jquery

Friday, 24 June 2011

Home Win Draw and Away Win Percentage Review

I have spent some time looking around the internet to see what systems are used by punters to try and give themselves an edge against the bookies. Using the stats available to me I have started to analyse % win ratios for the 4 English Leagues. As I always say this data only shows what did happen and not what is going to happen and therefore this should only ever be used as a guide when looking for that next winning bet.

If we look at the past 10 seasons as as whole for all leagues we see something quite interesting, and probably not too surprising to the masses.


We see that the % of home wins far outweighs the % of draws or away wins. Its always interesting to me to look at why this is the case, and to this day I don't fully understand it however its a fact - year on year there is always over 40% of games are won by the home side.

Here is the break down for each league for the past 10 seasons.
The Premier League consistently comes out on top with an even higher average for home wins. So the moral of this story is not to back too many away wins and draws as there aren't that many as a percentage! However if you do find one make sure you have backed it!
So what about days of the week? do we see a different pattern when we look at days of the week?

Here we can see the results as actual games rather than % - the trend follows for a Tuesday in the same vein as a Saturday, the difference is small enough to suggest there is no difference for a midweek game over a weekend game.

I read recently that someone placed 10 bets at the weekend, they selected 2 teams such as Manchester United and Celtic and then added a game to draw. They picked 10 different games to provide a chance of picking up some winning bets. You need approx 3 of the 10 teams to draw in any one week to make a small profit. We see from the chart above that its rare to see the number of draws to be over 30% for any one season so I don't think this tactic would work in the long term. However I was curious to know what return I might have got had i done this last season.

Rather than pick 2 teams I just picked Manchester United as they were the team with the most home wins and I only did the check for weeks where Manchester United were at home. My matches for the draw were all of League 2 - or at least any League 2 game occuring on the same weekend as Manchester United playing at home. What I found was that we had a good start. Infact in Manchester Uniteds 19 home wins, using this system would have paid out in profit on 6 occasions. There would have been 3 occasions where there was no bet and one one occasion no draws in the league corresponding to the home win. so 6 out of 16 is a 37% hit rate, which is slightly higher than our draw model however the net effect at the end of the season was a £117 loss. The reason for picking League 2 was because they had the most draws last season.

So although this plan looks like a good one, finding the teams to draw is not as easy as it looks, even if you take the view that a team has to draw at some point! Now should you get lucky and get 6 or 7 draws in a weekend then the profit is certainly there, but on statistics I don't think this is one to follow in the long term.

2 comments:

  1. Hello Martin, just wanted to say thank you for this article, is is exactly what I was looking for and answered part of a puzzle I'm going through, but I still have another question along this line and was wondering what you think or any knowledge you could share on this...

    The question is (and I know it most likely is a basic question)... Are there categories/types of games that these percentages are more or less accurate than others? I see you did the days of the week which was great info. One part of your article also that mentions premier games to be the most accurate for the home wins, so I guess what what I am wondering is what are the percentage variances of home wins for different types of matches. I know friendly games for example might not be to serious for a club. So lets say if a premier game home win rate might be 11% then the same club at home in a friendly might be only 6%? These are only numbers I used to illustrate and have no idea if there is any difference or not just very curious if there is any difference.

    Thanks for reading this, Cheers, Rob

    ReplyDelete
  2. Since over 40% of fixtures end up as home wins,this means that around 60% end up as draws or aways.. Even allowing for the fact that one would have to double one's stake to cover the latter the odds are often very favourable and one can exclude obvious home wins. Worth a try?

    ReplyDelete