Expected goals explained and how to use it to win bets

Before expected goals came along, football betting was relatively primitive. Shrewd punters watched game after game, hours of highlights, and studied the form guide to produce their best guesses.

If you’re focusing on the short term, past performance can indicate what’s to come, but it doesn’t help you get ahead of the trends.

Expected goals, devised by Howard Hamilton in 2009, assign value to every shot taken on the football pitch.

Not only is xG the most reliable method for interpreting whether or not a team created enough to win a match, but it’s also crucial to predicting future performance.

Developing a firm understanding of expected goals and how to apply them to betting, be that a more complex model or a one-off prediction can take your betting to the next level.

 

What is expected goals?

Every time a team takes a shot, it is assigned a value. Shots from tight angles or long distances are less likely to produce a goal, whilst a penalty is about as clear-cut a chance you can hope for.

That’s common sense because just by watching the game we get a sense of what constitutes a high-quality chance and what’s a speculative smash from distance.

But what if you could attach a definitive chance of success to every shot, and what use does that have in football, and more specifically football betting?

Expected goals are a product of data on over 300,000 sample shots - and growing - including distance, angle, area of the box, proximity to other players and the goalkeeper, and more.

The sum of shots produces a team’s expected goal value for a match, whilst the outcome of those shots adds to an ever-growing pool of data that creates the value.

For example, we know a penalty is worth around 0.76 goals. Why? The data on every penalty taken since expected goals began shows that 76% end up in the back of the net.

The same principle applies to one-on-ones within the width of the goals, headers in a crowded box, curled efforts from 30 yards, tap-ins inside the six-yard box and more.

More recently, xG has worked its way into the game through sports performance analysts who monitor every move of professional players, including the quality of their chance creation.

If a team is producing a high xG every match but failing to score, the manager knows his team is functioning the way it should, but it may mean extra shooting practice for the number nine.

 

Can you use expected goals to make money betting?

Understanding expected goals and how to apply them to football betting will make you a more successful bettor overnight.

The ageing cast of Match of the Day may scoff at it, tempering comments with ‘for what it’s worth’ regularly, but in layman's terms it’s worth money.

People who haven’t watched a minute of football in their life, let alone played the game, use xG to make large amounts of cash from betting.

If you were to take the Premier League table, including goals scored and conceded, and place it side by side with the expected goals table, you’ll be able to draw conclusions with little analysis.

If Liverpool are top of the league but rank seventh based on xG, it’s safe to assume there may be some regression to the mean shortly.

Upon further inspection, you notice the Reds have rattled in 50 goals so far this season, yet their xG tally reads around 40, which tells you they’re overperforming.

That’s only sustainable for so long and whilst everyone is rushing to add Liverpool to their Saturday acca’s are very slim odds, you lie in wait for an opportunity to bet against them at big odds.

Bookmakers also use expected goals, to the point that a team's tally from the previous season largely informs how they price the outright markets the following season.

But even in their wisdom, there are cracks in the market. Despite their understanding, they accept that Liverpool will win the majority of home games and need to be priced as such.

With the majority of punters preferring to back home favourites, it doesn’t hurt the bookies to go big on the underdog with the majority of their profits coming from bets on losing accumulators and bets on favourites.

Basing your bets on the expected goals table versus the real McCoy is the most basic use of xG, but even that will be enough to improve your fortunes.

Something about xG betting systems?

If you’re handy with a spreadsheet, you can go one step further by building your predictions using expected goals for, against and the league average.

By collating your data, be it from this season, the last 100 matches or several years, you can input every team’s expected goal values into a table.

Using a method called Poisson distribution, you could be just a few simple sums away from producing a score prediction based on chance creation rather than recent form.

You’d be surprised how often you can predict a shock that no one else saw coming. Betting on underdogs means betting at bigger prices, and that means winning fewer bets.

However, you’re now finding value in the betting market, which is more important than how many Ws you’re sticking in the win column because it will allow you to win over the long term.

How far down this particular rabbit hole you wish to go is up to you, so keep in mind there are much more simple applications.

One thing to look out for is teams that struggled during the first half of the season. You will always find a handful of teams that have produced a wealth of good chances and failed to convert them, leaving them languishing in the bottom half.

A switch of manager or a new number nine in the January transfer window are just two potential catalysts for change, but the chances are these teams will produce more wins in the second half of the season.

All of this to say, you have to be very open-minded about the possibility of the remainder of a team’s season looking very different to what’s gone before.

By this point, the bookmakers, and the betting public alike, may be down on this team’s chances, leaving you to swoop in and back them at exaggerated odds.

Equally, xG is a useful tool for mid-table teams that struggle to maintain a run of form across extended periods.

This can be particularly true of Leagues One and Two, where mid-table clashes often feel like a toss of a coin, and you’ll regularly see both sides priced around 15/8 to win.

Identifying a team winning games but producing low xG, or, on the contrary, a team struggling to score but producing a wealth of goalscoring chances, you know the script is about to flip.

You may even be able to become a specialist in a specific league, or on certain terms, as you get to grips with what informs their flip-flopping form.

The impact of individual players on expected goals

Some things will never change in sports betting. You can fiddle with xG til the cows come home, but it doesn’t count for much if a team’s top scorers are out.

Understanding the impact of individual players in the context of xG goes one step further to help you develop a broad understanding of what will allow you to utilise it properly.

In some cases, we know a player’s absence doesn’t make much difference. Other times, one clinical finisher may be the only reason a side lives up to their xG expectations.

Every team’s xG is made up of individual xG values for each player. Likewise, it’s important to consider the role of a specific defender or even the goalkeeper.

As we’ve already touched on, you can delve as deep as you like. Teams can produce drastically different goals against values based on what defenders are on the pitch.

Top goalkeepers are capable of bucking the trend and conceding fewer goals than xG would lead you to believe they should concede.

In the case of the world’s best strikers, it’s rare to see them match their xG tally or score fewer. Typically, they’ll go above and beyond, and they’ll do it year after year.

Separating Harry Kane from your Herbie Kane is just one small facet of the game that will allow you to correctly assess how sustainable a team’s level of performance is.