New user here: How am I doing?

[waytogo78247]

Hello all,

The 14 days trial was nowhere near long enough as in that period must have watched 25+ hours of video's and ended up banging my head on a brick wall. However, I decided to give it a go and although still in practice mode thought I would give Southampton FC a years subsription. Not sure that will help them as I heard that because of current restrictions on distancing that they had to use wheelie bins in their training sessions to stop the spread of infection. Problem was the wheelie bins won the training session 4-2!

OK, I have betting racing for nearly all my life and have done very well from it, but trading is a different animal. I usually have a grasp of what the public will bet ( in general ) and made a general algorythm that uses all races and picks a selection in about half to two thirds of all races UK/IRE that I think will get some betting interest. This is from using racing data and not past market data.

I started to test this on Monday as the first three days of the week are modest in liquidity and it appears even worse right now with what I can only guess that a lot are saving their cash for Cheltenham

I bet the whole range of odds, so as this is a short few weeks test I used liabilty stakes (percentage) of just £300 (I am even tight when in practice mode). This also will give me a better idea of what is happening in a much shorter period than convention level stakes.

I only go in the last 20 mins of the market and set the bet to back the liabilty to £300 whatever the price and ask for an ungreedy 4 ticks offset lay.
If fully matched at any time Close and Green.
If not matched in the last few seconds then cancel and re-lay at current odds then green.
No stop loss used as I am expecting these horses to get some action and huge drifts even on outsiders I can't see happening very often, it at all.

I hope I have explained that clearly?

I have one left to go today and in three days have results of:

Runs 45
Wins 33 73.33%
Profit 137.81

Not mind blowing and a tad dissapointing as I wanted 80-85% success.

I already noted the low liquidity this week and even a Kempton race (Nov stakes) not long ago had only £34k in the market at five minutes to post!
IRE racing, which I have noted on quite a few videos as not being reccomended, today was poor. With 4 trades at Wexford losing 3 (-£42.11) and 1 gain (+£3.48) for a total -£36.63. These are of course in the full results given earlier.

I know this is a very short test so far but am I looking anywhere near the right direction for automated trading? Forgot to say that all this is set up in Guardian.

[beermonsterman]

Hello all,

The 14 days trial was nowhere near long enough as in that period must have watched 25+ hours of video's and ended up banging my head on a brick wall. However, I decided to give it a go and although still in practice mode thought I would give Southampton FC a years subsription. Not sure that will help them as I heard that because of current restrictions on distancing that they had to use wheelie bins in their training sessions to stop the spread of infection. Problem was the wheelie bins won the training session 4-2!

OK, I have betting racing for nearly all my life and have done very well from it, but trading is a different animal. I usually have a grasp of what the public will bet ( in general ) and made a general algorythm that uses all races and picks a selection in about half to two thirds of all races UK/IRE that I think will get some betting interest. This is from using racing data and not past market data.

I started to test this on Monday as the first three days of the week are modest in liquidity and it appears even worse right now with what I can only guess that a lot are saving their cash for Cheltenham

I bet the whole range of odds, so as this is a short few weeks test I used liabilty stakes (percentage) of just £300 (I am even tight when in practice mode). This also will give me a better idea of what is happening in a much shorter period than convention level stakes.

I only go in the last 20 mins of the market and set the bet to back the liabilty to £300 whatever the price and ask for an ungreedy 4 ticks offset lay.
If fully matched at any time Close and Green.
If not matched in the last few seconds then cancel and re-lay at current odds then green.
No stop loss used as I am expecting these horses to get some action and huge drifts even on outsiders I can't see happening very often, it at all.

I hope I have explained that clearly?

I have one left to go today and in three days have results of:

Runs 45
Wins 33 73.33%
Profit 137.81

Not mind blowing and a tad dissapointing as I wanted 80-85% success.

I already noted the low liquidity this week and even a Kempton race (Nov stakes) not long ago had only £34k in the market at five minutes to post!
IRE racing, which I have noted on quite a few videos as not being reccomended, today was poor. With 4 trades at Wexford losing 3 (-£42.11) and 1 gain (+£3.48) for a total -£36.63. These are of course in the full results given earlier.

I know this is a very short test so far but am I looking anywhere near the right direction for automated trading? Forgot to say that all this is set up in Guardian.

Peter did say things go pretty quiet just before Cheltenham.
Looks like you onto some thing good their the only way you will know for sure is if it holds the test of time I would guess .

[Derek27]

Using Practice Mode Responsibly & Avoid Betfair Restrictions

[waytogo78247]

Thank you for the replies guys. I was also not aware that Betfair cancels accounts for using practice mode. I have done 3 days of testing with poor liquidity and will see how it goes with the higher weekend markets. The main problem I was trying to avoid was Data Fee's from Betfair with too many requests to the API. I have never found anyone that can give me a precise answer on this, so here is what I do.

In Test mode.

My list of races are put up at around 10:00am on Guardian and the refresh rate is 1 second per race. Today there were 21 races with possible bets.
20 Mins before post they are put into the main Bet Angel screen @ 1/2 second.

That is it. All these are on UK/IRE racing were my usual bets have dropped to under £20 per day.

I also use MFP which I have for way over a decade and currently place around £8000-£10,000 per month in bets (not trading). This is from about 25 USA races per day that run for just 20 mins before off at a refresh rate of 2 seconds.

I hope that won't be a problem, especially with my low turnover in UK markets and using practice mode.

Checking my lifetime account page I have also paid them over £75,000 in commission,.

Peter does state in his video's that each market is different. maiden, Handicaps, low class, high class etc but I am using a generic method that usually finds well bet horses. Peter also states that you need a plan for each market and that is basically my plan for every market with a "bet down" selection.

[Derek27]

Thank you for the replies guys. I was also not aware that Betfair cancels accounts for using practice mode. I have done 3 days of testing with poor liquidity and will see how it goes with the higher weekend markets. The main problem I was trying to avoid was Data Fee's from Betfair with too many requests to the API. I have never found anyone that can give me a precise answer on this, so here is what I do.

In Test mode.

My list of races are put up at around 10:00am on Guardian and the refresh rate is 1 second per race. Today there were 21 races with possible bets.
20 Mins before post they are put into the main Bet Angel screen @ 1/2 second.

That is it. All these are on UK/IRE racing were my usual bets have dropped to under £20 per day.

I also use MFP which I have for way over a decade and currently place around £8000-£10,000 per month in bets (not trading). This is from about 25 USA races per day that run for just 20 mins before off at a refresh rate of 2 seconds.

I hope that won't be a problem, especially with my low turnover in UK markets and using practice mode.

Checking my lifetime account page I have also paid them over £75,000 in commission, but have made over £190,000 Gross.

If you've made £190K why do you need to use practice mode? Can you not use real money?

[rik]

The 20 request per second was only an issue before API streaming was introduced
With the 1000 transactions per hour they seem to be quite lenient as well, but with your settings unless you keep cancelling and readjusting a lot of bets you wont get over a 1000 anyway

[Anbell]

The 20 request per second was only an issue before API streaming was introduced
With the 1000 transactions per hour they seem to be quite lenient as well, but with your settings unless you keep cancelling and readjusting a lot of bets you wont get over a 1000 anyway

"Transaction Charges only apply to customers recording more than 5000 transactions an hour."

[waytogo78247]

The 20 request per second was only an issue before API streaming was introduced
With the 1000 transactions per hour they seem to be quite lenient as well, but with your settings unless you keep cancelling and readjusting a lot of bets you wont get over a 1000 anyway

"Transaction Charges only apply to customers recording more than 5000 transactions an hour."

What is considered a transaction? Is it a bet placed or an update of a price/race? This I cannot find anywhere on Betfair or get a clear answer so I can make a calculation.

[Anbell]

The 20 request per second was only an issue before API streaming was introduced
With the 1000 transactions per hour they seem to be quite lenient as well, but with your settings unless you keep cancelling and readjusting a lot of bets you wont get over a 1000 anyway

"Transaction Charges only apply to customers recording more than 5000 transactions an hour."

What is considered a transaction? Is it a bet placed or an update of a price/race? This I cannot find anywhere on Betfair or get a clear answer so I can make a calculation.

"7. Transaction Charges
7.1 This section only applies to the Betfair Exchange.

7.2 In addition to any commission payments, Betfair customers will be charged a transaction fee if they process a large number of transactions on the main Betting Exchange or on Exchange Games.

This charge is designed to limit inefficient transactional behaviour, particularly at peak times.

7.3 A ‘transaction’ shall include all bets placed and all failed transactions. Transaction Charges only apply to customers recording more than 5000 transactions an hour.

7.4 For the purpose of calculating Transaction Charges, an “hour” means a continuous period of 60 minutes based on the standard 24-hour clock system. Each new ‘hour’ will start at on the hour (e.g. 10:00:00) and finish at 59 minutes and 59 seconds past that hour (e.g.10.59:59)."

https://www.betfair.com/aboutUs/Betfair.Charges/

[waytogo78247]

Thanks that is something I have wanted to know for a lot of years OR I may have read this before.

7.3 A ‘transaction’ shall include all bets placed and all failed transactions. Transaction Charges only apply to customers recording more than 5000 transactions an hour.

Where I have taken "include" to be "on top of" something else like requests to the API for market updates etc. It really is not written very well and is not an "all inclusive" statement of what a transaction charge is.

So, I need not fret if I have guardian with 30 races updating more then once per second.

[Anbell]

Thanks that is something I have wanted to know for a lot of years OR I may have read this before.

7.3 A ‘transaction’ shall include all bets placed and all failed transactions. Transaction Charges only apply to customers recording more than 5000 transactions an hour.

Where I have taken "include" to be "on top of" something else like requests to the API for market updates etc. It really is not written very well and is not an "all inclusive" statement of what a transaction charge is.

So, I need not fret if I have guardian with 30 races updating more then once per second.

Getting data from BF is not a 'transaction' - only if you place a bet or change a bet, or try to place a bet that is rejected (e.g. if you 'take sp all' every refresh) or other things like that.

It was changed about 6 months ago to go from 1000 to 5000 per hour, but they broadened the definition to include failed transactions. There were some hours where I'd do 100,000 'transactions' an hour that I didnt mean to do, and werent accomplishing anything, just because of poor coding.

[waytogo78247]

I am getting there, eventually, to understand this.

Rejected bets?

If I have a bet in a market that is looking for something specific and that has not been met is that a rejected bet?

Ex1

I have a horse in a market that I do not want to bet til 20 mins to post time but it is in Guardian and that market will be refreshing all day.

Ex2

I have a horse where I want a minimum of 4/1. The same that it is in gaurdian and refreshing all day.

Ex3

I have a horse that I want to bet as soon as the book% hits 105% (or lower). This one could have been refreshing all night.

Are all or any of these classed as rejected bets as conditions were not met?

That would lead to a lot in one day.

[Anbell]

I am getting there, eventually, to understand this.

Rejected bets?

If I have a bet in a market that is looking for something specific and that has not been met is that a rejected bet?

Ex1

I have a horse in a market that I do not want to bet til 20 mins to post time but it is in Guardian and that market will be refreshing all day.

Ex2

I have a horse where I want a minimum of 4/1. The same that it is in gaurdian and refreshing all day.

Ex3

I have a horse that I want to bet as soon as the book% hits 105% (or lower). This one could have been refreshing all night.

Are all or any of these classed as rejected bets as conditions were not met?

That would lead to a lot in one day.

I think that you generally dont need to worry about the transaction charge. It affects fewer than 1% of accounts. This should be the least of your concerns, and if you get a bill it probably wont be for very much, and if you are a limit case you can generally get a reprieve on your first week of charges.

[waytogo78247]

Thank You Anbell you have been a great help and I have a much better understanding now than I have had in the last 15+ years.

[xtrader16]

I wouldn't expect your 'Practise' trading to be replicated 100% once you go on the "live" version of the software. It just doesnt work like that. You might have a stake of £300 that gets matched in practise quite easily but in reality it might not get matched at all in live trading.

My advice to you is, go into live trading as quick as you can and use £10 stakes and work it upwards to £300 but dont expect your practice trading to be replicated in Live trading conditions. Practise is there, I feel, to run stategies and get used to the software/automation rather than see how much you can win in a one off trade.

Good Luck.