***** excel triggered betting

[1/4 of a millionaire fund]

I'd agree with lewis, it's alot more reliant on your strategy at the end of the day rather than coding skills. Most of my excel bots just use simple formulas with a little vba thrown in. Sometimes the simplest of ideas are the best as the more complicated you make the bot the more likely things can go wrong or fail due to lack of liquidity especially for 'trading' type bots.

With a lot of strategies you need to look at the long term picture and not dismiss things just because they don't turn in a profit every race, even bots that break even can be useful for reducing PC charges and a little tweaking can easily turn them into profit.

[ferdinand]

Ok, all sounds good. I'm thinking that if it's possible to write simple profitable bots using excel, does that mean a dedicated off the shelf programmable bot can work even better if given a good strategy? e.g. market feeder or bfexplorer etc.

[Timstertoo]

Ok, all sounds good. I'm thinking that if it's possible to write simple profitable bots using excel, does that mean a dedicated off the shelf programmable bot can work even better if given a good strategy? e.g. market feeder or bfexplorer etc.

No I think the order of tweakability is like this:

Programming yourself > excel triggers with ***** > Marketfeeder pro etc

[Todoroff]

If I have already ordered a bot like Bf Bot Manager or BetBotPro or some other bot does this means that i am step behind ***** users.(if we talk for simple strategies)

[1/4 of a millionaire fund]

If I have already ordered a bot like Bf Bot Manager or BetBotPro or some other bot does this means that i am step behind ***** users.(if we talk for simple strategies)

Can't see any reason why they'd put you one step behind, the whole point of those programs is to simplify the botting process so you don't need coding skills. As long as they've been well written and have all the criteria you'd need hard coded within them there's no reason they should be at a disadvantage.

Ok, all sounds good. I'm thinking that if it's possible to write simple profitable bots using excel, does that mean a dedicated off the shelf programmable bot can work even better if given a good strategy? e.g. market eater or bfexploder etc.

Again I can't see why they'd have the edge over other programs but it'll all be down to what you need the program to do and if it can do what you need it to. More of a case of finding what suits you best when choosing these progs, if you have no excel or other programming skills then they may be the best way forward as you'll be much better off using your time to tweak any strategies rather than trying to learn how to code it within any program. On the other hand you may wish to bring in other data such as SP's or your own database which is where, for some, it's worth the effort of learning excel.

As long as it's possible to 'code' your strategy into the program then they'll all be pretty much the same in terms of speed, it's possible they may even be faster due to removing the excel intergration link but in real terms we're likely to be talking milliseconds not seconds and if your strategy relies heavily on speed a custom written app would be the way forward.

[lewismbet]

I think the only hinderance with an off the shelf bot is that you are slightly constrained as to controlling your bots inputs as you are changing the value of the bot dials rather than making your own dial [hope that makes sense]. Excel [or even better custom] will allow you to think a little more outside the box.

Something that I felt allowed me to make a big step forward was to move from a single strategy on every race to having a barebones which runs every race playing my main edge and a number of smaller, rarer strategies for unusual occurences or which may only fire a few times over a whole days racing.

If i have a new improvement which fires 5 times a day and averages 50p a time it is a 'useless' trading strategy for an individual however if it can run in the background of your main bot that is £17.50 a week. Then add a few more of these... Don't go looking for record breaking profit each race just something that ticks away in the background but is disciplined and willing to work for minimum wage.

[ferdinand]

The reason I thought an off the shelf bot might be better would be because it would include lots of variables to use, triggers and various betting methods - ie it would already implement all the hard work. I guess that bot makers don't necessarily know the perfect mix of ingredients to build perfect bots. They probably wouldn't be selling bots if they did.

lewismbet, I like that idea of having lots of strategies running at the same time. Do you find it easy to keep track of individual strategies?

[Todoroff]

I think with big bank almost every strategy can be profitable. MF you said that you have several simple bots and they bring you a few K per month. It is interesting to me, how big is your account, if it's not a secret of course.

[Timstertoo]

The reason I thought an off the shelf bot might be better would be because it would include lots of variables to use, triggers and various betting methods - ie it would already implement all the hard work. I guess that bot makers don't necessarily know the perfect mix of ingredients to build perfect bots. They probably wouldn't be selling bots if they did.

lewismbet, I like that idea of having lots of strategies running at the same time. Do you find it easy to keep track of individual strategies?

Yes if they knew how to program a winning bot they wouldn't sell it. Nevertheless a simple to use botmaking program is something that also provides value.

I think the botbuilding program that would offer you ultimate freedom and tweakability yet through an extremely intuitive interface that even a 4 or 84 year old would get would need an amount of programming hours that would make it hard to be profitable. The customer potential just isn't big enough. Which is a good thing for people like MF

I think with big bank almost every strategy can be profitable. MF you said that you have several simple bots and they bring you a few K per month. It is interesting to me, how big is your account, if it's not a secret of course.

Hi Todoroff,

I don't think this is true. Rather the opposite. If something is profitble with a big bank it will also work with a small bank. But the other way around you can't endlessly up the stakes as you get in trouble with liquidity. If you could some of the people on this forum (not me) would be independentely wealthy already.

I'd be very carefull automating stuff like forcing the market up or down with big sums of money. If someone catches on that could turn out to be a very expensive bot indeed.

And if the strategy (not involving spoofing) doesn't work with a small bank, it will definitely not work with a big bank.

[1/4 of a millionaire fund]

I think with big bank almost every strategy can be profitable. MF you said that you have several simple bots and they bring you a few K per month. It is interesting to me, how big is your account, if it's not a secret of course.

Be nice if you could use a big bank for every strategy but generally you only profit if the market is wrong, and whilst the market can be wrong on numerous occasions it won't be wrong to large volumes for obvious reasons. None of my bots will be exposed to over £100 a market at any time.

[lewismbet]

lewismbet, I like that idea of having lots of strategies running at the same time. Do you find it easy to keep track of individual strategies?

The way I overcame this was with two betfair accounts [fine with betfair if they have different deposit methods] although this have 20RPS between them not each. One account runs winning strategies and the other uses small stakes and is my test bed.

On the main account most of the code looks like: triggers - submission - monitor [stoplosses/exit variables] - close -reset.

Before I reset/re-arm [on ***** basically clearing the betrefs/trigger columns] I copy all relevant data such as stakes/ and backlay prices to sheet three and I have a column where the strategy number goes so that I can see what bets originated from each trigger conditions.

[I don't actually use this just now as I found I was generating alot of data but was rarely mining it for info]

Re: Big Bank balances.

My main bot can run with as little as £300 in the account and makes a very respectable profit compared to stakes/Bankroll. In fact I have found that I spend most of my time trying to get larger stakes on and failing even at relatively low stakes. Most edges occur in unusual situations where there may not be a great fill-rate/liquidity available.

[ferdinand]

RE strategy tracking: I'd rather have a hands off approach. I'm currently trying a method I saw in Colleen Magoo's automating exchange betting book to track strategies using account statements from the API.

I have made my bots record every betId in a database table along with the strategy name and a few other variables for possible later analysis. I have written a separate script that runs every hour to call getAccountStatement and record the betIds and P/L amounts for settled bets in another table. I have yet another script that can ask the database to cross reference the tables and generate an individual P/L trail for each strategy and plot it on a chart. So far so good, it seems to be working, though I'm not quite sure what happens on the statement when an official result is overturned etc.

[Berkington]

Hello, long-time lurker here. I've never used ***** and am not really sure what it is. Is it easy to get used to?

Can I just clarify that a screenscraper is used to take certain information off a site without having to do it manually? I hope I've at least understood that part alright.

This is something I've wanted to do for a while, but I'll obviously have to learn both VBA and *****. Is it do-able?