On-Chain Price vs Middle-Over Dot Balls: The Gap Between Hype and Process in Cricket Fan Tokens

One night last franchise season I had two tabs open on my laptop. One was a live pressure dashboard — middle-over dot-ball force rate, boundary efficiency, chase elasticity, refreshing every over. The other was the on-chain price chart of that franchise's fan token. Within ninety minutes of the match ending, the token climbed about fourteen percent. My process score did not move a single point. The two biggest reasons for that win were two full-tosses in the last two overs and a misfield in the deep — outcome, not process.

Something measurable had split open between what the token was celebrating and what my model was measuring. Put in spreadsheet language: the market was pricing outcomes, I was pricing process.

I work as a transfer market administrator. A large part of the job is finding the lag between price and performance. In 2026 I modelled Benfica's Enzo Fernández at eighteen million euros before the Qatar World Cup. After his Young Player award, Chelsea paid one hundred and twenty-one million. That lag was roughly six months long. The cricket fan token market runs the same machine, compressed from six months into ninety minutes. In T20 the compression is sharper still, because single-match outcome variance is far higher than in football.

Context: which door did blockchain walk through

A fan token is a blockchain-based asset issued in a club's or franchise's name. It has two layers. The first is utility — holders vote on some club decisions, get access to events, signed shirts, meet-and-greets. The second is speculation — the token itself is tradeable and its price moves. The Chiliz ecosystem and several other platforms operate on this model, and in cricket another large blockchain door is digital collectibles and ticketing, where FanCraze's ICC-linked drops were among the most discussed examples.

My interest is not in the votes or the shirts. It is in the second layer, in the pricing mechanism. What is a fan token to a transfer market administrator? It answers an interesting question: what does a community believe about a player's or a team's future, and how fast is that belief converted into price? I restate the question like this — does the market price process, or does it price events?

A question like that has to be falsifiable, otherwise the data is just decoration. So I worked with a bounded dataset — seven franchise cricket seasons, ball-by-ball tagged, across ILT20, SA20, the BPL and the PSL, more than eight hundred innings. I tagged it myself, meaning every dot ball, every false shot, every over's field setting sits in a separate field in my database. I follow a personal rule: the three-source verification rule — a number that does not reconcile across three independent sources does not go into print. And there is a deadline rule, aimed at my own temperament: publish before midnight instead of polishing a model forever.

I treat cricket as a harsher test bed than football. Innings split into phases, separate economies for the powerplay and the death overs, dew, the Impact Player rule, no draws — together they make the luck component of a single match much larger. In football ninety minutes lets process win slowly; in T20, a hundred and twenty balls can bury process under an outcome. That is precisely why a sentiment-driven pricing machine like a fan token should lean further in the wrong direction in cricket.

Core analysis: process score versus on-chain price

I work with four variables.

Middle-Over Dot Pressure (MODP) — the percentage of dot balls a bowling side forces per over between the seventh and fifteenth, weighted by whether the dot was enforced by the delivery or produced by the batter's error. Counting dots alone is meaningless: one dot ball can be reverse swing, another can be a batter losing his footing.

Boundary Efficiency (BE) — runs produced per boundary ball; a more stable gauge than strike rate, because strike rate is easily inflated by dot-ball count.

Chase Elasticity (CE) — runs added above par for every wicket lost between the sixteenth and twentieth overs.

Process Score (PS) — a weighted composite of the three, normalised onto a zero-to-one-hundred scale.

What my tagged database shows: teams with a process score above 62 won 71 percent of their next five matches. Teams below 48 won 29 percent. The sample is small, so I never call this a prediction — I call it a probability band.

Now the real question. Which of the two does token price correlate with more closely?

In my calculation, the correlation between match outcome and daily token price movement is about 0.61. The correlation with process score change is only 0.14. That gap is not small. The market prices outcomes and barely prices process. In transfer market language, that is a visible, systematic mispricing.