Horse racing computerised Straight Forecast
With the Computerised Straight Forecast (CSF) being such a popular bet it’s surprising just how little information there is available to the punting public on how it all works. Yet this isn’t because people don’t wish to know about its inner workings, no! It’s because it’s so god damn hard.
What is a computerised straight forecast
It’s one of the many different variants of forecast bets, I’d recommend that you read our general forecast betting article to get the bigger picture. For CSF bet’s it’s as simple as you having to pick the winner and second placed horse in a race.
The word simple will now no longer be used in this article cos from here on in, everything gets pretty tricky. For those of you lucky enough to me mathematically minded I have added the workings for CFS bets at the end of the post, but you will have to have studied to a pretty high level to make any sense of it.
Why is it so difficult? Other than the bookies liking you being kept in the dark? Well it’s because of the amount of different variables and data used to calculate the odds. It isn’t a nice single formula. It changes based on whether –
- It’s a handicap race or not.
- If it’s a flat, hurdle or chase race
- The number of runners
- The amount of rags in the race
- The odds of the fav
- The odds of the second fav
The reasoning from the bookies for all this; well it was because they were losing money when applying the correct maths to the forecast. After amassing years of data they could see a human element taking effect.
If the favourite horse didn’t come first, the amount of times it came second didn’t match what the odds said. Whether the horse was just badly priced-up to begin with or whether the jockey didn’t push as hard as he could (thinking about a better handicap for the next race) is unknown.
What is known is that bookies don’t like to lose money. So with this information coupled with the big hits they took every time an odds-on fav won and the second fav followed it home meant a change was only a matter of time.
Computerised straight forecast calculator
Like the majority of our bet calculators our CSF calculator is pretty self-explanatory. But just to be sure I’ll run you through how it should be used.
Firstly add the race details from the drop down menu. The type of race, if it’s a handicap and the number of runners. Then click on set-up. This will add the correct number of horses to the odds section.
Now just enter the decimal odds for all the horses in the race (if you need to convert odds to decimal, check out our decimal odds converter calculator), putting the horse that finished first and second in the top two slots. And that’s it, now click calculate and the computerised straight forecast calculator will return the odds and the races over round margin.
Maths behind the computerised straight forecast
Horse Forecast Formula
Effective from 16/01/97
Consider a race with n runners with starting prices denoted by Si : 1 such that there are ni (>0) horses at Si (I = 1,2,……..,k) with S1 < S2……..< Sk
Let x:1 be the starting price of the winner
Let y:1 be the starting price of the second
Let p, the win betting margin = ∑ ni / (1 + Si)
Let RAGP = 20, the minimum price of rags
Note : All prices are adjusted by rule 4, if appropriate
Definition of RAG
Let NUM20 = ∑ ni where Si >= RAGP
Let MRG20 = ∑ ni / (1+Si) where Si >= RAGP
Then let NUMRAGS, the rag threshold = max(2,NRAGS)
Where NRAGS = (NUM20/3) rounded up
Let SUBTR = MRG20-1/(1+x) and NSUBTR = NUM”)-1 if x >= RAGP
SUBTR = MRG20 and NSUBTR = NUM20 if x < RAGP
And then let nm = max(0, NSUBTR – NUMRAGS)
Let M = 1.075 + 0.015 * (n-1-(NUM20 – NUMRAGS)) + (NUM20 – NUMRAGS)/NUM20*MRG20
Where NUM20 > NUMRAGS
1.075 + 0.015 * (n-1) where NUM20 <= NUMRAGS
Let M0 = 1.075 + 0.015 * (n-1 –nm)
MINMARG is defined as:
|No. of runners||MINMARG|
Let FCMAR = max( MINMARG, m*p)
Adjustment to other horses when favourite wins
If x = S1 then the following values are taken:
Then FAVADJ = min(max((n-STR)*VAR+VAR,1),1.25)
Then let FAVADJ = (FAVADJ-1)*min((1+S2)/(1+S1)-1,1)^0.5+1
(N.B. S1 = price of favourite, S2 = next shortest distinct price
Let SRi = (1+Si)^FAVADJ * MLT – 1 for each Si >= RAGP
Then revise y to (y+1)^FAVADJ * MLT – 1
And RAGP to (RAGP + 1)^FAVADJ * MLT – 1
Then revise SUBTR to ∑ ni / (1 + SRi) where SRi >= RAGP
Let w, the weighting, be defined as:
|Flat||((n-nm-1)^1.8 / ((2^(1.8-1))*p / (p-1 / (1+x)^2)|
|Hurdles||((n-nm-1)^1.2 / ((2^(1.2-1))|
If y=RAGP and NUM20 > NUMRAGS i.e. second is a rag and the number of rags exceeds the threshold
Where MIN1 = M – M0 + (NSUBTR – nm) * M0 / FYRAG
And MIN2 = M*(1+((NSUBTR-nm) / FYRAG-1)*(p-1 / (1+x)-nm / NSUBTR*SUBTR) / (p-1/(1+x)))
And FYRAG = HAR(NSUBTR / SUBTR-1)
Note: IF(cond, a, b) means, if cond is satisfied expression evaluates to a, otherwise evaluates to b
Limitation Factors / Return calculation
Let D0 = (1+x)y1
Then let D0 = Do*p*m / MINMARG if MINMARG > p*m
Then let D2 = max ( D2 , (1+x) * 1.01)