Three Seasons of Evidence: What IPL Auctions Keep Getting Wrong About Fast Bowlers

On a May night in 2026, Mumbai Indians defended 129 in Hyderabad to win the IPL final by one run. For the next fortnight I re-watched Mitchell Johnson's final four overs thirty-one times, drawing each field setting on graph paper. At sixteen I didn't know what I was looking for. I know now—I was watching to see where a bowler's price is set. Not in the auction. In the last four overs. That question eventually pushed me into three seasons of auction data.

The auction stage always tells the same story: fast bowlers are expensive, so teams buy them. But when I matched the money spent on fast bowlers across the 2026 to 2026 auctions against their actual death-over performance, the pattern inverted. More than half the fast bowlers who attracted the highest death-over premiums recorded an economy above 10 at the 20th over in the following season. This is not statistical accident. It is a structural flaw in how auctions work.

To understand it, you need to grasp what an IPL auction actually measures. It is not an evaluation process—it is a bidding process. No team calculates a bowler's "true" price; it calculates what nine other teams will bid. Auction prices reflect expectations, not performance. And expectations are built on old data.

The 2026 auction offers a clear case. A established fast bowler went for ₹10.75 crore on the back of three seasons of death-over economy between 9.2 and 9.8. Those are good numbers, but his best death-over season had come when two of his team's main bowlers were injured and he carried the pressure alone. The following season, given a clear role—the 17th and 19th overs only—his economy fell to 8.1 but his wicket count halved. The team had bought him as a death specialist; he was actually a bowler equally effective with new and old ball, provided he got consecutive overs.

Here my first contradiction begins. Teams treat death-over expertise as a stable attribute, when it is actually a context-dependent outcome. A bowler's performance at the 20th over depends on who bowled before him, what state the batter was in, and how the field was set. Bring him to a new team and all three change—but his price does not.

By the 2026 auction the pattern was clearer. One team spent ₹22 crore on two fast bowlers, both primary death bowlers for different teams the previous season. By season's end neither had recorded an economy below 9.5 at the 20th over. The cause was structural, not technical: their previous success rested on an experienced spinner building pressure in the 16th to 18th overs. In the new team that spinner did not exist.

Here an old habit returns. At sixteen I mapped the final four overs of an IPL final onto graph paper. Then I thought I was reading bowling plans. Later I understood I was watching how a one-metre shift in fielders changes a bowler's options. In that final, Krunal Pandya's 47 off 38 forced Rohit Sharma to move his ring fielders a metre at a time. The bowler was the same, the ball was the same, but the changing field transformed his option set. The same applies to auctions: the bowler is the same, but changing teams changes his "field"—and teams do not buy that field.

By 2026 I saw something I had missed. Post-auction analysis almost everywhere described fast-bowler prices as "irrational." But when I laid three seasons' data side by side, the irrationality became a pattern. For bowlers operating above 140 kph, auction price is determined not by wicket count but by how dangerous they look. This is an aesthetic valuation, not a performance-based one.

One direct consequence: fast-bowler prices rise while their roles remain undefined. A team buys a bowler for ₹15 crore, then bowls him in the 7th over because "he likes the new ball." Next match he bowls the 19th because "he can do death." Third match he is benched because "he has lost rhythm." That is not the bowler's problem; it is the team's planning problem.

An older line returns here—one I first wrote after logging sixty World Cup matches in a notebook: "Every transfer rumour is a map with one road missing." The same applies to auctions. Every purchase is a map missing one road—the road that made the bowler successful at his previous team. Teams do not buy that road. They buy the bowler.

Now my second contradiction, and perhaps the least discussed. Retention and the Right to Match card have turned the auction into a process where the best bowlers often go for the lowest bids. When a team retains three players, they leave the auction pool. The number of available fast bowlers shrinks, and those remaining inflate artificially. This is a market shortage, not a reflection of quality.

In 2026 I watched a specific case. A team paid ₹8.5 crore for a fast bowler who had played only 14 matches in two previous seasons. Why? Because death-bowling options were scarce that season and three teams were chasing the same profile. He played nine matches and conceded at 11.4. That was not his failure; it was the auction's.

The pattern is not confined to the IPL. When I examined domestic cricket—Ranji Trophy and Syed Mushtaq Ali Trophy—the same structure appeared. A domestic fast bowler bowls the 20th over for his team, but in the IPL he bowls the 16th because the team has no other experienced option. That is a role transformation the auction price does not measure.

I want to draw a clear distinction. Auction prices are not wrong—they measure a different thing. They measure availability, expectation and market scarcity. The problem is that teams mistake those prices for performance forecasts. When they do, they convert their budget into an asset they cannot deploy correctly.

There is another dimension I find under-discussed. Fast bowlers' auction prices should be discounted for injury dependence, but they are not. Across three seasons, roughly 40 per cent of fast bowlers who fetched more than ₹10 crore played fewer than 10 matches the following season. Injury is a structural risk for fast bowlers—but auctions do not price it. A team paying ₹12 crore for a fast bowler is effectively expecting 10 to 12 matches for that money, but never says so.

An old line applies: "I trust the pattern until the pattern starts explaining itself too neatly." Three seasons of data show a pattern, but it leaves a larger question open: is this a flaw in auction structure, or in cricket's habit of compressing a bowler's performance into a single statistic? My leaning is toward the former, but I do not dismiss the latter.

Now the part that makes me uneasy. Over three seasons I built a database of sixty matches—toss, dew arrival, powerplay run rate, middle-over wickets, chase outcome. When I matched it to auction data, one thing became clear: teams that define their fast bowlers' roles explicitly get better results with lower auction spend. The pattern was clearest for teams that assigned fixed over blocks—one bowler 1-4, another 5-8, another 16-20.

But there is a counter-argument I cannot ignore. Some analysts argue death-over expertise is a myth—that a good bowler bowls well in any over. That is not entirely wrong. Pressure in the 20th over is different, but it is situational, not a fixed attribute. A bowler with a good yorker will land it in the 20th over too. The question is whether the batter expects it. That is a psychological game, and it cannot be measured at auction.

So are teams making a mistake? I would say they are asking the wrong question. They ask "how good is this bowler?" when they should ask "in what role will this bowler be best for my team, and is that role the most efficient purchase in today's auction?" The second question is harder because its answer is not on the auction floor—it comes from your net sessions, your squad structure, your bowling plan.

One more thing is missing from this debate. Auction prices are a public ritual, and public rituals always distort private information. When a team pays ₹14 crore for a bowler, it is not just buying a bowler—it is signalling to fans, board and sponsors that it is "serious." That signal is a cost, and sometimes it exceeds the real cricketing cost.

I have seen this pattern elsewhere—in empty stadiums. When cricket returned in 2026 to empty stands, I built a log of sixty matches. One thing was clear: without a crowd, a bowler's real value showed not in wickets but in consistency. The crowd turns a bowler into a hero; the empty stadium turns him into a data point. Auction prices are still built for the crowd.

I want to leave one final concern, framed as a question. If auction prices are structurally distorted, and teams know it, why do they not change? One answer is that they want to, but a structural reason stops them—strategists and bidders are often different people. When you sit at a table bidding ₹10 crore, you are not thinking about the role you defined two months ago. You are thinking about which team you are beating.

So the most useful question is not who went for how much. It is this: after the auction, over the next three weeks, do teams define that bowler's role the way they had defined it in their own minds before the auction? In my three seasons of data, the answer was almost always no.

Another auction is coming. Teams will outbid each other, and fast-bowler prices will rise again. If you look at those prices and think some team knows something you do not, think again. They know what you know—they are just measuring it in a different direction. The real question is whether any of them will deploy it in the right one.