Expected Goals: Another Metric to Help Punters Gain the Edge they Need28th February 2019 / Julian
It will come as no surprise to anybody that football is all about the goals. Forget the crunching tackles, the Rabona crosses, the back heels and the tiki taka, the beautiful game is ruled by those who put the ball in the net more frequently than others.
Or is it?
There is a growing train of thought that dictates it is the efficiency with which teams score goals compared to the number of shots they have, which ultimately governs success. And you only need to witness the ruthlessness that Manchester City have shown in front of goal in 2016/17 to date to know that’s true:
The column highlighted in red shows the conversion rate of a shot on target to a goal – and here Man City top the pile. Their current success levels are no surprise to anyone as such.
The new way of thinking in football data analytics collates information about such things as ‘conversion success rate’ of shots to goals; here’s an image that reflects this from across European football’s top four divisions in the 2015/16 season:
In short, the closer and more central to goal you can get the ball prior to shooting, the more chance you have of scoring.
As keen football viewers this makes sense to us: how many times do shots from distance actually nestle in the back of the net, rather than flying hopelessly into the crowd behind the goal? The highlights packages only show those that do end up beating the keeper; the ones that fly into Row Z very rarely make the edit.
This data has actually informed the style of coaching and tactical nous of many of European football’s most noted thinkers. For instance, when Pep Guardiola’s team have the ball in wide positions they very rarely cross it; instead, they work it inside and hope to pass the ball into the zone in and around the penalty spot for higher percentage shots. You will also note, if you watch carefully, that his Manchester City side rarely shoots from distance: that’s because of the low percentage turnover of goals. He’s a smart cookie, is Pep.
This is the basic premise behind the Expected Goals model, which assigns each shot a rating for its likelihood of finding its way into the old onion bag.
Creating Your Own Expected Goals Model
One of the things we love the most about football betting is the little models and systems we can put into place that will help us to take back some of the bookmakers’ advantage. They know that the beautiful game is also an unpredictable one, and so any opportunity we have to find some useful insights should be taken.
The Expected Goals model affords us this chance, as we can see how efficient a team is in front of goal, rather than using their overarching ‘shots per game’ or ‘shots on target’ data, which in many instances simply isn’t descriptive enough.
A player shooting from 30 yards, whose effort is comfortably saved by the goalkeeper, will still record a shot on target for his team. If this happens five times in 90 minutes then you might look at the stats and think ‘wow, Team A dominated Team B.’ But the reality of the situation is that this might not actually be the case.
As we have seen from the graphic above, getting the ball into central zones prior to shooting is essential in giving your shots the best possible chance of finding the back of the net. And this is the crux of the Expected Goals model; attaching a numerical value to the ‘quality’ of each shot at goal.
There are various different statisticians and football analysts utilising this kind of theory, but unfortunately there is no central resource for data: i.e. none of the data companies out there (who shall remain nameless) allow humble members of the public to access their statistics. So, Expected Goal data needs to be manually recorded whilst watching the games.
That’s a shame, as it may be enough to put off some punters, but we really can’t enthuse enough just how useful this kind of model is for adding extra efficiency to your betting.
How to Make Your Own ExpG Model
The beauty of the modern game is that every single moment is captured on camera. If you have a subscription to one of the key football channels on TV – happy days, you’re in business. Even with an internet connection and an inquisitive mind you can usually find footage of the weekend’s action, and this is the key weapon in your arsenal when constructing your own ExpG model.
What you will need to do is make a note of each of the shot locations recorded for your chosen matches/league. We don’t need to do any complex calculations here – we have seen from the image above where the ideal shot locations are in terms of scoring goals more frequently.
Next, we need a graphical basis with which to work….something like this:
The coloured boxes are the ‘danger zones’, and the numbers attached to each a points rating given for each shot at goal from this quadrant. So, as you are recording the shots that fly in during the match that you are watching, you will be slowly building a portfolio of information about each team. It will take five shots from outside the area to equal one from inside the central part of the six-yard-box, and as we know this is reflective of the success rates for each.
What does this mean for bettors?
If you are serious about long term betting profits, then you will already know that one of the key areas to focus upon is value bets; e.g. teams that are undervalued or overpriced in the market.
What the Expected Goals model does is identify teams that are efficient in front of goal, so even if they are going through a bad run of form we know that – long term – they will bounce back and pick up plenty of points.
Conversely, if a team is yielding a high number of ‘high percentage’ shots, as identified by the ExpG calculations above, then we can recognise that they *should* concede plenty of goals and thus lose a higher proportion of matches as a result.
This model helps to identify efficiency, and that is a central metric for any bettor seeking continued success in the market.
The tip is based on the personal opinion of the author. No success is guaranteed. Please gamble responsibly. 18+
* All mentioned odds were valid at the time of writing. Betting odds are subject to fluctuations. Please check the current odds with the respective bookmaker!
Read the latest football predictions:
- Wycombe vs Tranmere Prediction & Betting Tips, 20th November 2019
- Newport vs Grimsby Prediction & Betting Tips, 20th November 2019
- Belgium vs Cyprus Prediction & Betting Tips, 19th November 2019
- Germany vs Northern Ireland Prediction & Betting Tips, 19th November 2019
- Scotland vs Kazakhstan Prediction & Betting Tips, 19th November 2019
- Netherlands vs Estonia Prediction & Betting Tips, 19th November 2019
- Bradford vs Shrewsbury Prediction & Betting Tips, 19th November 2019
- Gillingham vs Sunderland Prediction & Betting Tips, 19th November 2019
- Swindon vs Cheltenham Prediction & Betting Tips, 19th November 2019