Generally, most football teams perform in line with their recent results history. This means that in general they will tend to lose against better teams, and win against poorer teams. The quality of the teams is reflected by their position in their league, when the season has stabilised and ‘all other things are equal’.
Now, we could take the basic league positions as the guide to form, but this can change on a day to day basis for reasons unrelated to the team itself – for example by the results of other teams. So, we need to have a slightly more sophisticated system of assessing team performance which takes account of recent results (but how recent?). That is the first part.
Then, we need a way of assessing each match in advance to arrive at a likely outcome, ideally being able to put a number to this so that we can compare one match with another and decide which is more likely to be a home win, a draw or an away win. In this way we can arrive at a ranking for each of the 49 matches on a British coupon (which may of course cover Australian football matches during the British summer). That is the second part.
Analysis of the 2009-2010 British football season gives us an idea of what the average outcomes are. Over the whole season (40 pools coupons), 45% of matches were home wins, 26% were away wins, and 27% were draws (score and non-score draws combined).
So, with a team performance measure, a way of comparing matches and the above statistics, we can start to ‘home in’ and where the draws might lie (or, for that matter, the homes and aways, if that is your betting preference).
Overall these are just averages – each week will be different and there will be some unexpected results.
So, to maximise our chances of winning, whether it is the treble chance or fixed odds, we need a method to spread our stakes. We do this using plans or perms, which enable us cover many combinations. After all, to forecast 3 draws from 49 matches on a random basis is quite a long shot (the odds are over 18,000 to 1). In a 10 horse race, you have odds of 10/1 of picking the winner. With fixed odds betting, the bookie will have adjusted the payout odds to account (initially) for the likely outcomes, and the odds will drift depending on the stakes being placed by other punters. So, whilst in practice we could stake say 10 cents per combination, that is a big stake for 18,000 lines and we would not cover it with a win on account of the fixed odds (even if the bookie would take the bet), though we would in all probability have many winning lines if there were say 8 draws in the results.
However, if we were to lay a bet of 3 draws from 10 matches (120 separate bets), or 5 aways from 10 (252 separate bets) then we would likely get much better odds. This is because the odds are much longer; however, if we pick our 10 draw forecast carefully, then we can reduce the odds considerably, and still have the possibility of multiple winning lines and making a profit.
(c) 2010 Phil Marks