$33,981 hit Toronto at 46c — 96% of one market's flow on one side

What we found

In today's whale-flow tape we logged 562 qualifying events across the board, totaling $1,049,166 in tracked dollars. One market accounted for an outsized share of the structure we care about: Boston vs Toronto.

$33,981 landed on Toronto at an average price of 46 cents. The number that matters is not the dollar figure — it is the split. 96% of the tracked flow in that market went to one side.

That asymmetry is the whole finding. Dollar volume alone is close to meaningless as a signal; big two-way markets can print large totals while remaining perfectly balanced, and balanced flow tells you nothing about information. A one-sided book is different. When 96 cents of every tracked dollar arrives on the same side of the same market, the flow is either coordinated, correlated, or informed — and all three are worth noting, though only one of them is worth paying for.

The average fill of 46 cents adds a second detail. The money went in at a price the market had below even. Whoever wrote those tickets was taking the side the book had priced as less likely, not piling onto a consensus. Directionally, that is the shape of a disagreement trade rather than a momentum trade.

What it means

Less than most people will want it to.

This is one market, on one day, inside a sample of 562 events. We are reporting a market-structure observation, not a projection — we have not claimed Toronto will win, and we are not going to. Our engine prices games; this note describes order flow. Those are separate objects and we keep them separate on purpose.

The honest framing of concentration data: it tells you where conviction sits, not whether that conviction is correct. Sharp money is wrong constantly. The reason we track flow at all is that the *distribution* of one-sided markets, measured over hundreds of them, has historically carried some information about closing-line movement. A single 96% reading is one draw from that distribution. It is not evidence on its own.

There is also a measurement caveat we would rather state than bury. Our tape captures the events we can observe, which is not the same as all money in the market. A 96% concentration figure is 96% of what we logged. If our capture is biased — toward a book, a bet-size band, a time window — the number inherits that bias. We have not yet published a study quantifying that, and until we do, treat concentration as a directional indicator, not a measurement.

What we would do with this internally: flag the market, watch whether the price moves toward 46 cents or away from it, and record the outcome for the flow-vs-closing-line study we are building. What we would not do is size a position on one reading.

What would change our mind: a large-sample study showing one-sided flow above ~90% fails to beat the closing line — at which point concentration becomes noise we should stop reporting.

21+ where legal.