Vilamoura04 Posted January 7, 2010 Share Posted January 7, 2010 Quote Link to comment Share on other sites More sharing options...
Chris2b Posted January 7, 2010 Share Posted January 7, 2010 Re: Match Ratings -snappy title coming soon hi vilamoura, i follow with interest this original method.;) Why not choose B'gham wich is a big value ! value is value you know Quote Link to comment Share on other sites More sharing options...
Grex Posted January 7, 2010 Share Posted January 7, 2010 Re: Match Ratings -snappy title coming soon I just LOVE rating systems...... what is particularly interesting about your figures is that you are predicting "value" or "near-value" on some of the hot favourites (Arsenal, Chelsea, Inter, Sevilla... even Fulham :loon). Normally rating systems conclude that favourites are poor value. I will be very interested to see how this pans out over the coming weeks. One comment..... I think it would be clearer if you expressed "value" as a percentage, rather than a decimal proportion ;) Quote Link to comment Share on other sites More sharing options...
Vilamoura04 Posted January 7, 2010 Author Share Posted January 7, 2010 Re: Match Ratings -snappy title coming soon hi vilamoura, i follow with interest this original method.;) Why not choose B'gham wich is a big value ! value is value you know Hello mate In the last week since I have been doing these picking the 'likeliest winners' only (and ignoring value) showed a good profit. Following the 'value' selections (and ignoring likeliest winners) showed a loss. The reason for this as Grex alludes to I think is that favourites are often indicated as being poor value. I understand the importance of value. But I also want a reasonable strike rate, I dont want to be betting selections I know will only win this fixture once in twelve seasons. Hence why I am hypothesising that mixing the two (both 'value' and 'likely winners') is the way to go. But I have only been doing this for a week which isnt long enough to come to any conclusions yet. We need to see how it goes over the coming months. It may be that the value selections are the ones to go with. Quote Link to comment Share on other sites More sharing options...
Vilamoura04 Posted January 7, 2010 Author Share Posted January 7, 2010 Re: Match Ratings -snappy title coming soon I just LOVE rating systems...... what is particularly interesting about your figures is that you are predicting "value" or "near-value" on some of the hot favourites (Arsenal' date= Chelsea, Inter, Sevilla... even Fulham :loon). Normally rating systems conclude that favourites are poor value. I will be very interested to see how this pans out over the coming weeks. One comment..... I think it would be clearer if you expressed "value" as a percentage, rather than a decimal proportion ;) Fulham are a team I dont trust at odds on. I still remember the Bolton game last month when just about the whole world was on Fulham. Im not sure about the Stoke pick either...but I guess thats the point of ratings systems - to take away the objectivity that can sometimes cloud the judgement. I will look into expressing value as a percentage Quote Link to comment Share on other sites More sharing options...
Vilamoura04 Posted January 7, 2010 Author Share Posted January 7, 2010 Re: Match Ratings -snappy title coming soon I should also point out that I am very suspicious when my odds are wildly different to bookies odds. If this is the case t is almost certainly because my odds are wrong, rather than all of the bookies have made a huge mistake. Quote Link to comment Share on other sites More sharing options...
Staines Passive Posted January 7, 2010 Share Posted January 7, 2010 Re: Match Ratings -snappy title coming soon Hello mate I understand the importance of value. But I also want a reasonable strike rate, I dont want to be betting selections I know will only win this fixture once in twelve seasons. Like the system but completely disagree with this statement. You shouldn't think about strike rate before profit, and if you decide that betting on value brings you the largest profit and believe you have a way of picking the value you have to follow it. And don't think of it like a one in twelve seasons thing, think of it as a one in 12 bets Quote Link to comment Share on other sites More sharing options...
Vilamoura04 Posted January 9, 2010 Author Share Posted January 9, 2010 Re: Match Ratings -snappy title coming soon Noted SP. But there is a conflict here which is a much bigger debate than this thread. These ratings systems all throw up findings that backing teams like Man United and Chelsea are not value bets. Despite this, if you backed Man U and Chelsea over the last few seasons you would be in profit. Would you rather have bet for value and be down, or would you rather be in profit? The trick in all of this is the judgement call in what to bet and what not to bet. Value indicators are only useful as indicators, you still need to keep strike rate healthy. Hence why will only be selecting bets that are both value, and likeliest winners. I note that the ratings threw up a value winner tonight in Forest, a selection which was not considered to be the likeliest winner. Interesting. The bets which so far survive the weather tomorrow are as follows. Does anyone know if the snow and ice has or will be reaching Italy and Spain? I know it was in Lyon yesterday as I work with a guy who was there. Arsenal @ 1.4 Sevilla @ 1.57 Atletico @ 2.75 If Inter can just move out to 1.25 they would be a bet also Quote Link to comment Share on other sites More sharing options...
lunatism Posted January 9, 2010 Share Posted January 9, 2010 Re: Match Ratings -snappy title coming soon Gotta agree with Staines Passive Personally, over the last few seasons, I've been leaning towards a mixture of the two - a system that highlights value, and tagging this to another system that pulls out selections. What it resulted in, was a slightly lower strike rate, but a much higher yield... And....sometimes bookies DO make mistakes Vilamoura04, so have faith in your odds! :ok Quote Link to comment Share on other sites More sharing options...
palaceman Posted January 9, 2010 Share Posted January 9, 2010 Re: Match Ratings -snappy title coming soon These ratings systems all throw up findings that backing teams like Man United and Chelsea are not value bets. Despite this, if you backed Man U and Chelsea over the last few seasons you would be in profit. Then the ratings are wrong. Quote Link to comment Share on other sites More sharing options...
Grex Posted January 9, 2010 Share Posted January 9, 2010 Re: Match Ratings -snappy title coming soon .....Does anyone know if the snow and ice has or will be reaching Italy and Spain? Sevilla @ 1.57 Definitely no snow in Andalucia, Spain (Sevilla) ;) Quote Link to comment Share on other sites More sharing options...
Vilamoura04 Posted January 9, 2010 Author Share Posted January 9, 2010 Quote Link to comment Share on other sites More sharing options...
Vilamoura04 Posted January 9, 2010 Author Share Posted January 9, 2010 Re: Match Ratings -snappy title coming soon Arsenal and sevilla fail at short odds but maybe not too surprising considering their missing players. I wonder whether to adjust the ratings to account for team news, but it's such a black art deciding how much missing players affect a teams chances. I think I may need to take on board Staines views on value, the value bets have done much better than the likely winners in the other games today. Again I am rapidly forming the opinion that you need skill in interpreting these ratings to make them pay. Atletico win at 2.75 though which is a nice win. Assuming 1point level stake bets the system is 0.25 points down on the three bets on day 1. Quote Link to comment Share on other sites More sharing options...
Vilamoura04 Posted January 9, 2010 Author Share Posted January 9, 2010 Quote Link to comment Share on other sites More sharing options...
Staines Passive Posted January 10, 2010 Share Posted January 10, 2010 Re: Match Ratings -snappy title coming soon You system like the look for Mallorca DNB perhaps? Come on man, give us a big price winner, stop playing the safety card :ok Quote Link to comment Share on other sites More sharing options...
Staines Passive Posted January 10, 2010 Share Posted January 10, 2010 Re: Match Ratings -snappy title coming soon You system like the look for Mallorca DNB perhaps? Come on man' date=' give us a big price winner, stop playing the safety card :ok[/quote'] On a side note I notice that Mallorca have won there last 2 games in madrid :eek Quote Link to comment Share on other sites More sharing options...
Vilamoura04 Posted January 10, 2010 Author Share Posted January 10, 2010 Re: Match Ratings -snappy title coming soon Im sure there is value in Mallorca, Im not sure about Madrids rating I think its affected by the 5 games Madrid last at the end of the season last season when there was nothing on the games (one of which was Mallorca). But even if my ratings are out by 20% in favour of Mallorca they should still be value. Dont think they will win though :lol Quote Link to comment Share on other sites More sharing options...
Djordje Posted January 10, 2010 Share Posted January 10, 2010 Re: Match Ratings -snappy title coming soon Dont think they will win though :lol No offense mate, but I think this sentence underlines problem with your selections method. I don't really understand what is your criteria in selecting your bets, but I think you're trying to find value in more or less popular bets, which is basically wrong approach in my opinion. I think you have to find some strict rules for bets selection and stick to it, even when opposing mighty Real Madrid. After 20, 50 or 100 bets you can alter and refine these rules in order to improve your performances. Good luck whatever you decide :ok Quote Link to comment Share on other sites More sharing options...
Staines Passive Posted January 10, 2010 Share Posted January 10, 2010 Re: Match Ratings -snappy title coming soon Agreed - if you trust the system to bet on anything you really have to go with it unless you have a real reason other than you dont expect it to win. My system today threw up and under 2.5 in the Barca game (currently 3-0 at half time) but you have to go with it. Quote Link to comment Share on other sites More sharing options...
Vilamoura04 Posted January 10, 2010 Author Share Posted January 10, 2010 Re: Match Ratings -snappy title coming soon The selection method which I have explained very poorly is as follows Qualifying bets must be BOTH the likeliest outcome AND value. For any given match 1. Look to see which of the three outcomes (home, draw, away) the model thinks has the highest probability. So for Leicester v Ipswich today the model gives Home/Draw/Away percentages of 54%/28%/18%. This tells me that the 'likeliest winner' is Leicester with a home win % of 54%. 2. Look across to the value section to see if the odds of this selection are at least fair odds. So in the case of Leicester if they are 54% likely to win odds must be at least 1.86. The odds were 2.1, which indicates the odds are at least fair. This is therefore a qualifying bet. If the odds were 1.85 or less this would NOT be a qualifying bet. Non-qualifying bets would be anything which is either a. not the most likely outcome OR b. is not indicated as value What I am trying to avoid is longshots like Tenerife which may be a few percentage points of value bets but are so unlikely to win its just not worth the wait. Agreed - if you trust the system to bet on anything you really have to go with it unless you have a real reason other than you dont expect it to win. My system today threw up and under 2.5 in the Barca game (currently 3-0 at half time) but you have to go with it. I did some research into La Liga and the premiership and found that over the last two seasons, blanket betting the unders in any game where the away team is 1.8 or less is a profitable strategy. I did start following this trend but then forgot about it. Quote Link to comment Share on other sites More sharing options...
Vilamoura04 Posted January 10, 2010 Author Share Posted January 10, 2010 Re: Match Ratings -snappy title coming soon sorry duplicate post Quote Link to comment Share on other sites More sharing options...
Vilamoura04 Posted January 10, 2010 Author Share Posted January 10, 2010 Re: Match Ratings -snappy title coming soon OK assuming level stakes betting of 1 point on each: If you had backed the 'most likely winner' ONLY (ie the match outcome with the highest probability) for each match today you would have profited by 4.69 points for 16 points staked. If you had backed the 'Value' ONLY (ie the match outcome which the model indicates the price is too high) you would be 1.12 points up from 16 points staked, mostly thanks to the big priced Rangers draw. If you had backed selections that were BOTH the 'most likely winner' AND 'Value', you would be 1.29 points up from a 10 point stake. It is the final selection strategy that I propose to follow. Quote Link to comment Share on other sites More sharing options...
Grex Posted January 10, 2010 Share Posted January 10, 2010 Re: Match Ratings -snappy title coming soon ........ It is the final selection strategy that I propose to follow. Sorry, I don't quite follow your logic.... the highest yield (over 25%) on today's results was obtained by backing the "most likely winner". The system you propose to follow only produced a yield of 13%..... I realise that we are talking about a VERY small sample only, but....... ????? Quote Link to comment Share on other sites More sharing options...
relf Posted January 11, 2010 Share Posted January 11, 2010 Re: Match Ratings -snappy title coming soon Would be good if you can do back-testing over seasons, which would give you a much better idea. But personally, I think doing back-testing on match ratings seems to be very tedious given that the ratings change after every match - either you need to be a number cruncher, or perhaps, you can formulate in a good program/software. (Alternatively, you can set a few strategies and observe them in the coming weeks to see which one performs better.) Quote Link to comment Share on other sites More sharing options...
Vilamoura04 Posted January 11, 2010 Author Share Posted January 11, 2010 Re: Match Ratings -snappy title coming soon Yeah backtesting would be too painful an exercise to consider. Im going to just watch over the coming weeks. Quote Link to comment Share on other sites More sharing options...
Vilamoura04 Posted January 11, 2010 Author Share Posted January 11, 2010 Re: Match Ratings -snappy title coming soon ProbabilityOddsValueOddsValueOddsValueHWDAWHDAMan City Blackburn0.610.260.131.45-11% 4.824%8.612% Man City most likely winners but value is indicated in the draw and away win. As Man City are the 'most likely winner' but not 'value', this is not a qualifying bet. Quote Link to comment Share on other sites More sharing options...
Staines Passive Posted January 11, 2010 Share Posted January 11, 2010 Re: Match Ratings -snappy title coming soon In this case what about Man City draw no bet? Man City are 1.2 for that and you make them 87% favs? Quote Link to comment Share on other sites More sharing options...
Vilamoura04 Posted January 11, 2010 Author Share Posted January 11, 2010 Re: Match Ratings -snappy title coming soon Technically its 87% for Man City-draw double chance rather than draw no bet, but I know what you mean. Not a bad shout. I will be on. Quote Link to comment Share on other sites More sharing options...
Vilamoura04 Posted January 11, 2010 Author Share Posted January 11, 2010 Quote Link to comment Share on other sites More sharing options...
Grex Posted January 11, 2010 Share Posted January 11, 2010 Re: Match Ratings -snappy title coming soon A friendly word of advice....... for the reasons you allude to in your post, steer clear of cup games ;) 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.