Smokin Joe Posted March 30, 2009 Share Posted March 30, 2009 Re: Not a system or strategy exactly but rather more help requested in developing one SJ - when are you counting from ? A1 is counting back a couple of seasons as in page I ( as i understand it ) OK. I understand now. I am looking at each season in isolation. That was my perception of what the author was doing originally (although I may be wrong). Muppet, my figures are the same as yours. Quote Link to comment Share on other sites More sharing options...
a1ehouse Posted March 30, 2009 Share Posted March 30, 2009 Quote Link to comment Share on other sites More sharing options...
Smokin Joe Posted March 30, 2009 Share Posted March 30, 2009 Re: Not a system or strategy exactly but rather more help requested in developing one Forget about it, A1ehouse. I have tested it and the results are poor - a loss of 4.4% over the course of 1130 games this season. Quote Link to comment Share on other sites More sharing options...
Kanga Posted March 30, 2009 Share Posted March 30, 2009 Re: Not a system or strategy exactly but rather more help requested in developing one Damn ... sorry to waste your time guys ...was getting a bit hopeful there. Quote Link to comment Share on other sites More sharing options...
a1ehouse Posted March 30, 2009 Share Posted March 30, 2009 Re: Not a system or strategy exactly but rather more help requested in developing one Our survey says [url=http://www.jinglemad.com/e107_files/public/1193322240_6527_FT50091_wrong_answer.mp3]http://www.jinglemad.com/e107_files/public/1193322240_6527_FT50091_wrong_answer.mp3 Interesting discussion tho! Kanga - these things have to be explored and I'd rather it before the hard earned is placed :rollin. Quote Link to comment Share on other sites More sharing options...
muppet77 Posted March 30, 2009 Share Posted March 30, 2009 Re: Not a system or strategy exactly but rather more help requested in developing one Great collaborative investigation. This is what this forum is about and too much individual work is not always the best way. I have certainly taught myself a thing or two in excel that can be used again. Quote Link to comment Share on other sites More sharing options...
Smokin Joe Posted March 30, 2009 Share Posted March 30, 2009 Re: Not a system or strategy exactly but rather more help requested in developing one Great collaborative investigation. This is what this forum is about and too much individual work is not always the best way. I have certainly taught myself a thing or two in excel that can be used again. Agree. Anyone any other ideas for backtesting? :) Quote Link to comment Share on other sites More sharing options...
a1ehouse Posted March 30, 2009 Share Posted March 30, 2009 Quote Link to comment Share on other sites More sharing options...
Matthew Posted March 30, 2009 Share Posted March 30, 2009 Re: Not a system or strategy exactly but rather more help requested in developing one Good collaborative work here........not something people have seemed keen on in the past. I think you should be coming up with your own predictor variables rather than trying backtesting a load of tosh someone found on the internet somewhere. I went for a complicated mix. Quote Link to comment Share on other sites More sharing options...
happygooner Posted March 30, 2009 Author Share Posted March 30, 2009 Re: Not a system or strategy exactly but rather more help requested in developing one From a laying perspective the English leagues appear to be a nightmare with so much inconsistency that barely one choice of an entire Saturday card makes it into the selection criteria I am currently looking at. Quote Link to comment Share on other sites More sharing options...
muppet77 Posted March 31, 2009 Share Posted March 31, 2009 Re: Not a system or strategy exactly but rather more help requested in developing one just ran 3 seasons into the current one , looking at a season's worth of 'form' as suggested. ignoring the first season, the others had a yield of 0%. the final season individualy had a yield of -7%. when cutoffs were set to <0 and >0 the yield again was 0%. again i suspect that the number of goals being strongly correlated to points does not help the model. Quote Link to comment Share on other sites More sharing options...
a1ehouse Posted March 31, 2009 Share Posted March 31, 2009 Re: Not a system or strategy exactly but rather more help requested in developing one just ran 3 seasons into the current one , looking at a season's worth of 'form' as suggested. ignoring the first season, the others had a yield of 0%. the final season individualy had a yield of -7%. when cutoffs were set to <0 and >0 the yield again was 0%. again i suspect that the number of goals being strongly correlated to points does not help the model. Basically you'll be looking at every single game. I did a bit more digging today, what does your back testing look like when you restrict the odds to between 1.4 and 2.1? Quote Link to comment Share on other sites More sharing options...
muppet77 Posted April 1, 2009 Share Posted April 1, 2009 Re: Not a system or strategy exactly but rather more help requested in developing one -7% YIELD with the cutoffs and odds criteria this season, and -3% yield when cutoffs are 0 and 0. :sad Quote Link to comment Share on other sites More sharing options...
happygooner Posted April 24, 2009 Author Share Posted April 24, 2009 Re: Not a system or strategy exactly but rather more help requested in developing one Sound logic being the cornerstone of a good system let me just run something past the forum to hopefully back up what I am thinking. A complicated combination of both, no matter how complicated, seems to be the best way. So... If a dominant home team in past h2h encounters (lost no more than 1 in last 6 encounters at the same ground) actually lost the last h2h battle (won or drew previous 5) then my thinking is that logic suggests they will be fired up next time around to exact some sort of revenge or at least put in a solid performance. If a dominant home team in last 6 games this season loses their last home game to make it 1 loss in last 6 then logic suggests they lose their air of invincibility at that ground and that teams will be thinking we can do this...instead of....oh no we're going to get killed. By the same token if a crappy away side wins their last to go 1 win in last 6 they will instantly be buoyed by this and go into the next encounter thinking they can continue where they left off previously no matter the opposition. Ie in the second two examples there is more chance of an upset than the first. Agreed? Essentially what research I have done so far shows aspects of recent home/away form and past h2h home/away encounters to be equally important but......perhaps the parametres of when unexpected losses occur and when teams should be left alone/avoided are slightly reduced when looking at h2h as opposed to recent form against all teams in the league. Am I making sense? You know how when you've already decided something in essence but just want somebody to back your decision up:lol Thoughts and opinions most welcome:unsure Quote Link to comment Share on other sites More sharing options...
happygooner Posted April 25, 2009 Author Share Posted April 25, 2009 Re: Not a system or strategy exactly but rather more help requested in developing one Bumpety bump bump bump I'm off to London today but would be good to get some thoughts back on the above post when I return:ok Quote Link to comment Share on other sites More sharing options...
Sunshine Avenue Posted April 30, 2009 Share Posted April 30, 2009 Re: Not a system or strategy exactly but rather more help requested in developing one The big problem is getting a favorable price, or we just end up in side Betfair or bookie 'margin'. Getting hold of late breaking information, before the betting public, is the area to add on to your research. ie team news, players rested, sick, late injury to stars, particularly Goalie, center back, doesn't affect Premiership much, but it does the rest. Is ref a compulsive whistle blower ? if so 6-7 mins less play per game. We have to think unconventionly to gain an 'edge' Quote Link to comment Share on other sites More sharing options...
happygooner Posted April 30, 2009 Author Share Posted April 30, 2009 Re: Not a system or strategy exactly but rather more help requested in developing one The big problem is getting a favorable price, or we just end up in side Betfair or bookie 'margin'. Getting hold of late breaking information, before the betting public, is the area to add on to your research. ie team news, players rested, sick, late injury to stars, particularly Goalie, center back, doesn't affect Premiership much, but it does the rest. Is ref a compulsive whistle blower ? if so 6-7 mins less play per game. We have to think unconventionly to gain an 'edge' Yeah but in trying to produce a system that works right across the board in leagues we know nothing about and with teams we know nothing about this just isn't feasible. Yet I do believe the potential is there for us to work on nothing but stats. 2 months testing has proved favourable and wherever else I have looked it is mostly good. Only a couple of areas of uncertainty and this h2h business is one of them.:unsure Quote Link to comment Share on other sites More sharing options...
.thumb.JPG.8c5a15b710f7a4188ce438df64354c1e.JPG)
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.