A major factor in my development of these systems stemmed from a growing sense that far too many people are losing on Betfair and only a tiny percentage of people are making massive profits.
There is a growing feeling that professionals are taking advantage of less successful punters in the same way that Bookies have been doing for decades.
My key objective was to develop systematic strategies that would enable the intermediate and advanced punter to dramatically improve their betting experience and start to make their hobby increasingly profitable.
A true redistribution of the winning potential that should be available to everyone through Betfair.
How far a punter takes this experience and whether or not it changes their lifestyle is totally up to the individual. What you put in is what you get out. No one in their right mind should believe claims of instant wealth via any Betfair system.
Intermediate Betfair punters will have figured out by now that there is NO “system” that will work all the time. Even less is there a “system” which will work all the time, and can also be automated through computer programs.
Unfortunately as long as Betfair exists, which will be for a long time to come, new and more outrageous “systems” will be churned out to supply the demand created by exasperated and frustrated punters.
What is undoubtedly true is that if you are a form enthusiast then there is certainly value to be obtained by specializing in a specific type of racing and building good foundation knowledge of form and racing patterns.
What is also true is that through careful pre race technical analysis and a thorough staking plan, that using powerful computer systems and software a punter is now able to swing the advantage strongly in their favour.
Way before Betfair appeared on the scene, professionals have been looking for realistic systems that they could incorporate to automate their betting.
The problem with betting shops was that up until 2002 tax was charged on every bet placed. It was impossible to “build” a position and the betting shops would quickly become suspicious of people attempting to work systems and make their wagers profitable. Heaven forbid.
When Betfair appeared in 2001, a revolution happened almost overnight.
Betfair only charged commission on net winnings and even more radically it placed punters in direct opposition with each other by offering the facility to bet on a horse to lose.
This was a true innovation which betting professionals and city traders immediately saw as an opportunity to trade on sports events in the same way as they would with day trading and spread betting on financial markets.
By strange coincidence tax on wagers was removed by betting shops as early as 2002 and the tax was switched to become tax on the bookmaker’s profits.
Simple trading, also known as “scalping” or “hedging” on Betfair as such has already been covered extensively.
Most novice punters will quickly get to grips with the theory of trading and I use ladder trading in my work should liabilities on horses need reducing in a Book or Dutch, or generally to increase profitability of a position while time remains pre race.
The feedback I was getting from punters was that trading on the Betfair interface was very difficult on a day to day basis and as a general rule I have to agree. Punters would frequently tell me how they were exhausted trying to make £10 a day profit from basic trading using the Betfair interface.
Steamer identification immediately pre race is, and will remain a difficult and instinctive business.
WOM doesn’t always work. Strange anomalies move prices in different directions as robots, professionals and big hitters attempt to manipulate the market, and of course the average punter without SIS has no idea what is going on in the ring, or as the horses move down to the post.
In 2004 Betfair launched their API facilitating programmers to access the Betfair data in a timely fashion and enabling third party companies to design their own interfaces and automated betting systems
Automated Betting Systems are what you will have frequently heard bandied around online as “bots” (robots)
Bet Angel is a collection of bots and interfaces.
The interfaces cover all the well known aspects of technical analysis including ladder trading.
Advanced Trading Features include:
• one click trading,
• moving averages,
• enhanced interface charting,
• trade calculators,
• stop losses,
• trailing stops,
• offset betting,
• fill or kill orders,
• matched bets,
• saved bets,
• betting logs,
• timers
• a host of other features.
Literally everything a professional day trader would require to cover all aspects of Betfair day trading.
The bots themselves include Bookmaking, Dutching and a “rules” interface to configure Automation rules to trigger events across all the bots and interfaces should certain conditions arise.
As I set about developing a functional specification for the systems I realized that there were two basic requirements that had to be achieved.
1. The systems had to be basic, rule driven and able to be activated in a routine and consistent manner
2. The systems must provide daily betting opportunities that can deliver an achievable percentage daily return on capital.
In designing and testing the systems it soon became apparent that to achieve this I had to use ALL Bet Angel features for trading, bots and automation triggers.
The Betfair API technology itself is called SOAP; it allows access via a host of disparate systems. Programmers can select their language of choice, from java, .net, C++, VB, the list goes on.
Let’s start with a simple home truth with respect to Bots and enhanced interfaces. NEITHER Bots nor interfaces can GUARANTEE profitable betting systems.
Bots are simply automation of systems that you would have performed manually via a standard or enhanced interface and manual input operations.
If the manual system would have failed then by default so will the bot.
As I have already explained, there are NO paper based racing systems which work ALL the time without human intervention, therefore by default there can NEVER be a bot which can fully automate a system that works all the time.
Bots can be built around two basic categories.
1. The first is “Form Study” (fundamentals of the horse and the race itself)
2. The second is “Technical Analysis” (trading price movements and fluctuations in the market)
Most professionals would argue that hypothetically an effective and feasible bot would be one that targets “value” in a race.
For example, if you had the datasets available you could develop a computer system that calculates a predicted SP for all horses in a race based on previous form.
If the Racing Post or Betfair analysis then displayed a significantly lower SP pre race, then it stands to reason that you could configure a simple application that tells your bot to watch this race throughout the day and automatically place a bet should the SP for this horse, in this race, rise to what you consider should be its correct SP.
The problem is that “value” is in itself a subjective concept.
This was proven in 2005 when Betfair ran a test on the results of some of its datasets to ascertain the validity of betting on steamers as oppose to betting on drifters.
The results were somewhat surprising in that although fewer drifters won compared to steamers, in terms of profitability, betting to £1 level stakes, a punter would have made a significant profit by betting on drifting favourites, but would have made a significant loss by betting on steaming favourites.
Where i or even a computer system see value in a horse pre race, another professional punter or computer system configured with different rules may not see the same value.
External factors including :
• the other horses in the race,
• the going,
• the jockey,
• the trainer,
• the course,
• the draw,
• fitness,
• new from owners, stable lads, commentators, pundits
Any of these can render my “value” meaningless to the outcome of the race itself.
An automated bot cannot provide analysis of pre race changes and activity in the ring, all it can do is look at form prior to the day of the event and price fluctuations.
The reason no-one has ever successfully implemented a fully automated form based bot is that there are too many factors that can change pre race in respect of
• non runners,
• injuries,
• jockey changes,
• changes of conditions on the track,
• commentators and pundits,
• other external news and so on,
Any of these in reality would prevent any professional punter in the ring from making a bet.
Meanwhile an automated “value” system would place the bet for you regardless.
So we can see that attempts at full automation of racing systems have so far ended up in the pile labelled GIGO (Garbage In Garbage Out).
Similarly, automated Martingale strategies where stakes are increased after each loss, in the knowledge that eventually a horse will win are doomed to failure once a run of 5 or 6 failures is encountered, the stakes required are either unattainable or your pot will have deceased.
In Part 2 I continue with a deeper analysis of how my systems developed from specifications through design and into testing.
Mike J Davies is a Computer Analyst, LSE Day trader, Betfair Day Trader and Advisor and developer of
