The Auction's Blind Spot: The Middle-Overs Bowler the Market Still Cannot Price

On the final night of the last franchise auction I wrote two names side by side in my notebook. One was a death-overs specialist — a bidding war worth several crores, two franchises trading blows, and a deal that finished very close to a record. The other was a middle-overs spinner — base price, near silence, a single raised paddle. Same screen, same evening, a four-to-five-fold gap in price. And yet in my ledger sat 42 matches of phase-level data, in which the true contribution of these two roles draws almost the opposite picture. The first xG notebook taught me that a number can be a confession. Auction money makes its decisions in the mood of the room, in the chant, in the final-over highlight; the match's real accounting is written between overs seven and fifteen, where the camera rarely points. I have watched cricket for years, and for the last eight I have logged every ball by phase. This piece is one page of that ledger — one specific observation, one specific question, and an honest answer to it. Method first, because the greatest enemy of data is a vague definition. A T20 innings is 120 balls. I split it three ways: the powerplay, overs 1–6 (36 balls); the middle overs, 7–15 (54 balls); the death, 16–20 (30 balls). The split is not built for my convenience. There is a real reason behind it. The powerplay carries the 30-yard circle — only two fielders can be outside, so the ball travels to the bat. In the death overs the fielders spread, and the batter takes risk. In the middle overs there is no pressure from either end — the ball is squeezed into the middle, the field sits near the boundary, and the spinner goes to work. What stands out is that nearly 45 percent of an innings is bowled in the middle overs. It is the largest, most repetitive, most data-rich phase of the game. Yet almost the entire premium of a franchise auction is spent at the two ends — on the powerplay hitter and the death bowler. My method is simple but strict. I do not publish a claim without at least 15 matches of evidence — a rule learned from my 2026 xG notebook. For every bowler I calculate phase-adjusted economy, meaning his performance relative to the league's average run rate in that phase. I read the matchup index — left-hand versus right-hand, spin versus pace, the character of the pitch. And whatever the model cannot explain, I write down as well. I trust the baseline before I trust the breakthrough. Let me unpack the pricing. A death bowler is rarely given more than four overs in a match, and in many matches two of those land in the last two overs. Say he bowls 24 balls in an innings. Those 24 balls decide his reputation as a specialist. In a sample of 24 balls, one six shifts the economy by nearly a run an over; one boundary more than that. In the language of data science this is an extremely volatile, low-sample measurement — and it is precisely this measurement the market pays the most for. The tape explains the number; the number explains the tape. Watch the tape and you see that bowling at the death is largely a reaction to the batter taking risk. The batter knows the fielders are out, and he will try. So death economy is substantially a function of the opponent's decisions, not standalone proof of the bowler's skill. The bowler who concedes 12 in the last over and loses the match may well have kept it alive in the previous 34 balls — nobody remembers that. Now look at the middle overs. Fifty-four balls. That is, the same bowler bowls more than twice as many balls in this phase in the same match, if given a full quota. More balls means a bigger sample, and a bigger sample means more stability. One simple calculation from my ledger: the match-to-match variance of middle-overs economy is roughly 30 to 40 percent lower than that of the death overs. But this is not only about stability; it is about impact. A T20 match's story is often written between overs 7 and 15, when two sides squeeze each other. If the run rate drops below 7.5 here, the batting side must take extra risk in the last five overs. Risk means wickets. Wickets mean collapse. In my collected matches, wickets taken in the middle overs sit near 40 percent of the total — more than at the death, because the death fielding setup leans toward containing runs rather than taking wickets. Whose impact is this? Mainly the spinner's. In the middle overs the ball is old, seam movement is minimal, and the pitch offers grip. The way bowlers like Rashid Khan or Sunil Narine tie a batter down in this phase does not show up only in the wicket column — it shows up in the batter's strike rotation, in the pressure on the next man, in the collapse of the batting order in the last five overs. When a wicket is not a single-ball event but a chain of the next ten balls, its value cannot be captured by the wicket column alone. After the Impact Player rule came into the Indian Premier League in 2026, the middle-overs calculation became more important still. Where a team once had to give four overs to its fifth bowler or a part-timer, it now brings in an extra specialist. That means the density of quality in the middle overs has risen, and so has the cost of error. The side bowling well in this phase concedes 15 to 20 fewer runs across 20 overs — often the margin of the match. Try a test here. Take a team's scorecard and separate what it conceded in the last two overs from what it conceded between overs 7 and 15. In my 42-match sample, middle-overs run concession correlates more strongly with win or loss than the final overs do. In other words, those who build a team around the drama of the last over are building around the loudest but least reliable part of the match. Now the matchup index. To judge a middle-overs bowler you need to know who is batting against him. When a left-arm spinner bowls into the angle of a right-hander, the turn and the bounce work together. These matchups are visible in a single match, but across a six- or seven-match series they are what bends the course of an innings. Yet almost nobody in an auction looks at this subtle accounting, because the television screen shows only strike rate and economy. I am not denying the value of a death specialist of the Bumrah kind — that would be reckless. But the question is on how many balls his price is being set. The release-clause structure and the wage bill are the real story here: a franchise spends the bulk of its budget on the role with the smallest sample, and the least on the role with the largest. That structure of the budget is an opportunity for a sporting director, because a mispriced market is always somebody's gain. The picture clears further when we look at supply. Middle-overs spinners are plentiful — every domestic league, every subcontinental setup produces them. High supply, low price. But high supply does not mean low marginal impact. If a bowler delivering 36 to 48 balls in a match concedes one run per over less than the league average, that is four runs across four overs — the equivalent of saving eight runs in two death overs, but with far more reliable evidence behind it. There is another angle I have noticed lately. On high-scoring pitches the middle-overs role is changing. Where 200-plus totals are routine, giving a spinner four overs in the middle becomes riskier, and teams lean toward pace in that phase too. That shift has not yet been reflected in auction prices — the market still treats the middle overs as a spin-only role. Let me add a caveat, because this is a rule of my ledger. Empty stadiums gave football the control group it never wanted — and that post-Covid lesson taught me that without context a number is half a truth. The same holds in cricket: before reading any bowler's middle-overs record, you must know which stadium, what age of ball, and under what scoreboard pressure he bowled. Compare economies without separating these three variables and the exercise is nearly meaningless. So is the market wrong? The plain answer: no, the market is incomplete. The market is rational about what it sees, and blind to what it does not. Crowds rise in the final over, social media clips it, the franchise owner remembers it. If a data department wants to surface the value of the middle overs, it must convince the owner why a maiden in the 12th over is worth more than one in the 19th. That is hard work, because a maiden has no highlight package. Here the contrarian question arrives, and I apply it to myself. Suppose I claim the middle-overs bowler is undervalued. That could be true for two reasons — one, his impact really is greater; two, his impact is smaller but the market's inattention has made him cheap. The second reason is really a psychological market error, not a data error. The distinction matters, because if the second is true, then the more franchises enter this ledger, the more the price rises, and the advantage erases itself. One more possibility should stay open. Middle-overs economy may look low because the fielding setup in that phase is defensive — that is, not the bowler's skill but the tactic is producing the number. In that case calling the bowler undervalued would be wrong; the system itself keeps the role cheap by making it replaceable. In my sample I have not fully separated these two explanations, and that should be written down. A control group is just patience with a purpose — and right now I do not have it. I will also state clearly what could falsify my claim. First, if middle-overs wickets turn out to be an equal or smaller share than death wickets, my central assumption weakens. Second, if at high-scoring venues middle-overs run suppression becomes unrelated to winning, the role's importance falls. Third, if in small samples middle-overs economy turns out to be equally volatile, the stability argument collapses. If my claim fails any one of these three tests, I will change the number, not the story. Now a simple question that in my view should sit at the centre of auction strategy. If a team targets an average of 160 across 20 overs, which four overs give the most control? Four powerplay overs bring quick runs, but quick wickets too. Four death overs decide the result, but uncertainly. Four middle overs are the place where you can take the tempo of the match into your own hands — if you have the right man. The market still does not keep a separate line in the budget for him. In my view there are two things to watch in the next auction cycle. First, whether the price of middle-overs specialists begins to rise — if it does, the market is learning. Second, whether teams increase their use of pace in the middle overs, especially on flat pitches. The intersection of these two trends will tell us whether the market is actually learning from data, or merely imitating. One thing is worth remembering here. Cricket's data revolution has so far been built mainly around batting — strike rate, boundary percentage, phase scoring. Bowling analysis lags behind comparatively, because bowling success depends heavily on the batter's decisions. That asymmetry is the real cause of the market's blindness. A data department that brings only batting metrics will never price the middle-overs bowler correctly. A final thought. I am not blaming the auction hall. The hall is a market, and every market speaks its own language — here the language is the highlight reel. The franchise that can read past that language into the ledger will hold, over the next three seasons, an advantage nobody else can see, because nobody else wants to look. Every transfer rumor is a dataset waiting for a primary source — and the middle-overs bowler is still waiting for exactly that primary source. The number, in the end, throws up one question: if nearly 45 percent of an innings' balls, 40 percent of its wickets, and almost all of its match control are written under one role's name, why is that role priced last of all? The answer may not lie in the data but in the market's habit. And changing that will not take a season — it will take one sporting director willing to read the ledger.