Jump to content

A.I. Football ~ 27% yield


Recommended Posts

Re: A.I. Football Thanks for the reply Machine. I must admit that I am surprised that you use only odds. One thing I disagree on is that past results have no bearing on present results. If that were so then all we need do is use a pin. And of course the odds you use are the result of an analysis of the standard stuff and your choice of teams are the result of an analysis of past odds if you are using an AI programme. I must admit the scheme I have just started is simply based on odds as detailed study of other factors seems to get me nowhere. May your good run continue

Link to comment
Share on other sites

  • Replies 111
  • Created
  • Last Reply

Top Posters In This Topic

Re: A.I. Football

5 draws for the weekend blackburn v norwich 3.7 @ betvictor oldham v preston 3.4 @ boylesports leyton o v yeovil 3.4 @ betvictor portsmouth v derby 3.6 @ boylesports port vale v bristol rovers 3.5 @ boylesports 5*10pt stakes
Not a good as last week but a profit all the same ;) oldham v preston 3.4 @ boylesports won leyton o v yeovil 3.4 @ betvictor won staked 50 returned 68 profit 18 overall staked 950 lost 61 won 34 return 1171 profit 221 yield 23.26% :ok
Link to comment
Share on other sites

Re: A.I. Football

4 draws for the weekend all @ Betvictor west brom v villa @ 3.5 bury v oldham @ 3.5 bristol r v accrington @ 3.4 kidderminster v mansfield @ 3.5 The usual 10pt singles :ok
Just the 2 draws in west brom & bury Staked 40 returned 70 profit 30 overall staked 990 lost 63 won 36 return 1241 profit 251 yield 25.35% :ok
Link to comment
Share on other sites

Re: A.I. Football

just the 1 this weekend Tranmere v Scunthorpe 3.6 @ Boylesports Usual 10pts :ok
Well looks like a good ending to the season with a win. Staked 10 returned 36 profit 26 overall staked 1000 lost 63 won 37 return 1277 profit 277 yield 27.7% 'He who enjoys doing and enjoys what he has done is happy' Thread closed :) :ok

Link to comment
Share on other sites

  • 6 months later...

Re: A.I. Football

Hi methuselah Quite the contrary I only use 1 variable....... 'ODDS' They say 98% of punters lose and why is that? Because as you say they are all using the 'standard stuff' (past data) and making there ratings doing the same regression graph or trying to find win %ages through poisson distribution,form reading etc etc. Why should a collection of past results have any bearing on present results, 'you walk the same path you end up in the same place', It didnt work years ago so why would it start to work now, yet you still see thread after thread trying it,with most reaching the same inevitable conclusion. The only thing you know that is 100% nailed on in advance before any event are the 'ODDS' on offer, all the works been done for you by the bookies. Its not the 'holy grail' nothing is but I personally find using the actual odds as a datum and looking for outliers gets me closer to that elusive profit than inventing odds using rating/poisson %ages,or form reading. Plus Im a lazy punter so I have automate everything so I only have to click a mouse,no boring form reading,multiple data collecting etc... simple is good :lol I may be wrong but I believe Slapdash uses a similar method in his BOG trotting threads, using the betfair market as a datum to pinpoint possible outliers in the BOG markets, and to a point Aidymac using W.O.M (weight of money) as his datum to pinpoint possible mistakes in the market (though Im not a fan of W.O.M as it can be misleading as small amounts of money can have a massive impact on the odds in weak markets) If thats the case then there does seem to be more mileage in this type of approach than the ratings/poisson form reading etc. Cue Irate ratings,poisson, form reading value punters demanding Im hung drawn and quartered then burnt as a heretic :lol :ok
This was exactly what I was talking about and you disagreed......I take it that you have failed on this type of betting now since you started to disagree with me on it.
Link to comment
Share on other sites

Re: A.I. Football

This was exactly what I was talking about and you disagreed......I take it that you have failed on this type of betting now since you started to disagree with me on it.
You can't argue with the success of the bets on this thread though - 27.7% yield from 100 bets by the looks of it :clap
Link to comment
Share on other sites

  • 2 weeks later...

Re: A.I. Football B101 just noticed this

This was exactly what I was talking about and you disagreed
No it wasnt you were talking about value betting... estimating a probability of a team winning then seeing if you can get that price estimation from a bookie. Totaly the opposite of how I bet on football. I use 1 thing and 1 thing only the odds given BEFORE the game by the bookie Once that is input into the NN prog it tells me all I need to know (well it did till my prog went belly up). No ratings, no form analysis, no looking to find made up prices.
.I take it that you have failed on this type of betting now since you started to disagree with me on it.
See above
He told me this was not possible
No I didnt ,I disagreed with the concept of value betting in the thread you put up.
like a football match between England and Brazil, you can only guess on the probability. There are too many factors involved to make a precise objection prediction possible.
The highlighted text from that thread just emphasizes the nature of value betting, a total guess, all be it an educated guess but at the end of the day a guess is all it is. Yet punters still go on about it in the pretense that they know exactly what they are on about which I consider bullsh*it and I will say as much
Link to comment
Share on other sites

Re: A.I. Football ~ 27% yield

The bookmakers only have human beings to compile their odds - with a bit of mathematical help.There are more people on the other side of the fence trying to outsmart them' date='and some of them are actually up to the task.[/quote'] So your telling me that odds compilers have sat there doing every single set of odds they have on their site :eek the mind boggles :eyes They use WOM to balance their books on the net hence the different odds than the fixed odds coupons and they would need thousands of compilers just to keep up with all the differing amounts of bets coming in from everywhere. Im sure the major books have there own bit of software and an algorithm which sorts it for them in milliseconds, only using compilers when really necessary on certain things. Anyway that wasnt the reason for my post to argue the pros and cons of compilers.
Link to comment
Share on other sites

Re: A.I. Football ~ 27% yield It scrapes all the pre match odds then runs them through a NN prog which matches against previous patterns then checks against a CLT (Central Limit Theory) type prog. The scraper still works its somewhere between the NN output & the CLT input that im getting an error I think, as scraper and NN work ok together and CLT output is ok when used with another prog so it is somewhere in the middle im summising. The whole coding is thrown together to say the least (self taught) and its got more bugs in it than a tramps vest, my lads been messing about with it for a while to try and iron things out and is getting bits and bobs out of it but no real volume. Ive totally lost interest in it too much messing about so I leave it to him. Sad thing is it seemed to do the job while it worked :(

Link to comment
Share on other sites

  • 5 months later...

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


×
×
  • Create New...