BACK/LAY by liability, in play BOT.!shock!

[ShaunWhite]

How else does one backtest a strategy that needs a huge sample to show logterm positive results.

Backtesting is using saved data to do your analysis. You'd usually use part of your data to develop your strategy (perhaps in Excel) and a seperate subset of stored data to check you're not just backfitting the first subset.

Trying things in live markets is a trial.

It really shouldn't be the case that you need a "huge" sample to show a trend, unless you're doing very low margin hft stuff but in that case a "huge" sample is turned over in a day or two.

It might be worth you trying to work out what your RoI might be ie profit / (back stakes + lay liabilities) to determine how long a clear result might take to show up. Bigger margins will show up quicker. It varies hugely but a reasonable automated strategy might be in the region of 0.5% - 5% RoI. Anything less and it's hardly worth doing and anything showing much better than that is most likely a temporary anomaly or an error in your calcs.

The way people run long trials without getting restricted is by also doing a reasonable amount of actual bets. Betfair isn't just a playground, we all expect it to be fast and responsive and not bogged down by 000s of people just using the data stream to try random things.

[decomez6]

How else does one backtest a strategy that needs a huge sample to show logterm positive results.

Backtesting is using saved data to do your analysis. You'd usually use part of your data to develop your strategy (perhaps in Excel) and a seperate subset of stored data to check you're not just backfitting the first subset.

Trying things in live markets is a trial.

It really shouldn't be the case that you need a "huge" sample to show a trend, unless you're doing very low margin hft stuff but in that case a "huge" sample is turned over in a day or two.

It might be worth you trying to work out what your RoI might be ie profit / (back stakes + lay liabilities) to determine how long a clear result might take to show up. Bigger margins will show up quicker. It varies hugely but a reasonable automated strategy might be in the region of 0.5% - 5% RoI. Anything less and it's hardly worth doing and anything showing much better than that is most likely a temporary anomaly or an error in your calcs.

The way people run long trials without getting restricted is by also doing a reasonable amount of actual bets. Betfair isn't just a playground, we all expect it to be fast and responsive and not bogged down by 000s of people just using the data stream to try random things.

I see the point of restricting usage to people using practice mode . I will endeavour to put in live bets every single day, regardless of the amount.
‘Back fitting the first subset? You mean repeating the same strategy in the same sample will give Out skewed results? Or does that increase biases towards my expectations, meaning, iam expecting it because I have seen it and I forget to follow the evidence?
Reducing the noise to a range of btwn -5% — 5% .needs A lot of filtering and a great market research tool. The only problem is the quality of any data collected. True and valid results depend on it.
Betfair will not allow enough data for anyone to make a LADDER SIMULATOR. This begs the question how will a backtested strategy behave in an organic , living and reactive market ,When all it knows are static data points.
That aside , standard deviations are dependent on huge samples. I will need to know my extremities, what is the maximum loss and max profit over a given period of time This also needs a lot of storage for a database .
Not unless I find a great market edge, by default or design, I think it’s a Herculean task which can only be achieved once I master guardian And look into the ‘big brother “excel.
I appreciate your advice , thanks again.

[ShaunWhite]

‘Back fitting the first subset? You mean repeating the same strategy in the same sample will give Out skewed results?

No. What I mean is, let's say you have data for 2018 and 2019.
You would look for a strategy that worked in the 2018 data, and then test to see if you got the same results in the 2019 data.
If you just put 2018 and 2019 together in one test, you have a higher risk of backfitting, ie finding that winning combo of parameters in any data that give you winners, but won't necessarily work going forward.

The only problem is the quality of any data collected. True and valid results depend on it.

True

Betfair will not allow enough data for anyone to make a LADDER SIMULATOR.

Not true. What makes you say that? You can save the entire API data stream if you have the means to do it, many do. Befair licensing restrictions prevent that feature from being avaiable in commercially available software.

You can kickstart your data by downloading Betfair data here -> https://historicdata.betfair.com/#/home

Or if you wanted to look at the API then you can do that here ->https://docs.developer.betfair.com/disp ... omq5qye0ni

Or you can search the forum for threads about using BA to collect data.

This begs the question how will a backtested strategy behave in an organic , living and reactive market ,When all it knows are static data points.

Static data is enough to give you the foundations of a strategy, you don't need millisecond data because 1s data will average out enough for some basic analysis. Then some people do move on to a simulation, but the most common way is to then do a very small stakes live test.

You really need to design a strategy around your capabilities, it's perfectly possible to find a working strategy in quite modest data for use in pre-race markets, tennis, football etc etc, but if you think you're going to create some sort of high speed in-running strategy, you either need to seriously improve your data gathering and analysis or be prepared to try things live and refine them, with real money.

Automation will require you to have a reasonable R&D budget you can burn on trials, it's not easy or quick and most people who head that way have also developed some basic ideas first by manually trading for months or sometimes years. Coming up with a plan out of nowhere and then expecting it to just make money is setting yourself up for disappointment. Enthusiasm is great but being realistic is much more useful. It's an ultra competitive world, 000s of people involved and many at levels of sophistication you wouldn't believe, and that's why so few people, iro 5%, who try trading can make it work for them.

Try your idea manually first, if you can't make it work using the tv pics and your live brain it's going to be harder still with a primative bot that's got its eyes closed.

[decomez6]

[quotNot true. What makes you say that? You can save the entire API data stream if you have the means to do it, many do. Befair licensing restrictions prevent that feature from being avaiable in commercially available software.e][/quote]

agreed, i take that back, fake news is real
why is there no vendor offering , trading ladder simulators, with a market depth similar to live trading? this would sort out the waiting for races and clogging up betfair data streams , with 000o traffic.
on a side note,what are the effects of crossmatching on the VWAP and BSP? the matched volume on a selection is only an estimation and not dead on accurate, meaning its not a mirror of weight of money waiting to be matched .ofcoourse diregarding the spoofs and scratched trades.
i keep asking myself , am i visible to all traders when on practice mode? do other traders see my orders. if so , this is a form of un intended spoof, because my orders have no monitary value and yet other traders are making a value judgement with an assumption of the latter being ¬`real orders.

[sniffer66]

[quotNot true. What makes you say that? You can save the entire API data stream if you have the means to do it, many do. Befair licensing restrictions prevent that feature from being avaiable in commercially available software.e]

agreed, i take that back, fake news is real
why is there no vendor offering , trading ladder simulators, with a market depth similar to live trading? this would sort out the waiting for races and clogging up betfair data streams , with 000o traffic.
on a side note,what are the effects of crossmatching on the VWAP and BSP? the matched volume on a selection is only an estimation and not dead on accurate, meaning its not a mirror of weight of money waiting to be matched .ofcoourse diregarding the spoofs and scratched trades.
i keep asking myself , am i visible to all traders when on practice mode? do other traders see my orders. if so , this is a form of un intended spoof, because my orders have no monitary value and yet other traders are making a value judgement with an assumption of the latter being ¬`real orders.
[/quote]

Trades placed in practice mode do not show up in live. Think of the consequences if that were to actually happen

[decomez6]

Try your idea manually first, if you can't make it work using the tv pics and your live brain it's going to be harder still with a primative bot that's got its eyes closed.

[/quote]
this way pleaase the blind bots are self fullfilled prophecies. i guess i live and learn....i will definately take the manual root and see if i can achieve the 5% ROI. iam currently taking a very close look at DALLAS example bots. thats the way to LAS VEGAS!

[decomez6]

[quotNot true. What makes you say that? You can save the entire API data stream if you have the means to do it, many do. Befair licensing restrictions prevent that feature from being avaiable in commercially available software.e]

agreed, i take that back, fake news is real
why is there no vendor offering , trading ladder simulators, with a market depth similar to live trading? this would sort out the waiting for races and clogging up betfair data streams , with 000o traffic.
on a side note,what are the effects of crossmatching on the VWAP and BSP? the matched volume on a selection is only an estimation and not dead on accurate, meaning its not a mirror of weight of money waiting to be matched .ofcoourse diregarding the spoofs and scratched trades.
i keep asking myself , am i visible to all traders when on practice mode? do other traders see my orders. if so , this is a form of un intended spoof, because my orders have no monitary value and yet other traders are making a value judgement with an assumption of the latter being ¬`real orders.

Trades placed in practice mode do not show up in live. Think of the consequences if that were to actually happen
[/quote]
CHAOS galore, with plenty plenty swingers!

[decomez6]

‘Back fitting the first subset? You mean repeating the same strategy in the same sample will give Out skewed results?

No. What I mean is, let's say you have data for 2018 and 2019.
You would look for a strategy that worked in the 2018 data, and then test to see if you got the same results in the 2019 data.
If you just put 2018 and 2019 together in one test, you have a higher risk of backfitting, ie finding that winning combo of parameters in any data that give you winners, but won't necessarily work going forward.

The only problem is the quality of any data collected. True and valid results depend on it.

True

Betfair will not allow enough data for anyone to make a LADDER SIMULATOR.

Not true. What makes you say that? You can save the entire API data stream if you have the means to do it, many do. Befair licensing restrictions prevent that feature from being avaiable in commercially available software.

You can kickstart your data by downloading Betfair data here -> https://historicdata.betfair.com/#/home

Or if you wanted to look at the API then you can do that here ->https://docs.developer.betfair.com/disp ... omq5qye0ni

Or you can search the forum for threads about using BA to collect data.

This begs the question how will a backtested strategy behave in an organic , living and reactive market ,When all it knows are static data points.

Static data is enough to give you the foundations of a strategy, you don't need millisecond data because 1s data will average out enough for some basic analysis. Then some people do move on to a simulation, but the most common way is to then do a very small stakes live test.

You really need to design a strategy around your capabilities, it's perfectly possible to find a working strategy in quite modest data for use in pre-race markets, tennis, football etc etc, but if you think you're going to create some sort of high speed in-running strategy, you either need to seriously improve your data gathering and analysis or be prepared to try things live and refine them, with real money.

Automation will require you to have a reasonable R&D budget you can burn on trials, it's not easy or quick and most people who head that way have also developed some basic ideas first by manually trading for months or sometimes years. Coming up with a plan out of nowhere and then expecting it to just make money is setting yourself up for disappointment. Enthusiasm is great but being realistic is much more useful. It's an ultra competitive world, 000s of people involved and many at levels of sophistication you wouldn't believe, and that's why so few people, iro 5%, who try trading can make it work for them.

Try your idea manually first, if you can't make it work using the tv pics and your live brain it's going to be harder still with a primative bot that's got its eyes closed.

THis is GOLD. you just shortened my learning curve by a WHOLE pint !

[decomez6]

You could use a chart like this to plot out all the swings
viewtopic.php?f=54&t=16632

i modified the back bet to stake at 100% net stake, everything else remains unchanged,
this were my results :
download/file.php?mode=view&id=17770

download/file.php?mode=view&id=17771

[decomez6]

live mode with a £10 stake. showing similar swings , i greened up at the first instance,

[elofan0]

It does throw up some great results in practice mode ...why do i feel it will take the real cash if let it run free

[decomez6]

It does throw up some great results in practice mode ...why do i feel it will take the real cash if let it run free

thanks for the feedback ,
its behaving like a hidden monster. i have used it to make another monster. iam currently testing both. keep testing , two hands are better than one when taming a dragon ! this may be my ticket away from corona virus infested 9-5 jobs.

[elofan0]

hope so for you, im still trying to work out its behaviour why it does well in certain races then the odd one it takes the cash but no system or bot would ever be 100% ...... it jumps suddenly when least expected so maybe the profits would outlay the losses will keep watching it for now ....the easiest part would be to trade out before the race i know it has done that a few times but throws bigger profit in running in practice mode .. not tried with real cash

[elofan0]

Unsure how this could be back tested cos it has a mind of its own lol

[decomez6]

Unsure how this could be back tested cos it has a mind of its own lol

i have the guradian version and crazy numbers of :lay by liabilty 2 back by liabilty :results are too big for me to comprehend what is going on. how does applying one single rule to the whole market ,give such big variations.
out of this shock i made a modified guardian version :lay by liability 2 back by netstake : and that was managable .the variations are not crazy,a small bankk will handle the swings.
THEN! modified it further and tweaked it into one crazy beast, and all over sudden this video makes sense to me:
[https://www.youtube.com/watch?v=DZEiNaXVWSw][/url]

i am working on the new branch and will take it all the way.