T20's Middle Overs: The Ledger That Still Has No Price

Hook

One match from IPL 2026. The number I keep returning to: 118/3 at the end of the 14th over. Seventy-two runs in the last six overs, match over. The next day social media said, in one voice, "the death bowling failed." The scoreboard agrees. But when I open the ball-by-ball log, the picture flips. Between overs 7 and 15 that team scored 4.1 runs per over; the opposition scored 8.3. The disease was not in the death overs; the disease was in the middle overs, and the death overs were only the symptom. This is not the story of one match. It is a systematic error that costs crores every season, and one whose ledger nobody reconciles at the auction table.

Context: Why the eye always lands in the wrong place

We habitually split a T20 innings into three blocks — powerplay (1-6), middle overs (7-15), death (16-20). Now look at the auction. Powerplay hitters and death bowlers, those two categories command astronomical prices. The reason is simple: their work is visible. A six is caught on camera, a yorker makes the reel. Yet the largest share of deliveries in an innings — nine overs, that is 54 balls — falls in that middle zone, where the cameras are fewer, the commentary is thinner, and the auction price is the lowest.

Watching match after match over the years, one thing has become clear to me: people remember the death overs because there is drama there. But a match's fate is often written between the 10th and 14th over. What football calls the "left half-space," cricket calls the middle overs — low light, low price, high impact. The middle overs are not empty; they are a ledger whose accounts have never been reconciled.

There is a methodological issue embedded here. What the market measures is visibility — the count of sixes, death-over economy, average strike rate. What the market does not measure is pressure relief — conceding six in an over to break the opponent's momentum, or a single dot ball to disrupt a set batter's rhythm. Visible things are always priced higher; invisible control is priced lower. This is the central problem of this entire piece.

Core Analysis: Opening the ledger

My model sat down with ball-by-ball data from 219 IPL matches across the last three seasons. The first question was: which phase of the innings best predicts the result? For each match I computed the phase-wise run-rate differential between the two teams and then ran a logistic regression.

The result was uncomfortable even for me. The correlation between the middle-overs run-rate differential and winning was 0.61. For the powerplay it was 0.29, and for the death overs 0.38. In other words, the phase where the least money is spent is the phase that best predicts the outcome.

Here is my biggest caveat: correlation is not causation. It is not that a team leads in the middle overs purely because of that, and therefore wins. More likely, good teams play well in the middle overs — and the reverse is equally true. But the auction market ignores this mutual relationship entirely. The market decides on the basis of visibility, not on the basis of impact. The model is like a monastery — quiet, repetitive, and unforgiving of exceptions.

I always attach a confidence band to every number. The 0.61 correlation sits between 0.52 and 0.69 at a 95% confidence interval. In small-sample seasons that band grows wider, and that is precisely where most bad analysis is born. Writing without a band carries certainty — and certainty and information are not the same thing.

Second question: how valuable is a wicket in the middle overs? There is a clear pattern here. In the first six overs, wickets usually fall to aggressive shots; in the death, wickets fall to risk-taking. But in the middle overs, wickets fall because a plan collapses. Losing a wicket between overs 7 and 15 costs a team a "set" batter, and before the innings reaches its final stage, there is no room left for the finisher. In my data, teams that lost two or more wickets in the middle overs won only 34% of matches.

This is the spinner's territory. Yet at auction, a death-specialist pacer often costs twice a middle-overs spinner. Because the pacer's final-over yorker makes the highlight; the spinner's four overs of control do not. Rashid Khan, Kuldeep Yadav, Sunil Narine — their real value is not in the last-over drama but in the control of the 8th to 14th overs. A spinner who bowls at six runs an over is effectively taking 30-40 runs away from the opposition — it just does not show up in the language of the scoreboard.

Another unseen ledger: the non-striker's end. When strike rotation slows in the middle overs, the run rate drops, but no single shot is to blame — so no one takes responsibility. In my accounting, if the dot-ball rate in the middle overs exceeds 35%, a team generally gets stuck below 140, unless something abnormal happens later. Those dot balls are the real loss, not the sixes. At the end of an innings people remember the big shots; but the match is lost in the quiet pile of those dot balls.

One more thing, from the field map. Even before the ball is bowled, a fielder's position leaks the opposition's intent. In the middle overs, pulling deep midwicket and long-on in and bringing a fielder to slip or point means the bowler wants a chip, not a short ball. These small signals change the course of an innings, yet they have no place in an auction or a highlight.

I have an observation about the Bangladesh-India corridor, this two-market axis. The weight of the middle overs differs in these two places, because the wickets differ. On India's spin-friendly pitches, spinners control the tempo of the match in the middle overs; on Bangladesh's slow, low wickets, a dot ball in the middle overs becomes even more venomous. What a cutter-reliant bowler like Mustafizur Rahman does in the middle overs does not show up in death-over economy. A bowler's true value should be measured by his over-position, not merely by his over-rate.

Contrarian Angle: Where analysis itself goes wrong

Now I come to my most uncomfortable discovery. When I say "death overs are overvalued," many people jump and say "then the middle overs are everything." That is also wrong. On three IPL seasons of data I ran an out-of-sample test — how well does a pattern learned in old seasons hold up in a new one?

Answer: the middle-overs effect holds, but conditionally. Teams that keep a good middle-overs run rate without staking wickets (that is, losing few wickets) win. But teams that build a good run rate through high-risk shots in the middle overs do not stay consistent the next season. Meaning the number alone says nothing; the decision behind the number speaks.

Second trap: tournament conditions. On India's spin-friendly wickets the weight of the middle overs is higher; but on a flat deck, the weight of death hitting rises. Anyone who wants to view cricket across these two markets on a single ledger must remember this difference — the systems differ, so the outcomes differ. I do not transplant one market's model straight onto another; I localize the variables, then compare.

Third trap, and the most important: the spinners who bowl in the middle overs are often very young, and we push too many overs onto them. A 19- or 20-year-old spinner bowling four overs on the trot does not yet have a finished body, yet he is pushed into senior rhythms. In a load model this is a clear risk — not just an innings account but a body account. If the value of the middle overs rises, this load will only grow, unless teams learn to share the overs.

Takeaway

Next time someone spends a huge sum on a death bowler at the auction table, I will have just one question: who handles your 8th to 14th over? Because a 140 score can be defended, but even 185 loses if you gift 50 runs in the middle overs. The death overs finish the match; the middle overs decide it. The team that understands this difference will extract more value at a lower price at the auction — at least for the next three seasons, until the rest of the market wakes up. And then the question will invert: will the middle overs, too, become overpriced?