- This topic has 78 replies, 10 voices, and was last updated 18 years, 10 months ago by
.
-
Posts
-
In tonight’s Championship fixture, Burnley(+7) are slightly favoured, so I would have to be interested in Skybet’s offer of 9/4(if you can get on). They should be no more than  7/4.
Recommended bet: Burnley to win 20pts @9/4 with Skybet.
 Result: 1-1, loss 20pts
Bank now 1,295 pts (295pts profit)<br>Edited by Artemis at 9:08 am on Feb. 16, 2005)<br>
(Edited by Artemis at 5:09 pm on Feb. 18, 2005)
Artemis,
I’ve just had a look at this in RFO.
To help me, what’s your thinking on Hartlepool/Luton – ie what price would you be looking for?
Thanks<br>
tooting,
I normally just do the Premiership and The Championship ratings, except for matches that are on the pools coupons, although there is no reason why the ratings should not work equally well in the lower leagues. Some might argue that the teams are not as consistent in these leagues: they have smaller squads and the absence of key players is felt more keenly than it would be in the highest divisions. At present, I am monitoring these lower league matches, but not recommending any bets.
Hartlepool are (+58), which translates as 5/6, so the current odds of between 5/4 and 7/5 are generous.<br> Hartlepool 2 Luton 3<br>(Edited by Artemis at 4:44 pm on Feb. 15, 2005)<br>
(Edited by Artemis at 9:09 am on Feb. 16, 2005)
Thanks Artemis,
makes sense now – I thought I’d done the ratings wrong – and kept staring at the Hartlepool price!
I think I may play around with this myself. Happy to share any info (by PM I guess).
Cheers,
tooting
I’d certainly be interested in any ratings formula that you devise. It’s quite likely that you might want to give more or less weight to each of the three main inputs<br>i.e. Master Rating(M), Home rating(H), Away rating (A) and Trend(T).<br>PM will be fine, if you don’t want to go public, but there’s not many people who look at this thread. <br>
(Edited by Artemis at 12:55 pm on Feb. 16, 2005)
Hey guys, I look at the thread and have been playing around with a few theories so don’t leave me out ;)
Des
Not much good tonight!
I have my own methodology for the Premiership that I’m comfortable with.
What I’m quite keen to look at is how your approach (or adapted form thereof) may perform in Cup ties, lower leagues and in the European Leagues – ie in matches where I have no real knowledge. I’ve set up a spreadhseet to monitor for rest of this season, plus a small r&d fund. Not sure I’ll really have the time to do it justice but I’ll start off with good intentions!
Happy to keep posting here (just didn’t want to hi-jack your thread with boring thoughts/questions!).
I’m hoping the thread promotes discussion and widens my knowledge of football ratings and betting, so I’m quite happy to give my views. It’s always worthwhile to hear other opinions, even though I might disagree.
It will be interesting to see how your r&d fund performs on matches outside the Premiership & Championship.<br>I think you have to be careful with cup ties in Europe, especially where there are two legs: I’m a bit happier when it’s a mini-league situation. The FA cup is fairly reliable but the League Cup may involve clubs fielding ‘reserve’ sides because they have no real interest in it. As a general rule, I would avoid cup games because the ratings are compiled strictly from league form only. International ratings are compiled only from competitive games, of which there are relatively few, so they may not be as reliable as domestic ratings.
What do you base your own ratings on (basically what info do you use to work them out)? I’m also interested in<br> the European Leagues, because they seem to offer better value, and have all the results but just wonder if the same principals would apply to England and Europe.
<br>Des<br>
The two things I like most about the RFO ratings at first sight are:
1. The over-lapping nature of the divisional ratings. Having watched Fulham rise from bottom of the lowest division to the Premiership in 6 years, I think I gained a decent feel for this. It’s provided quite a few decent bets in cup football. To see it documented in handicap ratings is very interesting and is an area I will concentrate on. (Artemis, of course your warnings re cup ties generally are well made).
2. The universality of them. Anyone brave or mad enough to try to rate across nations gets my attentions.
I’m keen to exploit both the above. As you say Artemis both are riskier propositions, but that’s what attracts me to them – they have much more scope for pricing errors, (including our own of course!).
<br>We’re also nearing the stage of the season where that old chestnut "they have to win" sides are up against "nothing to play for" sides, where the betting and weight of money can lead to strange prices. I’d expect your approach to be a dynamite at exploiting these.
Des911,
I don’t do my own ratings anymore because the RFO Index is very similar to the method I used to use.
The main difference between the Index and the league tables is that each game is rated on the result AND the strength of the opposition. The league tables are points and goals only.
I think the future of ratings may well lie in using more stats from each game such as goal attempts etc., to get a clearer picture of the relative abilities of two teams. I’ve always thought that a simple scoreline can be very deceptive. To illustrate this point, here is an extract of a post I made on the Brokendice website a few years ago:
<br>On the broader subject of handicapping football teams, I have a few ideas that might interest a few people.
About five years ago, I devised a method of handicapping teams by awarding between 3 points(for an unexpected comfortable win) and minus 3 points for an unexpectedheavy defeat. Between these two extremes, points were awarded according to the difference between the expected result and the actual result.
At the start of the season, I rated all teams in each individual division on a scale between 10 and 20 according to where they finished the previous season and how the market rated their chances for the coming season. <br>By the end of the season, the top teams were on a mark of around 25 and the bottom teams were on about 7. I only took league fixtures into account.
I did this for a few seasons and used the predictions mainly for the pools. <br>I didn’t win anything, but had quite a bit of fun doing it. I came to realise how difficult rating football teams can be – especially as only one statistic from the game is used – goals scored.
I watched many games on Sky and often believed that the final score was often a poor reflection on what happened in the game. I know that this is how it has to be: it makes the game exciting and unpredictable and I wouldn’t change it. However as a means of predicting what might happen in the future, the bare result is inadequate.
I feel that what’s required is some system of scoring notional points for attacking play, and using these notional points to decide which team really deserved to be rated the better side.
A suggested points system would be: <br>a goal from open play……10pts <br>a converted penalty/dead ball kick….7 points <br>an own goal……5 points <br>a shot/header on target…..3pts <br>a shot/header off target…..2pts <br>a missed penalty….2 or 3 pts <br>a goal disallowed for offside…..3pts <br>hitting the woodwork……3pts <br>forcing a corner………1 pt <br>awarded a free kick in opponents half…..1pt
At the end of the game, the points scored might be of the order of say <br>Man Utd 67 West Brom 21, whereas the score might be 2-0.
For rating purposes, I would then award 100 points to every game played,and apportion those points in the ratio to how each team performed in terms of notional points.
Man Utd would get 67/(67+21) share of the 100 points and West Brom would get what remains. <br>i.e Man Utd would get 67/88 of the 10 points……76 points <br>West Brom would get 100-76 =24 points
The ‘form book’ would show the game with the appropriate statistics and the notional points and the rating out of 100 for each team for that game.
It would need someone at every game to keep score and some agency to collect the information and publish it. I’m sure it would be a very useful tool, which could be analysed in different ways. There would still be strange results in terms of the score, but the other stats might show if a team was genuinely unlucky.
-end of posting——–
tooting,
The end of season games will probably throw up some excellent value bets. Kevin Pullein in the Racing Post, who is always worth listening to, has proved that the odds-compilers almost always over-estimate the ‘nothing to play for’ effect, whereas the ratings will ignore it.
Artemis, do you go in blind and simply keep to the ratings as your guide?<br>If you do, i feel you could better your results by factoring in some crucial info.<br>It doesnt take long to find out injuries and other circumstances which can effect the outcome of a match.
I tend to put more emphasis on my personal x-factors than any system I have come across when producing bets. That goes for racing aswell.
At the moment I’m just relying on ratings, although I can see the benefits of making adjustments for personal x-factors. I’m always trying to devise objective methods that work and then perhaps try to improve them by bringing in more subjective factors at a later stage.
The validity of football ratings can be more questionable than horse race ratings because it is a team game and teams change, often markedly, from game to game. Perhaps this is why football matches are notoriously unpredictable, not to mention that the best team doesn’t always win.
Even if I had all the information available about team changes etc., I’m not sure I could quantify it so that I would be confident I was obtaining good value. I suppose the only way to do this is to see if you make a profit on the season.
I’ve worked out the ratings for this weekend’s games and extracted the teams who represent good value in terms of the formula I use.
They are:
Team(Rating)       Ratings Odds     Best Odds ÂÂÂ
Preston (+42)      6/5            13/8 drew 1-1<br>Man Utd (+88)      4/6             10/11<br> won 2-0<br>Sunderland(+50)    1/1            5/4<br> lost 1-2<br>Blackpool(+22)      11/8           5/2<br> lost 0-1<br>Hartlepool(+48)     1/1             13/8<br> won 1-0<br>Darlington (+7)     6/4             9/4<br> drew 1-1<br>Alloa  (+19)        11/8            5/2<br> [won 4-0/color]<br>Chelsea(+89)      4/6             10/11<br> lost 0-1
I have given ratings for the FA Cup matches, but would be unlikely to bet in those matches. Also, some of the games are in the lower divisions, so at this stage I will be concentrating on the matches in the Championship involving Preston on Friday, and Sunderland on Saturday for recommended bets.  <br>
(Edited by Artemis at 6:19 pm on Feb. 20, 2005)
Tonight’s televised match:
Recommended bet:
Preston to Win 30pts@ 13/8 (Skybet) draw 1-1, lose 30pts
No other bets for the weekend. The odds for Sunderland to win are now 6/5 at best. Not enough value(on ratings) to merit a bet.
tooting,
Well spotted. That one has slipped under my radar. I have Leeds(+24), so it looks like I will be wanting to Lay Wigan. The adventurous might want to bet on Leeds to win at around 7/2, but I prefer to keep the draw on my side for less reward.
Another recommended bet(thanks to tooting):
LAY Wigan to lose 50pts@1.98 (Betfair) Wigan won 3-0, lose 50pts
<br>Bank now 1,295 less 80 pts lost over the weekend = 1,215pts(215pts profit
(Edited by Artemis at 10:11 am on Feb. 18, 2005)<br>
(Edited by Artemis at 6:26 pm on Feb. 20, 2005)
Like you Artemis I find Alex Deacons ratings invaluable when looking for soccer bets. I also like the draw on my side so am a layer rather than a backer. For what is worth my lays today are:
OXFORD UNITED<br>CAMBRIDGE<br>SOUTHEND<br>RUSHDEN<br>DUNFERMLINE<br>RAITH<br>ROSS COUNTY<br>EAST STIRLING
- You must be logged in to reply to this topic.