AndrewKent Posted February 1, 2011 Share Posted February 1, 2011 Quote Link to comment Share on other sites More sharing options...
davidc Posted February 1, 2011 Share Posted February 1, 2011 Re: Steele's Perm Betting System Nice to see some selections form all your hard work Andy. Good luck. :hope Quote Link to comment Share on other sites More sharing options...
AndrewKent Posted February 1, 2011 Author Share Posted February 1, 2011 Re: Steele's Perm Betting System Thanks David, I really appreciate that. I've enjoyed learning and I'm a bit of a stats geek anyway so it's been enjoyable but yes, it does feel nice to arrive at this stage. I'm well aware that there's still probably a long way to go though! Quote Link to comment Share on other sites More sharing options...
AndrewKent Posted February 1, 2011 Author Share Posted February 1, 2011 Re: Steele's Perm Betting System Wrexham 2 - 1 Altrincham :cheers Manchester Utd 3 - 1 Aston Villa :cheers Watford 1 - 1 Crystal Palace :wall Stockport 1 - 1 Chesterfield :cheers Histon 1 - 0 Fleetwood :wall Current Bank = -£10.00 Okay so the first set of picks lost but I am not too upset. My picks weren't a million miles out and I knew there was still work to be done. I'll give it a few more days of posting this kind of thing up before I review it and see what needs to be changed. Kind regards, Andy Quote Link to comment Share on other sites More sharing options...
clay747 Posted February 1, 2011 Share Posted February 1, 2011 Re: Steele's Perm Betting System better luck next time. If you place a double chance bet on the away teams you can simplify all the bets to just one accumulator bet. Its been a long time since I read the the Paul steel book, how do you convert the power ratings into selections? Quote Link to comment Share on other sites More sharing options...
AndrewKent Posted February 1, 2011 Author Share Posted February 1, 2011 Re: Steele's Perm Betting System Hi Clay, this is the kind of input I am looking for. The next selections I will make, I will look into making the Draw/Away a double chance. I can cover this in depth in special posts but I am happy to cover the power ratings briefly. From the table, you can see the ratings for all three systems, Power Ratings, Game Form and Penetration Plus. The rule base follows thus: BANKERS Filter matches that are >= 5 for Game Form **From the remaining games** Filter matches that are >= 0.5 for Penetration Plus **From the remaining games** Select the 3 highest scoring games for Power Ratings DRAW/AWAY WIN Filter matches that are Quote Link to comment Share on other sites More sharing options...
clay747 Posted February 1, 2011 Share Posted February 1, 2011 Re: Steele's Perm Betting System BANKERS Filter matches that are >= 5 for Game Form **From the remaining games** Filter matches that are >= 0.5 for Penetration Plus **From the remaining games** Select the 3 highest scoring games for Power Ratings do you factor the home win odds in at all? for example, the fourth highest game may only be out by a few power rating points, but is offering much better odds? Quote Link to comment Share on other sites More sharing options...
AndrewKent Posted February 1, 2011 Author Share Posted February 1, 2011 Re: Steele's Perm Betting System I don't but that could be a good idea. Filtering all power ratings over 1 and then selecting the 3 highest odds? Quote Link to comment Share on other sites More sharing options...
clay747 Posted February 1, 2011 Share Posted February 1, 2011 Re: Steele's Perm Betting System I don't but that could be a good idea. Filtering all power ratings over 1 and then selecting the 3 highest odds? You could try that, its common sense that if you have two games with the same rating, you should pick the game that has the highest odds first. so you could expand that to say a margin of %5, and see how that works. the route most people take is to look at the historical data for each rating and try to work out the odds you need to be profitable with a given rating. Monkeys nest gives some good advice on how to do this in your other thread. Quote Link to comment Share on other sites More sharing options...
AndrewKent Posted February 2, 2011 Author Share Posted February 2, 2011 Quote Link to comment Share on other sites More sharing options...
sjt91 Posted February 2, 2011 Share Posted February 2, 2011 Re: Steele's Perm Betting System So, based on this kind of information, rather than picking the 3 highest scoring Power Rating matches as bankers, I should be picking the games where the odds show a good chance of success? Cheers, Andy Well no. Since if you pick those with the lowest odds, then you get a lower payback on the ones you do win. The point of betting is to make money, not to have a high strike rate. If you back at even stakes games with average odds of 1.1 and have a strike rate of 90%, you're still onto a loser. You need to find value. You should use your system to find out fair odds, and then when the odds exceed that, bet. You may choose to do this only within a certain odds range, or only for home teams and so forth, but the key is finding value. This is where the ratings system comes in, and ultimately its accuracy will determine how profitable (or not) you are. Quote Link to comment Share on other sites More sharing options...
ChiChi92 Posted February 2, 2011 Share Posted February 2, 2011 Quote Link to comment Share on other sites More sharing options...
Drallep Posted February 6, 2011 Share Posted February 6, 2011 Re: Steele's Perm Betting System Hi Andrew, Seems a poor Villareal performance and a strange Bayern Munich game scuppered you yesterday. I was just wondering where you collated your 10 years of results from? Quote Link to comment Share on other sites More sharing options...
PAW Posted February 7, 2011 Share Posted February 7, 2011 Re: Steele's Perm Betting System I too operate a system based around Steele's game form/penetration ratings, but I backed Koln (and also Osasuna) so we are obviously interpreting the output very differently Andy, you started another very interesting thread about regression analysis, are you applying any of those findings here? The ratings might highlight the most likely winner but in my view that is not necessarily what you should be putting your money on Quote Link to comment Share on other sites More sharing options...
AndrewKent Posted February 7, 2011 Author Share Posted February 7, 2011 Re: Steele's Perm Betting System Hi guys, Bit of late response on my part. Busy weekends and work have kept me from looking at this system as closely as I had hoped in the last few days but I have a little time right now so I thought I would keep the ball rolling along: Barcelona 3 - 0 Atl. Madrid :cheers Marseille 1 - 0 Arles :cheers Villarreal 0 - 1 Levante :wall Bristol Rovers 2 - 4 Brighton :cheers Köln 3 - 2 Bayern :wall So again, two bets let me down. I think from what I am seeing is that the system is worth following up on but as the feedback I have had suggests, there is still a lot of fine tuning and testing to be done. Drallep - All the data has come from www.football-data.co.uk which I collate into a database. The analysis has been outputs from this that I work on in Excel. PAW - In terms of the regression analysis, I get the concept of it, I have built working models in Excel/Access/VBA that work however I am a bit lost in terms of how best to use it. This is more to do with my own inexperience with football betting. Actually PAW, if you use this system, I would love to pick your brains about how you use Paul's ratings. It could be that have overlooked something which as you said, is impacting the picks that I am making based on how I am interpriting the ratings. Andy Quote Link to comment Share on other sites More sharing options...
PAW Posted February 7, 2011 Share Posted February 7, 2011 Re: Steele's Perm Betting System Ok whilst I think there's some great ideas in that book, IMO it goes wrong in the way the systems are applied. The way Steele does it, well you might as well jsut sort the fixtures in ascending order of odds and back the top 5 So instead I let the odds decide which teams to back, and the 'fair odds' calculations I use were obtained through regression analysis of the game rating/penetration systems (or slight variations of). As for his power ratings, well I don't think they tell you anything more than a quick glance at the league table does. Even though I do still compile them, after that I completely ignore them :) Quote Link to comment Share on other sites More sharing options...
AndrewKent Posted February 7, 2011 Author Share Posted February 7, 2011 Re: Steele's Perm Betting System So you didn't take the power ratings into consideration at all? Bummer really because it was the hardest to automate! So, in terms of a cut-off point for ratings, Steele mentions that anything greater than 5 for Game Form AND greater than 0.5 for Penetration Plus (or -5 and -0.5 if you are looking at the away team) is beneficial. Is this what you look at? Also, as there are two ratings there, how do you set up your regression? How do you combine the two so that you can analyse the 1X2 that have happened with these ratings? I've done regressions with one rating before but never with two working together. I'd love to know what you look at. Andy Quote Link to comment Share on other sites More sharing options...
PAW Posted February 8, 2011 Share Posted February 8, 2011 Re: Steele's Perm Betting System I'm not saying you should give up on the work you've put in, I just couldn't find any edge in using the power ratings myself. To me (especially the way they're used in the book) they just point out the bleeding obvious, and put you onto the sort of selections that are overbet week after week. For the systems I do employ, personally I don't use cut-offs since all my betting is price driven. It doesn't bother me how little chance the ratings give a team of winning, if the price is right then that's who I back As for the regression analysis, my knowledge of stats doesn't extend to anything fancy like combining ratings, so I just came up with a pricing formula for each system. If the available price passes each individual test then it's up for consideration. I hope you continue with this thread, but ultimately (and I think one or two others have alluded to this elsewhere too) I think experimenting with the ideas rather than applying Steele's selection method verbatim will be the way forward Quote Link to comment Share on other sites More sharing options...
AndrewKent Posted February 8, 2011 Author Share Posted February 8, 2011 Re: Steele's Perm Betting System Ah, so having done your regression analysis on your ratings, if the bookies are offering value odds based on said rating, you will consider it? Is there anything specific that you do in your regression analysis that I should know of? Or is it a simple case of looking at the match outcomes for each rating and carrying out the regression analysis from there? Quote Link to comment Share on other sites More sharing options...
AndrewKent Posted February 10, 2011 Author Share Posted February 10, 2011 Re: Steele's Perm Betting System Hi folks, With a little bit more time on my hands today I thought I would post up some of the analysis that I have carried out. I've taken the ratings systems that I have applied to each match and carried out some Resgression Analysis to try and find out what "Value" odds would look like for each rating. I'll start with Game Form. Game Form Without going into massive amount of detail on this, Game Form ratings are points that have been scored in the last 5 games for Wins and Draws, which have been weighted depending it the team is playing at home or away. A high score means a higher chance of a home win, a low score means a higher chance of an away win and a draw would be around 0. I'm not interested in predicting the draw at this stage because I'm finding it quite difficult, but perhaps this is something I can come back to in the future. Now from this, my next step (I guess!) would be to ONLY bet on games where the odds offered by the bookies exceed my value odds. Of course, I have two other ratings systems to take into consideration (which I will carry out Regression Analysis on) but I thought I would share this with you all. Please feel free to comment and tell me what I should be doing next! Andy Quote Link to comment Share on other sites More sharing options...
Mulkis Posted February 10, 2011 Share Posted February 10, 2011 Re: Steele's Perm Betting System Wow! Thats impressive. R2=0.94! Looking forward for ROI%:hope Quote Link to comment Share on other sites More sharing options...
AndrewKent Posted February 11, 2011 Author Share Posted February 11, 2011 Re: Steele's Perm Betting System Okay, Second attempt (my computer crashed whilst I was typing, most annoying :@) I have been thinking on what has been said in this thread, especially the feedback from PAW about the Power Ratings system telling us nothing more than what a quick glance at the League table would tell us. However I still think there is value in the Power Ratings system, so I want to apply it in a different way. The I had a thought: "Isn't an away win against Celtic harder to achieve for Hearts than an away win against Hamilton?" The Game Form system applies weighted points for each previous result (eg. 3pts for a home win and 5pts for an away win). What if we take that and multiply it by the Power Rating of the opposition going into the match? So in my example, If Hearts beat Celtic away, it's going to give them more points than if they beat Hamilton away. Power Ratings are built from Goals Scored/Conceded so by combining these two systems, I'm taking into consideration: Recent Form Seasonal Form Goals Scored/Conceded The proof is in the pudding as they say so I shall set about testing this approach and come back to you all. Does this IN PRINCIPLE sound like a logical approach? Quote Link to comment Share on other sites More sharing options...
Recommended Posts
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.