Arbusers

Hi there,

I have been following this forum for the past 8 years but I have been pretty inactive over that period. I started arbing around 10 years ago and to begin with made some extra pocket money each week which I was really pleased with. Over time I started to make a little bit more and got to a point where it was almost a part time income. Roughly 6 moths ago after seeing 'Value Betting' popping up again and again I paid some more attention to this and after thorough research decided to give it a go. I found this to be substantially more profitable and my accounts tended, on average, to last a lot longer before being shut or severely limited. This being said, after quitting full time work to gamble as a living, over the last four months I have found that eventually they do still get limited and shut because at the end of the day you are always betting on value and over the long run will end up in profit on the accounts.

Whilst disappointed that this couldn't be converted into a full time living as it was, it got me thinking about how bookmakers generate the odds and whether or not I could try and do this better than them. If you are consistently betting on the one bookmaker who is out of line and offering value it is fairly easy for them to highlight you as a pro gambler, however, if you find prices where the whole market is out rather than just one bookmaker this will certainly go noticed.

For example lets imagine across the board bookmakers are offering the following in a two way market (darts, tennis, snooker)....

Player A - 3.00
Player B - 1.36

Now lets say I have calculated the odds for player A should be 2.25 and therefore value could be had at a whole range of different bookies. Bets could be split across numerous bookies to avoid detection but also I imagine they wouldn't be to bothered about taking larger bets anyway as they are in line with other bookies and can always lay off some liability elsewhere.

In a nutshell IF you can generate betting starting odds you can make profit and have a much larger chance of making it sustainable.

With this info I decided to look into different ways of generating odds and creating predictive models. At the end of the day all it is is data (and lots of it) and how you use this data to generate odds and probabilities. A system that really took my fancy was the 'Elo Rating System' (worth a read if you arn't familiar with it).

I have picked a sport in which I feel it is popular enough to be allowed large bets but not too mainstream so that bookmakers will have super efficient predictive models and I have now gathered past results in the sport for the last 8 years (10,000+ matches). I have it all in a spreadsheet but I am struggling to convert this data into a Ranking system based on the Elo Rating System.

I was hoping someone out there could offer some guidance and if not I would also like to here your thoughts on what I am trying to do and whether or not anyone else on here is going down a similar path as I know some also gamble full time for a living.

Cheers,
Andy   

You will try to be beating people better than you, more experienced than you, and with inside knowledge of team selection, injuries etc

Plus even soft books (Well labs, 355s etc) are not that clueless when pricing up most popular events, and taking an out of line early price is the easiest way to get limited in a flash.

I wish you every success but dont bet the farm on it quickly!

dealer wins wrote: You will try to be beating people better than you, more experienced than you, and with inside knowledge of team selection, injuries etc

Plus even soft books (Well labs, 355s etc) are not that clueless when pricing up most popular events, and taking an out of line early price is the easiest way to get limited in a flash.

I wish you every success but dont bet the farm on it quickly!

Thanks for taking the time to reply,

In response, I think that it is this sort of negative thinking that holds many people back from really making serious money in this industry. All the stats that are used by bookmakers to generate odds are available to the general public if you look hard enough/are willing to pay. I would advise reading the book by Michael Konik, The Smart Money, which documents a journalist working for a professional gambler in Vegas. Basically he has developed a predictive model that is slightly better than the bookmakers in NFL and over the years make millions of pounds from his small edge.

I have selected a sport where there is no team selection, injury (to an extent) to factor in. It is Player A v Player B. This I feel makes the statistics slightly more accurate because there are less variables as to why the results ended the way they did.

In regards to the bookmakers in theory you could make the money available at betfair and then other people would take your odds and cover at soft books, so it would be them who have accounts closed whilst you still get your value odds.

Again I would love to hear your response because if nothing else I feel this is a great discussion point.

Cheers.

That's an interesting endeavour
More or less you are trying to beat the bookies at their home
Don't want to be pesimistic but they have the people , they have the know how , they have the means , they have the T&C's in their favour
I wish you luck , you 're gonna need it
What i would propose is to test your system thoroughly first and in case it's a winner use it with reliable bookies/exchanges

Thank you for this post of value Arbermac. I have some comments.

Arbermac wrote: I found this to be substantially more profitable and my accounts tended, on average, to last a lot longer before being shut or severely limited.

I agree that accounts do live (far) longer with Value Betting compared to other ways of sharp action. You mention that your accounts are being shut or severely limited, maybe this is a spot you should be paying more attention and change the way you operate. As you already know we are taking part in a number of Value Betting projects operating both manually and automatically. The shut ratio that we are experiencing is unbelievably low. So far only 1 bookmaker shut the account down in the manual project and absolutely no-one limited us in the automatic.
For this, I would (very generally advise) to stay in the big 3 markets, football, basketball, tennis, and within these markets avoid something that would piss off bookmakers, for example, women basketball Turkey. You mentioned darts, this is a small market IMHO.

Arbermac wrote: Whilst disappointed that this couldn't be converted into a full time living as it was

This is a decision point for you, and I m glad you selected to keep fighting.

Like arbitrage, trading, middling and other sharp action, Value Betting offers several ways to go. You have to find the right way. I understand that many people are reluctant to even try, being stuck in arbitrage. For me arbitrage is an old fashion, it can be done but Value Betting offers more with less. Gambling is evolving.

cortomaltese wrote: That's an interesting endeavour
More or less you are trying to beat the bookies at their home
Don't want to be pesimistic but they have the people , they have the know how , they have the means , they have the T&C's in their favour
I wish you luck , you 're gonna need it
What i would propose is to test your system thoroughly first and in case it's a winner use it with reliable bookies/exchanges

Thanks for the reply.

Basically yes, and many people do this and profit from it handsomely.

As well as the 10,000+ matches I also have the starting price of each player from Bet365. This means I can back test as much as I need until i've refined a predictive model that provides a consistant edge. The major problem is, as I mentioned above, getting the excel sheet to provide an initial rating system which can then be played with.

Cheers, If I do manage to get it moving forward I will update on the progress.

I agree that accounts do live (far) longer with Value Betting compared to other ways of sharp action. You mention that your accounts are being shut or severely limited, maybe this is a spot you should be paying more attention and change the way you operate. As you already know we are taking part in a number of Value Betting projects operating both manually and automatically. The shut ratio that we are experiencing is unbelievably low. So far only 1 bookmaker shut the account down in the manual project and absolutely no-one limited us in the automatic.
For this, I would (very generally advise) to stay in the big 3 markets, football, basketball, tennis, and within these markets avoid something that would piss off bookmakers, for example, women basketball Turkey. You mentioned darts, this is a small market IMHO.




Thanks for the reply.

I have been following the Value Betting thread and it seems to be going fantastically at the minute. I would be very interested to know which bookmakers you are using because regardless of the tactics used if you are a consistent winner on an account over a sustained period your account will be limited/closed.

I may have missed this but how are the bets selected for the Value Betting System, is it a case of backing odds out of line with the market or based more on statistical analysis? I will have another good read over it all when I get the chance.

Cheers.

More or less, all of your questions are already answered in the relevant thread, but let me repeat some things here.

O4S is manually placing the bets on our behalf. This project is based on statistical analysis. I m not allowed by O4S to reveal the strategies used. For this project we use the mainstream high street bookmakers, let me describe them as the big 7-8 so you all understand the names, together with an orbitx and Matchbook accounts provided by our supported agents. In addition, we use AsianOdds provided by AsianConnect88, but all major tools of this kind would do the job.
We now start to experiment (again) with Value Betting through only sharp bookmakers, Pinnacle, SBObet, Matchbook, Betfair. We need more time to have accurate results but so far so good.

We use a bot for Value Betting, currently operating with only 4 bookmakers. I prefer not to mention their names, but hey, these are 4 well known UK high street bookmakers. This bot is comparing the odds and automatically selecting bets when Value is detected according to certain mathematical criteria. We are pushing the bot developer to go public. Many of our members already know which bot I m talking about.

Outside the ABVI, we are using another bot that places Value bets in almost all well known UK, Malta, Gibraltar bookmakers. I receive the reports and I know the exact filtering of this project but I the developers of this bot are not allowing me to reveal them. This bot is not scrutinised by the software company that built it and it is about to go public. I do not know when this is going to happen as there is a lot of marketing enabled in this and everything must be done by the book. We are offering our insight and we take part in brainstorming for the development. We are pushing towards a soft version of the bot that will allow more people to jump in, instead of a hard core version that would wake up bookmakers.

Arbermac wrote: ...... convert this data into a Ranking system based on the Elo Rating System.

That was baked into prices 20+ years ago, unfortunately.

With the advance of data and predictive algorithms, what's being used today is an order of magnitude more sophisticated. For example, my models use about 1200 variables. For betting based on 3d positioning data, you could easily use hundreds of thousands.

If your system goes tits up, be consoled that they wrote great music. 

Chipmunk wrote:

Arbermac wrote: ...... convert this data into a Ranking system based on the Elo Rating System.

That was baked into prices 20+ years ago, unfortunately.

With the advance of data and predictive algorithms, what's being used today is an order of magnitude more sophisticated. For example, my models use about 1200 variables. For betting based on 3d positioning data, you could easily use hundreds of thousands.

Thanks for commenting.

Sometimes complexity is not always the answer. May I ask which sport you use models for and what success you have?

I feel that mainstream sports are very hard to beat but some of the less popular ones are definitely there for the taking. Theres a reason why when bookmakers bring out opening lines sometimes there is big variance in prices and this reason is that they use different methods to generate the prices. One of them will be producing more accurate odds than the other.