The Price of a Dot Ball: Why Dhaka and Kolkata Auction Floors Value the Same Batter Differently

On the last ball of the 12th over at Mirpur the gallery erupted, but the ball never reached the boundary. The four had come two overs earlier, the six would come in the next over. In this over, five of six deliveries were dots. The scorecard recorded 5(6) — a line nobody remembers the next morning.

In my ledger, that over has a name.

Before it began, the batter's strike rate read 148.6, and that was exactly the number floating on the broadcast graphic. By the end of the over, a second number had moved — his middle-overs dot-ball rate drifted from 38.4 to 41.2 percent. One number rose, so did the other. Television picks only one.

Two questions follow. First — how good is this batter, really? Second — which number is the market that will buy him actually looking at?

Let the ledger breathe before the narrative does.

Across the 2026 and 2026 BPL seasons I hand-logged 2,314 dot balls. Not from scorecards — from the stream. Over number, bowler type, whether the batter was on strike, and the required run rate at that moment, all recorded separately. I stopped writing match reports and started writing this way in 2026, when I manually logged 1,214 shots from Bengaluru FC's I-League season.

Three metrics carry this piece, and all three definitions come first.

Dot-Ball Rate (DBR) — dots divided by balls in a fixed over band. I keep overs 1–6, 7–15 and 16–20 separate, because a dot ball is not priced the same in each.

Boundary Dependency Index (BDI) — the share of runs coming from fours and sixes. Above 70 percent, a batter's run tally is an event-driven number, not a durable skill measure.

Matchup-adjusted dots — an adjustment for how good the opposing bowler was in that band relative to tournament average. Dots against Rashid Khan in the seventh over and dots against a part-timer in the seventh over are not the same delivery.

Sample: two seasons, 89 matches, 2,314 dot balls. That is small, and I will state the limitations separately at the end.

Now, the two markets. Dhaka's BPL auction and the IPL auction in Kolkata and Mumbai price the same player differently, and the gap is bigger than scale alone. The most expensive Bangladeshi in IPL auction history remains Mustafizur Rahman — Sunrisers Hyderabad bought him for INR 2.2 crore in 2026, after INR 1.4 crore in his first season in 2026. Eight years on, that ceiling has not broken. Yet at the 2026 IPL auction, Kolkata Knight Riders took Litton Das at around INR 50 lakh, near base price.

Put those two facts side by side and a question surfaces: is the market pricing talent, or pricing a specific slice of talent?

The price sits in the dot ball, not the boundary

I count the silence between the dot balls.

In my ledger, a dot ball in overs 7–15 costs a team total 1.86 runs on average. In overs 1–6 that cost is 1.12; in overs 16–20 it is 2.41. Middle-overs dots are not the most expensive, but they are the most buried — powerplay dots and death-overs dots both get at least a mention.

Now the comparison that matters. Two batters, both with a strike rate of 145. The first has a DBR of 32 percent in overs 7–15; the second, 44 percent. Same strike rate, same league, same season. In my adjusted model the first batter's side finishes the innings 14.3 runs higher on average, and at a 95 percent confidence level that gap sits between 4.1 and 24.5 runs. The difference is invisible in strike rate.

Strike rate is a ratio; dot-ball rate is a cause. The ratio makes two players look identical. The cause separates them.

Why does the gap widen? Because a middle-overs dot ball does double damage. A ball is spent, and the batter must take a larger risk on the next one. At the death, a dot's cost shows up in the runs. In the middle, the cost shows up later — exactly when the finisher walks in.

Role-adjusted pricing: three roles, no more

Before I price by role, I hold one rule: no more than three custom roles in one analysis, and every definition written down before I look at outcomes. Otherwise any cheap player can be slotted into a fresh role and declared a market error.

Three roles here.

Anchor-rotator — DBR under 30 percent in overs 7–15, BDI under 55 percent, strike rate above 130. This profile holds an innings together.

Powerplay specialist — strike rate above 140 in overs 1–6, but DBR above 40 percent in overs 7–15. This profile starts innings and does not finish them.

Death-overs finisher — above 1.65 runs per ball in overs 16–20, but facing fewer than 18 balls per innings on average.

In Bangladesh's domestic circuit the three profiles are not priced alike. Powerplay specialists fetch the most at auction, because their highlights are the easiest to cut. Anchor-rotators fetch the least, because their work is invisible on a scorecard. In my tracking, several of the batters with the highest middle-overs dot-ball rates are powerplay specialists whose BPL auction price climbs, while at least one anchor-rotator goes cheap despite his side finishing with a higher expected total when he is in it.

That is the mispricing. The market pays for a slice of talent, and that slice is the one standing in front of the camera.

The IPL market asks a different question. Nobody there asks how many runs you made; they ask in which over, against whom, in which matchup. Mustafizur Rahman's INR 2.2 crore is not for his cutter — it is for his death-overs matchup specialist profile. Varun Chakravarthy's market was built on exactly that logic: matchup, not score. Bangladesh's domestic auction still reads largely off the score.

Is the dot ball the cause, or the symptom?

Now to break my own argument.

The relationship between DBR and team total runs both ways. It is true that a batter who plays dots sees his side score less — but the reverse is also true: when the required rate climbs, batters take risk, and risk produces dots. My adjusted model tries to control for required rate, but control is never complete.

The second gap is matchup. Across overs 7 to 15 a batter faces roughly three bowlers, and their quality varies enormously by side. Across 89 matches in two seasons that variance is wide enough that my confidence interval on any single batter's DBR spans six to eight percentage points.

The third is a warning aimed at myself. Practising noise-adjusted scepticism daily breeds a habit — becoming the man who always says "actually" and corrects the room. So I publish my own override base rate: of my last 31 overrides, 17 were right and 14 were wrong. The record matters more than the magnitude.

The largest trap sits inside my own trade: keep inventing finer roles and eventually every cheap player looks like a hidden asset. If the arbitrage has not closed by the time the auction ends, the role was not a market error — the role was my own fabrication.

Limitations

Two seasons, 89 matches, 2,314 dot balls. I have no ball-tracking data, so line, length and field settings could not be measured separately. The adjusted model's confidence interval is wide, and no single match supports a conclusion. The stadium was empty; the numbers were not — but the numbers do not say everything either.

What I will watch next

Ahead of the next BPL auction I am publishing a prediction in advance, with a timestamp: the franchise that buys at least one anchor-rotator with a middle-overs DBR under 32 percent will, next season, score more than tournament average in the last five overs of its innings.

Published before the auction. After it, I will check the numbers — win or lose.

The question remains open. Does the market buying highlights not know what a dot ball costs — or does it know, and simply lack the instrument to price it?