Bookmaking Maths

[joe lively]

Hi all, I wonder if anybody can help with some basic maths

I understand that when the book value on the lay side exceeds 100% we can lay every selection for a green book using the correct stakes. Its working out these stakes that I would like help with.

Simple example would be:
Selection 1 - 2.00
Selection 2 - 4.00
Selection 3 - 4.00
Selection 4 - 10.00

The above gives me an overround of 110%. How do I work out how much to stake per selection to acheive a green book?

I know I can use arbcruncher etc but I would really like to know the maths so I can do it myself in excel. Any help would be much appreciated.

[GaryRussell]

I've made an example spreadsheet for you.

http://www.gruss-software.co.uk/Excel/LayBook.xls

[joe lively]

Thanks a lot Gary. I really appreciate you taking the time to do that for me.

Once again a great help.

[manager1]

Hi Joe,

The best way is to use the Betting Assistant in Payout mode which will calculate the stake automatically for you.
There is stake mode, liability mode and payout mode available at the top of the program. Just click on it until Payout is selected in red.

You don't have to worry about calculating stakes, the only worry is getting above 100%

[joe lively]

Hi Manager, yes i have used that method in the past but I wanted to incorporate it into excel.

I am on the understanding it is not possible to use payout mode through excel? Would be great if if you could.

[joe lively]

On another note Manager1, I have been playing around with a system recently that was inspired by one of your theories that you mentioned on another thread.

What i'm doing is monitoring laddy prices with Gary's live prices spreadsheet, then once the market goes in play I lay every selection at 10% lower than the last qouted bookies price. I have set unmatched bets to cancel after 6 seconds. I think I need to adjust that depending on distance.

I am regularly getting matched and although I am currently showing a small loss I am sure long term the theory will hold up.

Do you have any pointers on how to improve on this?

Cheers

[manager1]

Put your payout figure in cell Y2

In Cell S5 put this formula =IF(R5<>"",$Y$2/R5,"")

In Cell R5 put this formula =IF(A5<>"",Y5,"")

Assuming you are using the live bookies prices this should equate to payout mode.

[manager1]

On another note Manager1, I have been playing around with a system recently that was inspired by one of your theories that you mentioned on another thread.

What i'm doing is monitoring laddy prices with Gary's live prices spreadsheet, then once the market goes in play I lay every selection at 10% lower than the last qouted bookies price. I have set unmatched bets to cancel after 6 seconds. I think I need to adjust that depending on distance.

I am regularly getting matched and although I am currently showing a small loss I am sure long term the theory will hold up.

Do you have any pointers on how to improve on this?

Cheers

Yes, I don't worry now about the seconds, what I do is fire in lays for the whole field below bookies and cancel all bets as soon as the first one is matched.

I use this formula to do it.....

=IF(AND(A5<>"",$W$3>0),"CANCEL",IF(AND(A5<>"",$W$3=0),"LAY-IP",""))

Where cell w3 contains this formula

=SUM(W5:W42)

Now you'll find that most matches occur either as the race goes suspended or within a few seconds of reopening.

Don't forget to copy the first two formulae right down to say row 30 or you'll miss half the runners.

[joe lively]

Very helpful, i'll incorporate that into my spreadsheet. I'll give it a go on this afternoons cards. I had been wondering about cancelling the rest once one got matched. Thanks


With you not cancelling lays after a set time anymore, I am assuming then that your getting matched every race?
Looking at say a 7f race and you don't get matched for 15 seconds, would it be possible that we're simply laying horses that are genuinely going well?
Or do you find that even at this stage the horses chances are exagerrated by backers?

Thanks again for the help.

[manager1]

Joe, I find that the premise is right in what you're saying, but if you have the whole field running for you, someone at some early stage is going to make a quick wrong judgement. That error is what makes the profit.
Of course you will get matched on loads of eventual winners, and it's easy to put it down to letting the bet go too long, after looking into this, it seems that the horse would have won regardless in most cases, even if cancelled just after the off.

Basically it evens out and you have enough % on your side to make the errors outweigh your own.

Last night I got matched on a 20/1 and 50/1 horse in the same race at 15/1 and 40/1.
Neither won.

So there is around a 20% edge at bookies prices on those longshots often 30%, I just shaved off another 20% edge for myself without them being eventual winners.

Obviously you're not going to make 140% ROI in the longrun, because you will have many races where the odds on favourite leads, and wins.
You laid 10% lower but you won't get matched on those that fall, those that are slow away and those that miss the start altogether.

The edge is in the misjudgement and overexhuberance of backers early.

Later in the races, they are breakeven.

[manager1]

If you lay one horse and let it go, you will just be laying eventual winners.
That's why I lay the field and cancel after the very first one is matched.
It's the first horse an error is made on that's the qualifier.
That horse may win, but the error is in the odds the backer took on it.

[joe lively]

Very interesting post and on that basis I can see why you let them go until one is matched.

You mentiion the 10% I was using but not getting matched on fallers and slow starters, do you mean I should be aiming at closer to the bookies price?

I have also considered the possibility trading off some of the liability if possible, I decided that I may well be reducing overall profit on horses that eventually went on to lose anyway. But would be interested to hear your thoughts?

[manager1]

Quite frankly, I'm looking at this myself, but I'm not sure that it will return profit.

A lay is matched, either the horse will win or lose.
If the horse wins, you can't back it back for a profit (usually), so you'd be making a loss, if it loses you're giving away an edge. I think the number of times it wins will be offset by the margins of profit in trading out on losers.

Don't take my word on it though, I've only just started looking at it.

[joe lively]

Following your advice yesterday Manager1 I let my bets run until one was matched. Yesterdays results were positive and an improvement on cancelling after 6 seconds.

I'll run with it again today and try and post some results.

[manager1]

Great Joe.
I also had a very good day yesterday and also a good start today.
You must have some patience just like a bookie, two days ago I had a shocker and lost 20% of my bank, everything that shortened won, and anyone could have been excused for wondering if you're just laying horses going to win.
The way I proved it last year was to back the shorteners, and I did my proverbials

Yesterday I got back all the losses and made an additional 20%.

Today I laid a horse at 8/1 with the bookies at 7/1 on Betfair, the bookies firmed to 7/1, so I ended up laying at bookies prices.
The horse was unplaced.