The home of intelligent horse racing discussion

# Computer Straight Forecasts

Home Forums Tipping and Research Trends, Research And Notebooks Computer Straight Forecasts

Viewing 15 posts - 1 through 15 (of 27 total)
• Author
Posts
• #745
non vintage
Member
• Total Posts 1268

Whilst I imagine that the exact formula for calculating these is probably quite complicated, is anyone aware of whether it is available for inspection anywhere, or whether it should be available?

Obviously, different factors are taken into account – these would I guess include the over-round percentage, the field size, and the price of the favourite as well as the SPs of the first two horses. I’m also guessing it might include something around extreme draws (as for tricasts), however unfair this may be.

<span style=”color: Green;”>(Oops – missed a R4 here, so example from today not relevant – sorry!)</span>

(Edited by non vintage at 3:22 pm on Jan. 25, 2007)

#36613
non vintage
Member
• Total Posts 1268

tdk – i just added up the SPs and worked that out. thanks. will take out the details of the races…

still wonder whether there is not some obligation for the payout formula to be in the public realm?

#36614
Mr Frisk
Participant
• Total Posts 163

Non Vintage

I used to have a copy of it somewhere – I seem to recall that I phoned up one of the main bookies organisation (BOLA probably, as it was then) and they sent it to me.

I showed it to an acquaintance who happened to be a well-known professor of maths at the local university, and to my surprise, he said that it was relatively straightforward in strict mathematical terms, despite looking horrendously complicated to me (it ran to about five pages of A4).

Basically it’s the odds of the winner multiplied by the odds of the second, but then there are four or five stages after that where "tweaks" are made to shave the dividend: when there’s an odds-on favourite, for instance, or there is a well-known draw bias, like low  numbers at Chester.

Of course, you could argue that the odds of horses in low boxes at Chester already take account of the fact that they have a good draw, so there should be no need to tweak the dividend at all. You could, but they won’t listen to you.

#36615
Formath
Member
• Total Posts 1451

If it is any help this is what Nick Mordin in his book, ‘Betting for a Living’ circa 1992 says about the calculation’:<br>"….. this can be seen from the standard straight forecast offered by most bookies on big races, which operates in a similar way to the Computerised version. In this bet, 1 point is added to the price of the second horse before the odds of the first two are multiplied."

#36616
FlatSeasonLover
Member
• Total Posts 2065

So what is the reason given for shaving the dividend with an odds on favourite?

#36617
roland
Member
• Total Posts 302

agreed with formath, that i thought it was as simple as that.<br>which is why i thought there was always huge value when you get a race down to 2 horses, when 1 is very short but you fancy the bigger priced horse to win, as the added point is a larger percentage of that horses odds. Make sense? not to me!!

#36618
bluechariot
Member
• Total Posts 569

In the long run do Exactas with the tote and you will beat the csf most times

#36619
Glenn
Participant
• Total Posts 1981

The CSF doesn’t take account of the draw. The formula has been constantly revised (downwards) over the years and goes something like this:

(p1st*(p2nd+1))

divided by

the number of excuses they can come up with for the forumula favouring punters (harmonisation, related contingencies etc)

minus

current tax

minus

pre-2001 tax

minus

the rounding up of pre-2001 tax to 10%

minus

levy

minus

Lord Donoghue’s hospitality packages

minus

lizards’s xmas do expenses

minus

expenditure on brown envelopes

minus

repayments on the Kepmton redevelopment loan

plus

the drippings from Tom Kelly’s nose

<br>Here p is the price after it has been adjusted to reflect bookies’ wet dreams of what it really ‘should’ be – around 2.5% a runner.

(Edited by Glenn at 7:01 pm on Jan. 25, 2007)<br>

(Edited by Glenn at 7:03 pm on Jan. 25, 2007)

#36620
Ultimate Nightmare
Member
• Total Posts 326

it used to work on the bags dogs forcasts roughly ish, mind you that was many decades ago :biggrin:

#36621
Anonymous
Inactive
• Total Posts 17718

UN <br>It used to work, (the Nick Mordin formula), with reasonable accuracy on horse racing too.<br>If I recall correctly, didn’t they add the draw to the formula when someone cleaned up a couple of times on high draw combinations at Thirsk – could even have been a member of this forum?

#36622
Mr Frisk
Participant
• Total Posts 163

I think it was Patrick Veitch who cleaned up at Hamilton, won Ãƒâ€šÃ‚Â£250,000 when the top three stalls came in at big prices – and yes, not long after that, BOLA announced that the draw would be factored in to the formula.

I believe there was also an adjustment according to the total overround on the race ie. if there was a low profit margin, they would bump it up a bit, but if it was high already, well, that’s another matter…

#36623
Glenn
Participant
• Total Posts 1981

The tricast formula takes account of the draw, the CSF does not (unless they changed it since I last looked).

#36624
Mr Frisk
Participant
• Total Posts 163

Yep, you may well be right about that, it was sloppy to conflate one with the other. Although if the draw isn’t factored into the CSF, you have to wonder why not, since they get away with it when it comes to the Tricast.

#36625
Mr Frisk
Participant
• Total Posts 163

And since I’m in self-flagellatory mood, I think you’re right reet hard, it was Thirsk, not Hamilton.

#36626
wit
Participant
• Total Posts 2152

….and i’m pretty sure it was Paul Cooper and not Patrick Veitch.

i’ve got a VHS tape from a short-lived racing news series of tapes issued in the late 80s / early 90s where Zorro interviews the guy in a pub.

best regards

wit

<br>

Viewing 15 posts - 1 through 15 (of 27 total)
• You must be logged in to reply to this topic.