Auction Price, Pitch Price: The Number Nobody Codes in the BPL Transfer Window

The price rises from a strike rate at the draft table. One franchise watches a single innings — 74 off 38, three sixes, two long-ons — and writes "match-winner" on the sheet. My hand-coded ledger has eleven innings for the same batsman, and the picture differs: one boundary every 24 balls against spin after the powerplay, a chase strike rate falling from 129 to 108. One player, two prices — one auction, one pitch. In the BPL transfer window, the gap between those prices is the real story, and nobody codes it. The BPL market rests on four numbers: player category, base price, salary cap, foreign quota. Launched in 2026, the league began with an auction and later moved decisively to a draft system where franchises pick and the BCB fixes base prices. In practice, the real negotiating floor is not the auction hall; it is the category room. A player filed in the wrong category gets the wrong price, and that error rides the balance sheet for a whole season. The foundation is shaky. Categories lean on recent performance — a blurred mix of domestic and international output with no single verifiable store behind it. National Cricket League scorecards live in one place, Dhaka Premier League records in another, Under-19 and Under-23 in separate files, club records in private notebooks. No API, no feed, no standard schema. Out of that vacuum, prices are set not by evidence but by impression. I built my first dataset from zero in 2026. Watching every match twice, tagging events, 1,200 entries across six-hour keyboard sessions. That is why I wrote: I did not trust the BPL's numbers until I had coded them by hand. And hand-coding revealed the most important thing was not batting — it was bowling. The first myth to break was the sanctity of strike rate. In domestic T20, strike rate is the most visible number because a single innings can create it. Split those innings into powerplay, middle and death phases and the picture changes: many batsmen build overall strike rates in the powerplay, when the circle is compulsory and only two or three fielders sit deep. In the middle overs spinners bowl with an up-field ring and risk rises. In my 24-match hand-coded set, the average drop across those phases ran 21 to 25 runs per 100 balls. Franchises never write those 25 runs on the category sheet. Death-over economy is the second blind spot. It is the franchise's most expensive asset and the cheapest priced one, because tables show wickets, not economy. A bowler who concedes 22 in four overs without a wicket banks nothing. Yet strangling the last four overs from a scoring rate of 11.5 to 9.5 removes roughly 8 to 12 runs of expected score. A model without a decision is a diary, not a weapon: numbers that never earn a place in the XI may as well not exist. The category system itself is the structural weakness. Batsman, bowler, all-rounder — boxes so wide that the differences vanish inside them. A leg-spinner who bowls in the powerplay and one who bowls the 16th over are different products filed in the same room. An opener and a finisher share a box. The classification follows passport professions, not on-field roles. Wicketkeeping is the third neglected number. Glovework is barely measured in domestic cricket because stumping and catch events are not logged anywhere. A keeper's price is set by his batting position. Lateral movement, standing-back catches, diving stumpings off turning balls — no valuation exists, so no price exists. My 24 matches had no keeping events not because I erred, but because the source stopped there. The fourth gap is haste. The bridge from age-group to senior cricket is narrow. U-19 graduates enter senior domestic cricket before physical maturation, especially fast bowlers. Those physically ahead in age-group cricket succeed, and that success buys a BPL category slot. If workload data on 19-to-22-year-old pacers were tracked series by series, the first two years after a U-19 World Cup would show a sharp jump in over-load. Early maturity plus senior responsibility does not add injury risk — it multiplies it. Markets measure potential, not risk. Put those four gaps together and the BPL window looks less like talent discovery and more like an annual re-pricing of the same forty to fifty names. New names arrive through gap-filling inference, not proof. Who replaces whom is decided by who performed last year — itself a function of who got matches. Opportunity distribution and performance distribution blur into one label. A West Indies or Pakistan domestic middle-order batsman is easy to adjust for the BPL because ball-by-ball records are public. Adjusting a Bangladeshi batsman is harder because no measurable bridge exists between BPL and National Cricket League standards. The fix is institutional, not technological. In my own messy spreadsheets every match had two entries — one copied from the BCB scorecard, one coded from my own viewing. When they disagreed I hunted for which fixture it was and whose card was wrong. That verification is exactly what a shared, tamper-proof ledger would do if it existed: write once, timestamped, unrewritable, with the board, the franchise and the agent reading the same number. In cricket, the blockchain argument is not about cryptocurrency but about agent disputes — who owns a record. Right now Bangladesh is not fighting over ownership; the record does not exist. The most overlooked number is the release and injury decision. Window talk circles price, but squads change through release clauses and workload management. An all-rounder playing both roles accumulates batting load and bowling load that we never view as one volume. A franchise buying a single dual-role player looks economical, but in the final four matches that same player's power-hitting distribution drifts. No table tracks the drift. Here is my biggest tension. To verify numbers in 2026 I hand-coded 24 matches because no single source was reliable. That taught two contradictory lessons. A number you have not personally verified cannot support a confident decision. And a number that waits for full verification never gets built. Both are true. The difference is the threshold: decide in advance how much evidence each decision requires. A transfer threshold lower than a match-day threshold is not a sin, because the decision is itself a bet on incomplete information. One truth stands: what the market sells is not performance but the faulty memory of performance. Reports come from the BCB and scorecards; decisions come from franchises. Two different papers. A player consistent all season is underpriced at the draft table, because consistency is a series-level trait while the table rewards a single innings. A flashy knock makes history; a steady 40 makes a table. So the question is not the auction price. The question is whose data. If a team strategy says "pick openers on strike rate" — whose strike rate? Which season? Which phase? Verified where? Information nobody owns is worth nothing, and that is what franchises buy each year. Where a resource exists, markets settle. Where it does not, people gamble. Without a verifiable central store of all-format performance, every transfer window will repeat the same story — who is more believable, who is cheaper, who costs more. Paper sets the price, not the pitch. And the pitch will answer, but only next July, when the contract has already ended. If a franchise ever builds its own tracking, it will be team-level and player evaluation becomes defensive. If the burden of proof falls on the player, every out-of-form innings becomes extra pressure. What is needed is an open, timestamped record where what happened in which match is re-examinable. My most expensive spreadsheet was the 2026 one, where a single missing line among 1,200 events cost a match. Five years later, if someone in Delhi or Lahore wants a Bangladeshi bowler's death-over economy, they cannot find that line, because the fixture was never in any database. Data that is never recorded is never reused. In the next transfer window I will look at one thing, and it is not price — track record. A franchise that keeps its own database will make fewer category errors. Everyone at the table sees the same information. Not everyone keeps the same ledger.