Can you trade historical races in Bet Angel?

[jimibt]

Not sure how trading on past markets with simulated money is any better than trading on live markets with simulated money. Seems like a lot of work for the developers to give you no real benefit. Thing is, there are so many races every day, so plenty of opportunity to practice. When the markets close, spend time analyzing your data.

peter, you disappoint . *Seems like a lot of work for the developers...*. whilst i agree that there are many markets every day, tweaking rules live is only for the faint hearted - been there, not advisable.

for any product that evolves and becomes a world class leader (as BA undoubtedly has), work has to be done. not all of that work has immediate quantifiable benefit. for any product that is seriously presenting a rules based engine, a backtesting facility is a must (it's the same with any software platform. loadtesting, integration testing etc - all are key facets required in order to support the extension of features). for any product that wants to compete with other rules based applications (i won't of course mention what that product is called), it must look out to see if competition exists within that domain and see if there's a way to grab some of those customers who may use product B due to that particular offering.

peter, i know this is a pretty polarised topic and it's obvious where i stand on things but as i suggested in a previous post, there are simple(r) ways to allow a test market to be statically linked to the application via an editable xml file which the end user could use as-is or if more savvy, could edit to simulate a previous market condition. i'd be more than happy to lessen the developer workload by proposing that as a halfway house. that way, one half of the divide gets something to work with, the other half can safely ignore it.

[Bluesky]

This all sounds a lot more complicated than I was expecting. It sounds like jimibt is an IT professional unfortunately most of what he has said has gone right over my head, that's my fault btw not his.

I was hoping to just click on a couple of buttons and hey presto I could practice in simulation mode on some past races, like one can with financial markets.

The point someone made regarding there are plenty of races is a fair one, yes there are lots of races, but sometimes they are at inconvenient times. I thought if it was not too difficult or costly to implement pervious racing data then that would have been handy for newbies or even oldbies that are still loosing money.

It appears that this is a lot more difficult and expensive than I imagined.

[ruthlessimon]

Trouble is guys, it's not really the holy grail, it's better to model a market and project from there and focus on getting in before the move starts or focus or on your fill. Historical data is fine, but I don't think you will find an edge there.

Sorry to bring this back, but I still can't get my head around this - & I'm as open-minded as they come !

7 figure traders in the city, live & die, by being able to replay their strategies & models to hone their execution skills. I can't believe why BF trading should be so different.

[ShaunWhite]

Trouble is guys, it's not really the holy grail, it's better to model a market and project from there and focus on getting in before the move starts or focus or on your fill. Historical data is fine, but I don't think you will find an edge there.

Sorry to bring this back, but I still can't get my head around this - & I'm as open-minded as they come !

7 figure traders in the city, live & die, by being able to replay their strategies & models to hone their execution skills. I can't believe why BF trading should be so different.

I was lost at " it's better to model a market and project from there"

Can you point me (us?) to any further reading on how to do this? I've seen a couple things relating to financials but not for sports markets which can behave quite differently.

[ShaunWhite]

In fact, a whole series on how to develop, test and document strategies, from either a hunch or 'modelling' would be incredibly useful.

You've often given us all a fish, but what I want is the net....or a clue to where I can learn how to make a net

[ruthlessimon]

In fact, a whole series on how to develop, test and document strategies, from either a hunch or 'modelling' would be incredibly useful.

You've often given us all a fish, but what I want is the net....or a clue to where I can learn how to make a net

100% agree. I don't know why the forum isn't rife with this. The process of modeling & edge development should be everywhere. A trader who knows how to make an edge, will have an edge for life.

A quote from the chief trader at a London prop firm: "The timeline to profitability is a function of how well you debrief & record patterns"

[ShaunWhite]

Might be worth starting a new thread Si ?

And if it gets all mathsy, maybe you could act as a translator for those of us who spent more time smoking round the back of the bike shed than we did in class

[ruthlessimon]

Perhaps, but I'm nowhere near close to the profitability of Peter, so my process of edge creation obviously still has many many holes.

Although, the devil in me says, "A trader who knows how to make an edge, will have an edge for life." is the reason it's not shared - at least openly

[to75ne]

In fact, a whole series on how to develop, test and document strategies, from either a hunch or 'modelling' would be incredibly useful.

You've often given us all a fish, but what I want is the net....or a clue to where I can learn how to make a net

100% agree. I don't know why the forum isn't rife with this. The process of modeling & edge development should be everywhere. A trader who knows how to make an edge, will have an edge for life.

its a zero sum game, not such a good thing to give too much away, better to have lots of people not winning.

[ShaunWhite]

In fact, a whole series on how to develop, test and document strategies, from either a hunch or 'modelling' would be incredibly useful.
You've often given us all a fish, but what I want is the net....or a clue to where I can learn how to make a net

100% agree. I don't know why the forum isn't rife with this. The process of modeling & edge development should be everywhere. A trader who knows how to make an edge, will have an edge for life.

its a zero sum game, not such a good thing to give too much away, better to have lots of people not winning.

I'd agree if the market only contained traders, but there's bookies and punters out there who can pay a few quid extra to make the whole trading population a little extra if everyone is a bit sharper and more tooled up. Like the new BA version, everyones going to gain (apparently) so I hope that's at the expense of the bookies and punters and not each other.

I've a feeling that the whole trading ecosystem only functions because there's almost as many edges as there are traders, but most traders just sit comforably in the groove of their favourite few styles/methods; horse reversals, breakouts, sup/res, B2L, scalping, football draws, tennis comeback etc etc etc ...I suspect only a few really squeeze the pips out of every method. There's basically room for all of us.

Certainly if there weren't many ways to do this then the whole system would grind to a halt. So giving people the key to unlocking even more hidden nuggets would actually be a good thing Variety is the spice of life, and for the markets too?

[ruthlessimon]

Like the new BA version, everyones going to gain (apparently) so I hope that's at the expense of the bookies and punters and not each other.

That's a really good point I hadn't thought of that. In theory, it's actually mathematically impossible for everyone to gain.

But this is the thing. As edges become common knowledge, surely the person who had the edge originally is the first to realize it's fading away - & the first to go & adjust it. Similarly to a trend in a market. By the time the punters etc. are aware of the edge (trend goes exponential), the smart money/traders have already moved on to new ideas & avenues.

Could this graph accurately describe the lifecycle of an edge?

[Bluesky]

Sorry to bring this back, but I still can't get my head around this - & I'm as open-minded as they come !

7 figure traders in the city, live & die, by being able to replay their strategies & models to hone their execution skills. I can't believe why BF trading should be so different.

I agree 100%, I was actually shocked when I started looking at sports trading that this facility wasn't already available. Talking to several other sports traders a lot of them would welcome this option and many of them would be prepared to pay a premium (not a large fee) but something most people would consider reasonable to have this facility built in to BA.

If I ran a sports trading software company I think this option would be a 'game changer' and I would see if it could be introduced at a reasonable cost to my customers. I would also imagine that my competition would be extremely worried if I was to provide this option. I suspect they would very quickly have to do the same or go out of business.

[LinusP]

You can buy historical data from betfair so it would only involve BA implementing the logic to read/parse this data (it’s in streaming format so the logic is already there), don’t see how that is against betfair’s terms and conditions?

I and many others who program their own applications already do this and I assumed its why betfair offered historical data in the first place. At the moment its limited to those who can program, featuring it on BA would increase the revenue 100x, for many financial exchanges selling data is one of their main money makers.

http://app.ft.com/content/785092ec-33d8 ... ee5ffe5b5b

[spreadbetting]

Sorry to bring this back, but I still can't get my head around this - & I'm as open-minded as they come !

7 figure traders in the city, live & die, by being able to replay their strategies & models to hone their execution skills. I can't believe why BF trading should be so different.

I agree 100%, I was actually shocked when I started looking at sports trading that this facility wasn't already available. Talking to several other sports traders a lot of them would welcome this option and many of them would be prepared to pay a premium (not a large fee) but something most people would consider reasonable to have this facility built in to BA.

If I ran a sports trading software company I think this option would be a 'game changer' and I would see if it could be introduced at a reasonable cost to my customers. I would also imagine that my competition would be extremely worried if I was to provide this option. I suspect they would very quickly have to do the same or go out of business.

Betfair currently offer the pro data at a fee of £200 per month for the Horseracing,£200 for soccer etc. Surely if there was such a demand for what you're all asking some developer would have released something by now to cash in on that demand.

For BA to amend the app to import that data and act as some kind of 'practice' mode would them mean an additional cost added to the app to cover any development costs etc.


I remember the big clamour when PC was introduced and everyone saying they'd switch to Betdaq if only the software was available. I doubt the demand is really there to cover the cost or hassle of dealing with Betfair to introduce it to an existing app.

[Euler]

People seem to baulk at the cost of software to trade at £12.50 a month, so seeing people willing to pay £200 a month is a bit of a stretch TBH. Betfair also wants to charge us a hefty fee to make it available via an app as well. I can see the appeal to serious users.

I use data a lot to construct ideas but almost never to backtest strategies. Ideas come out of the data which I then test in the market. There is a huge leap from idea to a practical application, so all my best strategies have come from direct participation.

I think there is a better solution for customers. But I'll keep that under my hat for the moment.