After the Fourth Over: Fast-Bowling Workload, the Broadcast Calendar and Cricket's Invisible Bill

Last February I sat in the second tier of the Gabba in Brisbane for a night T20, a small notebook in my hand. The people around me jumped at every six; I was writing down only one thing, a speed log. One fast bowler's first over averaged 146 kilometres per hour. The second, 143.4. The third, 139.1. And in the fourth over, the one that carries the real weight of the innings, the one where economy rate decides the game, his average dropped to 132.6. Thirteen and a half kilometres of pace lost across four overs. The scorecard never records that. It records 2 for 28, an economy of 7.00. The data the scorecard cannot hold is often the data that matters most. The numbers were never the story; they were the trailhead.

I write this as a betting analyst who has watched cricket for twenty-six years, but for the last eight or nine a different ledger keeps running in my head while a match is on. We talk endlessly about how much money, how many cameras, how many broadcast hours cricket spends. Nobody opens the other account: how many overs, how many thousands of kilometres of travel, how many pairs of shoulders. This piece tries to open that account.

Some context is necessary, because fast-bowling workload is both cricket's oldest debate and its least measured one. Since 2026 the international calendar has taken a fixed shape. On one side, the 2026 ODI World Cup in India packed 48 matches into 46 days. On the other, franchise leagues now open two or three windows a year, each running three to eight weeks, alongside bilateral series, trophy tours and warm-up fixtures. In that structure, 50 to 70 days of competitive bowling a year is completely normal for a frontline fast bowler. Some reach 90.

I started with economy rate, but I had to finish with shoulders. The most visible fact is this: at an IPL auction, a fast bowler now costs what 150 kilometres per hour costs. At the 2026 IPL auction, Kolkata Knight Riders bought Mitchell Starc for INR 24.75 crore, then a record price in IPL history. At the same auction, Pat Cummins went to Sunrisers Hyderabad for INR 20.5 crore. A market that pays that for a shoulder never writes a line about who funds that shoulder's rest. The cost hides elsewhere: in national team press conferences, on domestic team sheets, and in the ticket price of the fan who came to see that bowler and was told he was being 'rotated'.

The first real accounting follows from this: a fast bowler's cost splits three ways, and only one party ever feels the whole bill. The three are the board, the franchise and the bowler. The board takes broadcast money, the franchise takes gate and merchandise money, the bowler takes the few overs in which his entire career sits. Over five seasons I have tracked three layers: rest days between matches, over-by-over average release speed, and the ratio of bouncers and yorkers per innings. Put together, a pattern keeps returning.

The first signal is calendrical. Average rest days between matches, for fast bowlers only, have fallen by roughly a third compared with 2026. What used to be six or seven days is now three to five. That figure combines international and franchise cricket. Add time-zone shifts and the picture worsens, because bowling in a new continent three days after a match means broken sleep, and the link between sleep debt and muscle recovery is close to settled in sports science.

The second signal is my favourite, because it happens inside the game, in plain sight. Inside a T20 spell, a fast bowler's release speed follows a defined curve, and the curve bottoms out in the very over where his value peaks. Last season I logged over-by-over speeds across twenty-seven T20 spells, from stadium and broadcast sources alike. The average picture: if the first over is indexed at 100, the second reads 98.2, the third 95.1, the fourth 91.8. Roughly an eight per cent drop across a spell. That sounds small, but at yorker and bouncer height, eight per cent is often the difference between a yorker and a full toss, and we all know what a full toss costs in T20.

One clarification matters, or the accounting stops in the wrong place. A speed drop does not automatically mean a bowler is 'tired'. The same bowler in a Test often holds pace into his fourth or fifth over of a spell, which shows the difference is mental as much as physical. In T20 a bowler must change plan three times in four overs, reset nine fielders ball by ball, and hold the opposition batter's swing pattern in his head. I have watched that mental load show up directly in physical pace, especially in spells that begin after the powerplay.

The third signal is tactical. I track the ratio of bouncers to yorkers per innings, because those two deliveries load the shoulder most. A bouncer is a full spinal extension; a yorker is maximum speed with total ankle and knee control. Over the last two years, yorker usage in the final five overs of IPL games has risen sharply, because models say yorkers concede least at the death. Every extra yorker is one more delivery the bowler cannot bowl within a safe shape. Here board and franchise interests point the same way, and the bowler's shoulder points the other way.

This is where I want to pause, because the most comfortable myth hides here. We blame franchise cricket for endless overs. Measure it, and a fast bowler can bowl 30 to 40 overs in a Test innings, more than his entire four-match T20 week. The number is not small. What differs is not the number of overs but the type of over, the pace demand, and the rest between matches. A Test bowler gets days after a Test. A league bowler gets another match in two days, and every ball feels like the last ball.

I know plenty of readers will say this is old news, players play because they are paid. True, and I am not denying it. My job is to ask why we always do the money accounting from one direction only. As a betting analyst I see something odd daily. The franchise that pays over INR 20 crore for a bowler makes no plan for resting him when he is on national duty. The board signs huge broadcast deals but has limited power to cut the number of matches in them. The franchise's only duty is to the owner and the fan. The bowler's only duty is to his own career. Four people, two payers, one receiver, and the receiver is told to be professional.

The real cost sits in broadcast time, not in delivery count. Broadcast revenue has multiplied over a decade. In the same decade, average fast-bowler career length has not grown. It has shrunk. Read those two numbers side by side and a question appears I still cannot answer: if audiences, broadcast money and stadium light have all grown, why has career life shortened? The likely answer is that the career belongs to the one party controlling none of those three.

Yet mapping my own data directly would be a mistake, and I made it myself. Looking back through three years of notes, not every spell I flagged as a pace decline was T20 or ODI. Some were Tests, some domestic, some practice games. Blaming one format or one league is easy and probably wrong. The harder job: every format has its own fatigue curve, and each must be measured differently. Tests need overs and days; ODIs need spell length; T20 needs the gap between overs. Mixing them produces a comfortable story, not a truth.

So the second layer of my model, published here, is narrower. I track four pillars: rest days, over-by-over pace decline, delivery-type load per innings, and travel between matches. Last season, holding only these four, the severity of decline did not rise in a straight line with match count. It rose with travel time between matches and with time-zone change. That is a new signal in my ledger, and it rarely appears in mainstream debate. A bowler's fatigue may not be built in the overs. It may be built at the airport.

The model also warns me. Take two fast bowlers with near-identical workload: same overs, same rest days. One breaks down; the other plays the season out. Workload cannot explain that gap. You need delivery type, action, foot position. Workload is a tendency, not a verdict. Without separating correlation from causation, analysis becomes a sellable story, and stories do not prevent injuries. Anyone saying on camera that the calendar is the sole cause of an injury is offering a position, not data.

That raises the contrarian question I put to myself: if we insist on a calendar crisis, should we measure intensity per minute of high-effort bowling rather than per match? In T20 a bowler sends down four overs, but every ball is close to maximum effort. Those minutes should be the core index. In my count, the number of high-intensity deliveries in a four-over T20 spell often matches or exceeds a fourteen-over Test spell. Measured that way, the problem may not be the number of leagues but the number of matches inside each league: scheduling, where teams decide who plays, while boards and owners decide who rests.

One cost that gets almost no airtime is replacement. When a fast bowler breaks down, who comes in? Usually a domestic bowler. He steps into international intensity, on unfamiliar pitches, often in the very match where the first-choice bowler is resting. In one season that may be a huge break for two or three players; as a rule it stops being an opportunity and becomes an unbooked cost. Teams rise and fall on the shoulders of bowlers who were not built into the ticket price, but assembled on an hour's notice. I call this cricket's silent subsidy.

The fan pays part of it too. I have sat in stands and heard the line: 'I bought a ticket to watch Starc, and now he's rested.' There is an economic truth in that sentence. When a fan buys a ticket, he is buying a specific spell, not a fixture. If the spell never arrives, it is an incomplete product with no refund. Owners know this, boards know this, and nobody prices that risk into the ticket. That is the community ledger I add to every preview: who pays, who is served, who carries the real cost.

Now the question at the centre of all this. Suppose we hold every data point together. Rest days falling. Pace decline measurable. Yorker usage rising at the death. Auction prices at records. Broadcast revenue climbing. Yet no single index answers my model's final question: why does the pace drop never show a high correlation with match outcome in my books? One reason. When we hold only outcome data, we think in averages. Every cricket data point belongs to a specific moment, a specific bowler, a specific over. Lose that specificity and analysis becomes broadcast decoration.

My job is to ask why. Why was this bowler brought back for a fourth over when his pace was already on the downslope? Why do teams use the word 'rotation' while the decision is taken in an owner's room, not in front of the bowler's medical panel? Why does nobody negotiate in writing with a broadcaster to add one extra rest day between matches? Those questions make poor headlines. But answer one of them, say whether moving from three to five rest days adds one per cent to fourth-over pace, and it matters more than a billion-dollar broadcast deal.

This is the end, but stopping here would be wrong, because I began in a Brisbane stand and I finish with one plain prediction. For the rest of this season I will watch one thing closely, still small in my ledger but growing: fast bowlers are starting to think in overs rather than matches. The plan becomes 'I bowl three overs today', not a full spell, because the data says every ball after the third over in a spell is the most expensive and the weakest. If that trend holds, in two years we will see bowling attacks where spells get shorter and spell quality rises. That day the scorecard will not say it, but heads will turn toward the airport, the night lights, and that stand where someone sat with a small notebook. In the next series I want to know whether more rest days lift fourth-over pace. Open your own notebook and run the count. It costs little, and the data is the most honest thing in the ground.