Getting matched at BSP

[Archangel]

Anyone know if you leave an amount to be matched at BSP , will it just match off whatever was available to lay at the BSP at the exact start of the race, and not keep it in play ?

[ShaunWhite]

Change your bet persistance to 'TakeSP'. It does what it says on the tin.

In low liquidity markets you occasionally find the bet lapses because there isn't sufficient money in the SP pool to settle your bet but that not a big deal with sp being close to 0ev.

[jamesedwards]

Anyone know if you leave an amount to be matched at BSP , will it just match off whatever was available to lay at the BSP at the exact start of the race, and not keep it in play ?

An SP bet won't be left for in-play but Betfair can occasionally be late enough to turn a race into play that the final BSP is affected quite significantly. I've seen races where Betfair have turned the race in-play with as little as a furlong left to run, and therefore the BSP on the winner could be completely different to the price at the actual off-time of the race. This only happens occasionally but US racing is particularly prone to these errors and I wouldn't touch BSP on US races as a result.

[Archangel]

Hopefully there will be enough liquidity to match large SP bets. I am currently working on a strategy with a 6.5% edge but that depends a lot on the bets getting taken at BSP

[jamesedwards]

Hopefully there will be enough liquidity to match large SP bets. I am currently working on a strategy with a 6.5% edge but that depends a lot on the bets getting taken at BSP

Just in case you aren't aware, your bet becomes part of the BSP calculation. So the higher your back bet the lower the final BSP will be, and the higher your lay bet the higher the final BSP will be.

An SP bet of a few hundred pounds can influence the BSP of a smallish market quite significantly. Obviously the bigger the market the lower the impact ratio.

[Archangel]

Just in case you aren't aware, your bet becomes part of the BSP calculation. So the higher your back bet the lower the final BSP will be, and the higher your lay bet the higher the final BSP will be.

An SP bet of a few hundred pounds can influence the BSP of a smallish market quite significantly. Obviously the bigger the market the lower the impact ratio.

Yeah we are talking 3 figure amounts on average markets and potentially 4 figures on larger markets

Thats interesting. So, in effect, if I am placing sizable bets at SP I could be negatively affecting my potential profit by reducing the BSP ?

In that case it might be better to just take price on offer in and around the off

[jamesedwards]

Thats interesting. So, in effect, if I am placing sizable bets at SP I could be negatively affecting my potential profit by reducing the BSP ?

That's my understanding of how it works. As soon as you become part of the SP market your stake affects the final BSP and the higher your stake the more it will affect it. https://promo.betfair.com/betfairsp/FAQ ... kings.html

[napshnap]

Just in case you aren't aware, your bet becomes part of the BSP calculation. So the higher your back bet the lower the final BSP will be, and the higher your lay bet the higher the final BSP will be.

An SP bet of a few hundred pounds can influence the BSP of a smallish market quite significantly. Obviously the bigger the market the lower the impact ratio.

...

Thats interesting. So, in effect, if I am placing sizable bets at SP I could be negatively affecting my potential profit by reducing the BSP ?

...

Yes, and you can't calc exactly how much your SP bets will affect SP.

In that case it might be better to just take price on offer in and around the off

This is questionable, you'll have to manage a matching of your bets (can be a pain in the arse in illiquid markets), no guarantee that your bets will be matched.

[decomez6]

Hopefully there will be enough liquidity to match large SP bets. I am currently working on a strategy with a 6.5% edge but that depends a lot on the bets getting taken at BSP

Taken at one particular BSP ? OR within range of BSP?
- I would take one BSP (x) Then test the strategy Using only x as the entry or closing price depending on your goals.
-then test the strategy on a range of BSP’s . + or - 10% of BSP (x).
This will then show you how well the strategy performs if you miss the BSP and end up getting + or - 10% of your asking BSP (x) price .
…if you are still profitable in this scenarios , then you can place a BSP limit bet , only accepting the bet if the BSP is within the accepted range.

From my observation ( I might be wrong ) . The BSP is usually + or -10% of the last traded price @ the off.
And as Shaun stated, it won’t matter as in the long run the -10 +10 = 0

[Archangel]

Taken at one particular BSP ? OR within range of BSP?
- I would take one BSP (x) Then test the strategy Using only x as the entry or closing price depending on your goals.
-then test the strategy on a range of BSP’s . + or - 10% of BSP (x).
This will then show you how well the strategy performs if you miss the BSP and end up getting + or - 10% of your asking BSP (x) price .
…if you are still profitable in this scenarios , then you can place a BSP limit bet , only accepting the bet if the BSP is within the accepted range.

Taken at the official BSP only. The edge calculation comes from backtesting around 6800 bets over 5 years.
(Total Profit / Total Risked) * 100
Average odds around 4.0 so wanting to keep towards the business end of the market where the liquidity is. Also some stress testing with max losing runs etc.
But as you suggest, it might be worth testing for a BSP range also

[jamesedwards]

Taken at one particular BSP ? OR within range of BSP?
- I would take one BSP (x) Then test the strategy Using only x as the entry or closing price depending on your goals.
-then test the strategy on a range of BSP’s . + or - 10% of BSP (x).
This will then show you how well the strategy performs if you miss the BSP and end up getting + or - 10% of your asking BSP (x) price .
…if you are still profitable in this scenarios , then you can place a BSP limit bet , only accepting the bet if the BSP is within the accepted range.

Taken at the official BSP only. The edge calculation comes from backtesting around 6800 bets over 5 years.
(Total Profit / Total Risked) * 100
Average odds around 4.0 so wanting to keep towards the business end of the market where the liquidity is. Also some stress testing with max losing runs etc.
But as you suggest, it might be worth testing for a BSP range also

The problem with the back-testing is that your stakes were not part of the market. eg All the BSPs would have been lower (by an unknown value) if you had been backing at SP, so you need to be confident there was/is enough edge to cover the impact of you entering the market.

[Archangel]

Taken at one particular BSP ? OR within range of BSP?
- I would take one BSP (x) Then test the strategy Using only x as the entry or closing price depending on your goals.
-then test the strategy on a range of BSP’s . + or - 10% of BSP (x).
This will then show you how well the strategy performs if you miss the BSP and end up getting + or - 10% of your asking BSP (x) price .
…if you are still profitable in this scenarios , then you can place a BSP limit bet , only accepting the bet if the BSP is within the accepted range.


The problem with the back-testing is that your stakes were not part of the market. eg All the BSPs would have been lower (by an unknown value) if you had been backing at SP, so you need to be confident there was/is enough edge to cover the impact of you entering the market.

The question is, how do you quantify the impact of my bets entering the market, or is that even possible?

[ShaunWhite]

It's not possible to work out your impact. But the advantage with bsp is that you gain 1/2 a tick vs taking a price, as SP is usually between last back and last lay price. Otherwise a price near the start will effectively be bsp because there's no particular bias, it's not consistantly better or worse. If it was that would be a strategy in itself.

Maybe try to get on with offers (and theoretically gain 1/2 a tick vs BSP) and leave any remaining unmatched at sp?

Start small and scale up until it fades? And hope someone else launches a bot doing the opposite and it's even better