Automating stakes at a cross-over point

[Bibbs612]

Hi there,

I'm looking to automate football matches by laying an 'over x goals' position just before ko, and then gradually legging out by placing a number of subsequent back bets. For example, laying £50 at 3.5, then backing 5 x £10 stakes at 3.7, 3.9, 4.1, 4.3 and 4.5.

However this presents a challenge with tick sizes, as the tick size changes when the price reaches 4. Therefore, I cannot merely place the 5 back bets at 'custom ticks above the best market price' as once the cross-over point of 4 is reached the tick sizes become bigger. In addition, I am expecting my entry point for the lay bet to be between 3 and 4, therefore the subsequent back bets presumably would need to take their cue from this original lay bet?

Thanks in advance.

Paul

[Dallas]

I'm not sure what the problem is?
It doesn't matter if you go through any cross over points when using the same stake to close the trade

[Bibbs612]

I'm not sure what the problem is?
It doesn't matter if you go through any cross over points when using the same stake to close the trade

Thanks Dallas. Apologies, I haven't explained that very well.

- So 30 seconds before the event start time, the Bot will place a lay bet of £50 on the over/under market, at the 'best market price' where the odds are between 3 and 4. Let's say in this case the lay odds are 3.5.
- At the same time, I would then like to place 5 x back orders of £10 each into the market (with a rule to 'keep all unmatched bets'), with prices increasing by 0.2 relative to the original lay bet. In this case that would be 3.7, 3.9, 4.1, 4.3 and 4.5.
- For the back bet rules, when I use the parameter 'Custom ticks above the best market price' using Ticks of 4 and 8, this places back bets at 3.7 and 3.9 respectively, which is what I want.
- However, the next back bet would be placed at 4.2 and not at 4.1 as I would like. This is because at the cross-over point the tick increments increase by 0.1 rather than 0.05, and therefore 12 ticks above 3.5 is 4.2 and not 4.1.

Is there a way that I can place the increasing back bets, 0.2 apart relative to the original lay bet, notwithstanding that the original lay bet could be anywhere between 3 and 4? Merely using 'Custom ticks above the best market price' with ticks of 4, 8, 12 etc won't do it, as the Bot won't know at outset exactly where between 3 and 4 the original lay bet will be placed.

Thank you.

Paul

[ShaunWhite]

Hi Paul, number of ticks will always do just that. I'm not sure how you'd do it in Guardian but I don't use ticks I always use implied probality offsets, ie if a horse is 23% chance I'll offset at 24% 25% etc.... And obviously those just need converting to prices (a slightly involved process due to rounding) . For me it's a much more logical approach than using the percentages implied by the step sizes at places Betfair have chosen to use.

Also for things that use price depth (as something like WoM would) I create 'phantom' ticks if my calculation includes a cross over, eg 3.9, 3.95, 4.0, (4.05), 4.1, (4.15) which have zero available and traded. It gives a slightly more accurate view of volume in relation to probability because as you say 1 tick below 4.0 isn't the same as 1 tick above it.

You might find some way to do what you want with a combination of those two ideas, but I'm not sure how with a limitted set of features.

[Bibbs612]

Hi Paul, number of ticks will always do just that. I'm not sure how you'd do it in Guardian but I don't use ticks I always use implied probality offsets, ie if a horse is 23% chance I'll offset at 24% 25% etc.... And obviously those just need converting to prices (a slightly involved process due to rounding) . For me it's a much more logical approach than using the percentages implied by the step sizes at places Betfair have chosen to use.

Also for things that use price depth (as something like WoM would) I create 'phantom' ticks if my calculation includes a cross over, eg 3.9, 3.95, 4.0, (4.05), 4.1, (4.15) which have zero available and traded. It gives a slightly more accurate view of volume in relation to probability because as you say 1 tick below 4.0 isn't the same as 1 tick above it.

You might find some way to do what you want with a combination of those two ideas, but I'm not sure how with a limitted set of features.

Hello Shaun,

Thank you for your suggestions. I hadn't thought of using implied probability offsets, so I'll look into that. And I'm not familiar with Weight of Money so will take a look at that also.

To be honest, I think I may just give it a go using the step sizes I indicated previously and just accept the marginal increase in odds. I think there may be a way to indicate that if the odds are at a specific price below 4 then a certain increment should be applied, but I would have to state what this was for every price between 3 and 4, which would involve a lot of rules! If I do find a way to overcome this I will post back here.

Thank again.

Paul