Late money: the money that arrives after your ticket is in

§1The price is decided after you have committed

When you stake into a pool, you commit an amount. You do not acquire a price. The dividend on your selection will be struck later, from a ledger that is still being written, by people who owe you nothing and do not know you exist. Everything on this page follows from that sentence.

The uncomfortable half of it is easy to state: if money arrives on your selection after you stake, the dividend you receive is lower than the number you saw when you staked. That is not a malfunction, a penalty or a mistake. It is the mechanism operating exactly as designed, and the same mechanism is what lets a pool move to a price no bookmaker would have needed to publish.

This is the point at which pool betting has least in common with the other two structures. A matched order on an exchange is a completed trade at a price; a fixed-odds bet is a contract at a price. A pool ticket is neither. It is a share in whatever the pot turns out to be.

§2The same 40,000, arriving two different ways

The clearest way to see dilution is to take one pool and move the same amount of late money to two different places. The pool starts at 250,000 gross with 46,700 on the selection in question and an illustrative take of 17.5 per cent, so the dividend would be 4.42. Now 40,000 arrives before the pool closes.

The pool grew by the same amount in both cases, the take is unchanged, and the net pool to be paid out is identical at 239,250. The only difference is where the money landed, and it changes the dividend from 2.76 to 5.12 — a factor of nearly two, in opposite directions, from the same 40,000.

For a bettor holding a ticket, that is the whole risk of a pool in one comparison: not that the pool might be small, but that the pool's shape is outside their control and is decided last.

§3Why the asymmetry is so steep

It is tempting to think of dilution as symmetric — that money on your selection hurts you and money elsewhere helps you, roughly equally. It is not, and the reason is in the formula rather than in market sentiment.

Money arriving on other selections adds to the numerator of the dividend and leaves the denominator alone. Money arriving on your own selection adds to both: it enlarges the net pool, but it also joins the group of tickets the net pool is divided by. Since the money on a single selection is normally a small fraction of the whole pool, the second effect dominates the first.

That ratio moves with the distribution, and it moves in the direction that makes a popular selection the worse place to be diluted. If the selection already carries a third of the pool, the money outside it is twice the money inside, so the ratio is around two to one. If it carries two per cent of the pool, the ratio is around forty-nine to one: the ticket is almost entirely at the mercy of late money arriving on the same selection.

None of this is an argument for or against any selection. It is a description of an exposure that is unique to pools, and that a bettor cannot hedge by checking a board more often. Board watching shows the exposure; it does not remove it.

§4Self-dilution: your own stake is part of the winning money

The stake you place lands on your own selection, so it is in the denominator of your own dividend. This is the pool equivalent of a market impact, and it exists in a bookmaker account exactly nowhere: putting 25,000 on a fixed-odds selection changes your potential return by nothing at all.

The table below takes the same baseline — 250,000 gross, 46,700 on the selection, an illustrative 17.5 per cent take — and adds a single stake to it, changing nothing else.

The pattern is the one that makes pool betting awkward for anyone staking in size: the cost of your own money rises as you commit more of it, and it is charged to your own ticket, not spread across the pool. A stake of a hundred units against a quarter of a million is invisible. A stake of a hundred thousand against the same pool removes more than half the dividend it was aiming at.

The practical translation is not a staking rule but a sizing fact: in a pool, the relevant question before staking is not only how much the selection is worth, but how much of the pool the stake itself represents. On the page on tickets and units the same arithmetic is applied to a ticket made of many combinations, where it bites harder.

§5Thin pools, where the distortion is largest

The dilution effect scales inversely with the money on the selection, so it is at its worst where pools are smallest: a novice race, a minor meeting, an obscure league, a market with a low turnover, or an exotic pool that has not attracted money. An illustrative pool of 8,000 with an illustrative 20 per cent take makes the point.

Your own 1,000 moved the dividend from 3.73 to 2.56 — a reduction of nearly a third — because in a thin pool your stake is a large share of the money on your own selection. Nothing about the event changed. No other bettor needed to disagree with you. The stake diluted itself.

This is also why a pool dividend is not comparable across pools without knowing the money on the winning selection. The same event, priced by two pools of different sizes, can pay wildly different amounts simply because one of them is thin enough for a single stake to move it.

§6What can and cannot be done about dilution

Almost nothing about dilution is within a bettor's control, and it is worth being blunt about which parts are.

What remains true at the end of all of it is the sentence at the top: a pool ticket is a share in a pot, and the pot's final shape is a collective decision made by people who have never heard of you, taken after you have committed.