The stats you need when betting on football

14 min

Analysing the mountains of available football data for betting can be extremely time consuming. Matchbook explains what you should seek out and what can be easily ignored.

Back in the nineties, ‘Statto’ was the beautiful game’s original stats geek. Sporting Harry Potter-style specs and a vile dressing gown, he filled the role of your socially awkward nerd on cult TV show Fantasy Football. At the time he was a figure of fun, yet Statto – played by sports broadcaster and gambler Angus Loughran – was ahead of his time.

Twenty years ago, stats weren’t as abundant and readily available as they are now. These days, we’re spoilt for choice and the sheer volume of information on squads and managers is truly mind-boggling. Harnessing relevant data and filtering out the rubbish is essential to making more informed bets and, hopefully, becoming a successful bettor.

Today everyone is a wannabe stats geek throwing out Bellerin’s top speed or Kante’s pass completion percentage like they are confetti at a wedding.

The fact that Tottenham playmaker Dele Alli covers almost 12 km every match is probably intriguing to fitness coaches but it’s hardly going to help pinpoint betting opportunities.

Indeed, many of the stats out there are probably not worth your attention. So what should you be looking for to make your betting decisions that bit more profitable?

Dele Alli battles for the ball during the Premier League match at the Liberty Stadium, Swansea.

The Basics

The fundamentals like recent form (usually the past six matches) should be your first port of call, as well as the head-to-head (H2H) record between the two sides. Although players and managers come and go, which makes some question H2H records as a valid prediction tool, certain clubs do seem to have their bogey sides.

Maybe it’s psychological, but it’s equally plausible it’s a clash of strategies and playing styles. For example,

Tottenham have only managed to emerge victorious from five London derbies against Chelsea in 50 Premier League encounters stretching back to 1992.

Similarly, Manchester United have only suffered three defeats at the hands of Sunderland in 31 clashes in the top flight. Furthermore, Everton have tasted victory just nine times versus Merseyside rivals Liverpool in 49 Premier league games and only mustered eight wins in 49 against Man United.

Koeman and Klopp shake hands following Liverpool’s 3-1 victory in the Merseyside derby.

To see how current and previous Premier League clubs have fared against one another since the top flight formed in 1992, check out this tool.

Goals are the goal

Average goals scored and conceded are also important stats when assessing the 1X2 market, as well as the likelihood both teams will score and the under and overs markets. For instance, goal-shy West Brom scored a meagre average of 0.89 goals in the 2015/16 season (the second lowest in the Premier League) whilst conceding 1.26 goals at the other end.

Indeed, matches involving the Baggies averaged just 2.15 goals per game, which is significantly lower than the season average of 2.8 in the Premier League. At the other end of the spectrum, Man City netted an average of 1.87 goals per match and shipped an average of 1.08 goals.

West Bromwich Albion rely heavily on Rondon for goals.

The discrepancies are even higher between the teams in some years. Matches in the 16/17 season over the first 25 games involving Liverpool produced an average of 3.4 goals. It was a similar figure for Bournemouth and Swansea games due to their porous defences.

Furthermore, certain clashes consistently produce goals but there are no hard and fast rules and you need to look at each game on its own merits. For example, you might expect derbies to be frenetic affairs played at 100mph with goal-scoring chances few and far between. But the stats vary considerably depending on the teams involved.

While the “fast and furious derby” rule is true of the Merseyside Derby with an average of just 2.26 goals per game in the Premier League era

It’s the opposite in North London….

Tottenham and Arsenal encounters average 2.83 goals, which is 0.13 goals above the norm from the Premier League and will have a significant bearing on how total goal markets are priced up.

Expected Goals and Shots

Naturally, injuries to key players – particularly forwards – will have a bearing on goal expectancy, so you need assess how goals scored and conceded are impacted by the absence of a side’s talisman or main striker. But luck (or variance if you prefer) is also a huge factor.

The natural variance in performance is why many bettors and football analysts are increasingly looking to Expected Goals (XG) as a key metric. A team’s XG is a measure of how many goals they would score on average given the number of shots on target they have been taking in certain areas of the field. It’s a fairly complex calculation, and if you’re interested in learning more you can try here or here.

Regular stats on XG can be hard to come by, but it’s worth following Michael Caley on Twitter as he publishes graphics showing XG from almost every major European game. You should also check out the Stats Bomb website and give them a follow on Twitter too. Or if you’re brave you could start compiling your own XG stats.

To keep it more simple you can focus on simple shots metrics. Even if a club is struggling to score, they may well be racking up above average shots on goal, so paying attention to total shots on and off target in recent matches could pay dividends. Indeed, you may encounter generous odds on the 1X2 market or there could be value to be had on the overs.

Stats to ignore

So if shots and XG are ones to watch for, possession is the opposite. While possession stats look impressive they should be treated with a certain degree of caution.

A side might enjoy 70% of the ball and yet are hopeless at creating and finishing chances, while the converse can also be true.

As 15/16 Premier League champions Leicester City demonstrated this by averaging just 44.8% of possession (only West Brom and Sunderland had a lower percentage), during their title winning season. You don’t need to keep the ball to win matches if you have the pace and quick passing to catch opponents on the break.

Jamie Vardy doesn’t care about possession.

Similarly, a side with an above-average corner count might lead you to think they are more likely to score than an opponent with contrasting averages. However, that’s not necessarily true.

The stats suggest only around 3% of all corners lead directly to goals.

At the midpoint in the 16/17 season in the Premier League the average number of corners for all home and away sides was 5.5 and 4.8 respectively. So goals from corners are pretty rare occurrences, which means these set pieces aren’t as advantageous as you might think.

Individual players stats are also usually not that relevant to the bigger picture. That said a player’s importance to the side can sometimes be underestimated, however, and comparing team performance against a combination of stats such as passes completed, tackles made and shots can sometimes be helpful. But tread with caution here.

The bigger picture

Of course, all stats should be put into context as there could be all manner of explanations for deviations from the norm. For example, a bad run or goal drought could be due to influential players on the side lines or they could have experimented with a new formation. A new manager often gives a club in the doldrums a much-needed boost.

Hull City manager Marco Silva is turning things around for The Tigers.

It’s also important to remember that all the stats you are studying are in the public domain. And professional football bettors will have even more data at their fingertips, as well as predictive modelling and people feeding them information from inside clubs.

As a result, you’re unlikely to find a mispriced Over 2.5 goals market, for example, but knowing the stats is as much about avoiding bad bets as finding good ones.

Focusing on more obscure markets and less high-profile leagues where the pros and layers may have missed key stats is something to consider investigating.

A few of our favourite resources for football statistics are: