Is Rule 4 fair?

The honest answer: sometimes it takes too much, sometimes too little. It depends on the price of the horse that was withdrawn and on the field that is left.

What the deduction is trying to do

If a horse with a 50% chance is withdrawn, everyone else's chance goes up. A fair deduction is roughly the withdrawn horse's chance of winning. A set scale by price is a rough guide to that. It does not know how good the remaining horses are, how big the field is, or what margin the bookmaker had built in.

A hypothetical to show the point

This is an illustration with invented numbers, not a forecast and not betting advice.

Two joint favourites at evens, a third horse at 10/1 and seven no-hopers at 100/1. Say a handicapper rates the favourites at 90, the third horse at 80 and the rest under 40.

• One favourite withdrawn. The other favourite, now only 10lb clear, might be a fair evens chance. The deduction (45p at evens in the scale we have) takes a big bite of the winnings, and the horse is no better a chance than before. Here the deduction is heavy.
• Both favourites withdrawn. Deductions of about 45p each would be added (up to the 90p cap). A backer of the 10/1 horse sees the odds shrink from 10/1 to roughly 1/1 or less after deductions, yet the horse might now be a very strong favourite. Here the deduction is small compared with the improvement in the horse's chance.
• A 100/1 no-hoper withdrawn. No deduction at all, and no loss.

The calculator on the front page lets you try your own numbers.

What the experts say

• A long-running 2008 Racing Forum debate began with Richard Hoiles's proposal that the deduction should allow for the bookmaker's margin, which the flat scale ignores. Posters were split on whether the present system favours punters or bookmakers.
• Geegeez has said the scale favours punters statistically (we could not open that page to check).
• Our own rough sums, which are inference, not published research, suggest the scale is about fair around 4/1, kinder to the punter at evens to 2/1 and at 6/1 to 14/1.

The 14/1 cliff

A horse withdrawn at 14/1 costs backers 5p in the £. A horse withdrawn at 16/1 costs them nothing. The real difference in chance is less than one percentage point (about 7.1% against 5.9% on raw prices), so there is a small hole in the scale that favours the bookmaker.

Each-way

An each-way bet is two bets of equal size: one on the horse to win and one on it to place. Under Rule 4(C) the deduction applies to your winnings on both parts, win or place, and your stake is still returned on whichever parts win. The calculator has an each-way tick box that shows the result if the horse won, placed, or did not place.

The place terms matter too. Each-way bets are settled on the actual number of starters, so a withdrawal that cuts the field can also cut the places (for example from 1/5 odds for three places to 1/4 odds for two, if the field falls from 8 to 7). Many bookmakers offer better terms than these defaults, so use the terms your bet was settled on.

If the withdrawal leaves the number of places equal to or more than the remaining runners, a "reverse Rule 4" can lengthen the odds on your each-way bet. Forum writers say this costs punters more often than the win-bet deduction helps them.

Lessons

• The deduction comes off winnings only.
• The same withdrawal can be good or bad for you depending on what you backed.
• The price used for the withdrawn horse is the odds current at the official notification (Tattersalls Rule 4(C)); a bookmaker moving that price just before a withdrawal can change your deduction, which is what happened in the Tango Sky case.

Sources: Racing Forum thread, page 2, Racing Post.