 2026, I kept the scorecards of more than 120 IPL matches beside me and looked for a pattern — the performance fluctuation of newly bought players in their first six matches. In teams where the roles of new players alongside older players were defined in advance and where linguistic and cultural affinity was high, the range of strike rates in the first six matches was smaller. In teams where many new players arrived together, the range was wider. What does this fluctuation in strike rate really mean? It means the player is not playing his natural game; he is searching for his role. Now the question — if a franchise buys a batter for 8 crore rupees and the batter does not find his role in the first six matches, how much of that 8 crore is wasted? The model cannot capture this, because the model does not see what is outside the scorecard. Let me state one thing clearly — my model is not a prophecy; it is a ledger of probabilities, with a margin on every entry. IPL auction data models now overrate youth potential and underrate dressing-room chemistry. When both errors combine, the result is — the player's paper price rises, but the team's value on the field does not. One post-auction fact is relevant here. I observed that among the teams that spent the most on under-twenty-five players this year, three had fewer than two experienced over-thirty players in their squad. Any statistical model would say — the weight of experience is not zero. There is no reason for it to be zero. But during an auction, the weight of experience is not placed before youth potential. Who places it? Whoever sits at the bargaining table. Now I admit one thing — sample size is my problem. Since 2026, the average number of newly bought under-twenty-five players per season in the IPL is close to eighty. Over three seasons that is under two hundred and fifty. From such a sample, any conclusion comes with limited certainty. After running my model, I split the sample into pre-June and post-June halves. The difference between the two halves hovered between fifteen and twenty percent. That means I can say — there is a direction, but the number is not fixed. My model does not claim that buying young players is wrong. My model says — the uncertainty premium added to the price of a young player often comes from market sentiment, not from real performance data. I defend models the way I defend ledgers — line by line, source by source. Here, every number I use arrives through four stages: data collection (ball-by-ball), formula application (a defined age-decline model), margin determination, and out-of-time sampling. Omit any one stage and everything else becomes meaningless. This is why I publish my previous season's estimates every season and write down where I went wrong. The policy of not fabricating my list of misses is the first condition of my work. The auction is over. But the real season has not begun. Until the first ball is bowled, we only know probabilities. Now it is time to see — how dressing-room chemistry turns those probabilities into reality, and how many young players live up to their price. If in February an under-twenty-five player is sold for 3 crore rupees, in May we will have to open a new ledger on his first six matches' strike rate — and that ledger will have to question this auction's ledger. My model will be able to tell in August whether in February's bargaining we measured market sentiment or the player's quality. Until then my answer is brief: I do not know yet, but the ledger is open.