- This topic has 11 replies, 8 voices, and was last updated 14 years, 6 months ago by
.
-
Posts
-
Does anyone know of the formula for calculating the CSF and where to find it?
The formula should be published for all to see as thousands bet on it each day and no-one appears to know how the dividend is arrived at.
You can request a copy of the formula from the Association of British Bookmakers.
The formula should be published for all to see as thousands bet on it each day and no-one appears to know how the dividend is arrived at.
…not even those with the formula in front of them.
The bizarre nature of the ‘formula’ was highlighted by the outcome of the 1000Gns.
The CSF for Jacqueline Quest (66/1) to beat Special Duty (9/2) was announced before the subsequent reversal of the placings and it proved to be less than the eventual dividend paid for 9/2 beats 66/1.
AP
Still waiting for a reply as to my request for a copy of the formula.
If it turns out to be as complicated as I’m anticipating then there really can only be one reason and that is the bookies are thieving bleeps!
Interesting point Factorman, I always thought it strange it wasn’t more widely available, despite its legendary level of complication. Would be interested to finally see it if you get hold of it.
I have been dabbling with forecasts for about 3 years now and have found the CSF to be slightly better than the Exacta, not least because you could sort of predict the dividend.
I use a very rough shorthand method for calculating likely CSF dividends before a race…it is not too bad but can be way out sometimes…when factors such as the draw etc. are built into the formula.
Basically it is simply to calculate the price of a double between the two horses in question but do not return the stake for the winning horse (first leg of the double) e.g
Forecast… winner 2/1 second 8/1
A normal double would be £1 x 3 x 9 = £27
CSF estimate………….£1 x 2 x 9 = £18
If 8/1 beats 2/1 then….£1 x 8 x 3 = £24
This seems to have a reasonable tolerance against returnsLooking at Goodwood which I picked at random from yesterday
Race 1
Estimated CSF £27.00
Actual CSF £27.96
Exacta £20.00Race 2
Est CSF £51.60
Act CSF £51.19
Exacta £63.00Race 3
Est CSF £12.25
Act CSF £12.77
Exacta £14.30Race 4
Est CSF £21.00
Act CSF £26.11
Exacta £29.50Race 5
Est CSF £40.00
Act CSF £39.60
Exacta £90.70Race 6
Est CSF £374.00
Act CSF £317.76
Exacta £139.80Race 7
Est CSF £22.00
Act CSF £22.37
Exacta £24.20As expected nowhere near the Exact returns but for the CSF the ‘rough’ estimate worked pretty well with 5 out of 7 returns with £1 of the actual CSF (this would be an above average performance in my experience). It is less accurate when short porice favourites win or a combination of long shots as in races 4 & 6 at Goodwood yesterday. Nevertheless I have found it a pretty reliable way to estimate CSF forecast returns before the start of races.
I had a copy a few years ago, it ran to about 3 pages of A4.
At first sight, it appears horribly complex, but I passed it on to an acquaintance who happens to be a Maths professor and his verdict was that in essence it is surprisingly simple.
The basic return comes from multiplying the odds of first and second, but then there are – or were, as it may have changed since then – several minor adjustments to take account of "extreme" results, such as an odds-on chance beating a 66-1, or two horses from adjacent stalls finishing first and second in a big field.
Returns also come from a curve rather than a straight line, so the ones in the big fat middle are closer to what you might expect than those on the margins.
Overall, and much to my disappointment, he concluded that there was little in the formula that could be construed as naked gerrymandering. The worst that could be said was that some adjustments – the one for the draw in particular – were over-enthusiastic arse-covering, not least because a well-known draw advantage will usually be factored into the SPs anyway.
Mr Frisk – The odds of the favourite, finishing position of the favourite and the overround have a massive bearing on the declared dividend.
You can also twist the odds in your favour by emplying a variation of the Harville formula, but not enough to make the bet pay higher than the expected outcome.
That’s interesting Shabby, have you kept a record of when your calculations are out of sinc?
Type of race, field size and the draw etc.No reply to request so far, shall give them the benefit of the doubt and assume my e-mail is still bouncing around in cyber space!
If there is a re-calculation for the draw if stalls adjacent to one another fill the places, when is it activated? automatically or by a controller? and what if the result has nothing to do with the draw but the horses earn their places on merit, is the punter penalised for forecasting the result spot on?
- You must be logged in to reply to this topic.