Between the lines

The world cup is over but there is no let up and the new English football season is fast approaching. BettingTools has been very successful in terms of providing profits from horse racing tips but in recent years the football side has things has fallen by the way side. So, I’ve decided to step things up again with my football analysis which will hopefully result in some decent profits from football betting next season.

Over the last couple of weeks, I’ve been playing around with the data from, programming formulas and converting the data into my own odds using the Poisson distribution formula. So far I’ve seen some good seasons in terms of ROI but overall the odds I’ve created have ended up extremely similar to those of the bookies. I guess this is to be expected. If you analyse the same kind of data that the bookies do and in a similar way, you are going to get similar results.

Initially this got me quite disappointed but two posts in particular have inspired me into thinking about the best way to approach the football betting markets or any betting markets for that matter. The first one was that of our very own Pinpoint and his handicap service blog post. There are so many statistics and variables in horse racing that it can be overwhelming but Pinpoint has been successful sticking to mainly one and that is speed. He has a couple of ways of analysing it, which provides tips for two different profiles but that is his main focus. Race times and speeds that horses have recorded are still not a big part of what’s provided in the mainstream or analysed by the majority of punters in UK racing.

After his speed analysis he uses his own intuition to make a decision on whether a horse that have shown good speed are worth a bet. He will account for going, course, mark, distance, trainer form, jockey, draw and listen to interviews with trainers to help him make his final decision. It dawned on me that I’ve been discounting my own intuition too much and this is often the best way to get an edge on the markets. Particularly in the lesser watched leagues.

I think that shots in football is like speed in racing, a key statistic that is still in its infancy in terms of using it to accurately predicts match outcomes. Things have moved a step further with expected goals (xG) but there is still work to do and I think the experts themselves are still struggling to convert these expectant goal figures into betting odds to significantly improve on what has proven to work best so far. Goals change games, they impact each other and the players involved in the shots/matches all make a difference and computing this information for a stats based model is problematic. This is where our intuition can excel. Was the shot lucky too or were there several times where a shot could have been performed but wasn’t that the data is missing etc.

It’s very difficult to create an accurate prediction model but much easier to find flaws in a model or flag up things that the model/odds is unlikely to have accounted for.

This pinnacle article is the second piece that has inspired me as it shows that data analysis and intuition are both important.┬áThe article highlights the fact that our intuition can outperform a model on a small amount of data but it’s difficult to apply a detailed intuition based analysis on thousands of markets however and this is where models have the advantage. It very astutely points out that it’s hard to believe in our own judgment/intuition mainly because we can’t always test it, at least retrospectively. By submitting our tips and thoughts to a tipping competition like our own though, we can go some way to testing our analysis.

When I priced up premier league matches using my own intuition and no data as a starting point, I saw ROIs of upwards of 20% in some seasons. A return almost impossible using a purely statistical approach but I was reluctant to increase stakes because I still didn’t trust my own analysis. It wasn’t “proven” still in my own mind. Unbelievable really.

Up until now I’ve chosen to focus on either a statistical approach or an intuition based one but never the two in unison for the same tipping profile.

It’s now clear to me that focussing on one key stat and analysing in-depth and then using your own intuition to see what the stats might not have accounted for or that reinforces the stats will allow us to make a very informed decision on whether a selection is worth a bet. This is a type of Bayesian analysis where the odds are based on historical data but moved one way or another slightly for every bit of new information.

A model doesn’t haven’t to yield a positive ROI for it to be of use. It can be a great starting point to then filter on other criteria to make a selection.

It’s inspired me to resurrect the market movers profile which is based on solid wagering information on the betfair exchange. Even if the strategy isn’t profitable overall it can still be very useful in helping us to decide on whether to bet or not.

So, to ensure an edge in football betting next season, I’ll be analysing shots in a superior programmatic approach than ever before to price up the matches, whilst also watching and reading about the matches themselves in greater detail to enable my intuition to have the final say.

Of course having faith in our intuition isn’t always the problem and occasionally we can have too much faith in it for a particular market. This cannot lose, or this will not be beaten are phrases we’ve all heard many times before. There will be times where we get it so wrong and times where luck has conspired against us but as long as we stake sensibly it won’t ruin us. It’s for this reason that I always prefer a level stakes approach theses days. Nearly all the long-term successful bettors I have encountered use level stakes and it will keep you much saner than the variance that can be experienced using a variable staking system.

Betting has become so much easier for me since I started to see the game as being about gaining an advantage over lots of markets. No one bet makes any kind of difference and I’m not too fussed about individual results. If I’m wrong on a bet it doesn’t hit me hard and I can move on much easier.

Bring on the new season. I’m ready for you!