Have you seen the movie The Graduate? Y’know, the one from the 60’s, that you used to be able to enjoy before you found out Dustin Hoffman was a bit of a creep.
If not, it’s about a smart kid who graduates (as you might expect) from college, comes home to his parents’ lovely house, and subsequently loses all direction in his life.
Perhaps the movie’s most famous scene, outside of all the Mrs. Robinson stuff, is this one:
Well, ladies and gentlemen, today I want to say one word to you.
Just one word.
Corners, You Say?
Corners aren’t sexy. In fact, they’re probably the most boring part of actual football.
The thing is, I believe they’re undervalued as betting markets. My reasoning is based partly on common sense, and partly on the odds available.
The common sense part is that… who the hell bets on corners?! Proper, serious bettors, perhaps, but I’d confidently suggest most punters stick to the more exciting markets: results, goalscorers, maybe over/unders, and so on.
That’s what made me first think about betting corners. I then went to check the Premier League odds on corners, and… well, I was surprised how basic they are compared to other betting markets.
Your classic over/under for goals is 2.5. The odds either side of that over/under 2.5 will vary wildly game by game.
Let’s take EPL odds from this weekend as an example.
Here are William Hill’s over/under odds for probable snooze-fest Burnley vs Watford:
And here are their odds for Chelsea vs Arsenal:
Bit of a difference, no?
Over the past five season, there have been 10.6 corners per game in the Premier League. Accordingly, the main over/under for corners is 10.
Here are the main corners markets for those same two games.
Burnley vs Watford:
Chelsea vs Arsenal:
On the one hand, we’ve got one game which should be a defensive slug-fest, with two lackluster sets of attacking players. On the other, we’ve got (what should be) an all-out offensive war, with constant, end-to-end attacking.
And yet… the corner odds are almost exactly the same.
It was the same story for almost all the Premier League games I looked at, regardless of matchup.
You need only a passing knowledge of football to know this shouldn’t be the case. More attacking = more corners. It’s as simple as that.
(I did research this just to check it was true, by the way… and it is. More on that later).
The point is that I think there’s a big opportunity here. The odds for the market seem off, and I doubt there’s a huge amount of competition to exploit it.
The hard part?
Actually doing so.
So, I know that I want to bet on corners. A lot. I have a basic Premier League betting strategy for doing so, but I want to make it completely clear that it’s very, very early days.
To bet on corners, I need to be able to accurately predict how many there will be in each game.
I know already that, on average, there will be 10 corners in a Premier League game. That’s all well and good, but obviously we need to take into account the teams that are actually playing a given game. That’s how we know how to go over or under 10 corners.
(In time, I’d like to get more specific, and bet in a tighter range – 8-10 corners, for example. For now, let’s leave it at the over/unders.)
At the moment, as I mentioned in the intro, I’m trying to evaluate which other statistics best feed into corner creation.
The two main stats I’ve looked at so far are shots on goal, and shots on target.
My basic reasoning is that more shots on goal = more pressure = more corners. More shots on target = more pressure and more saves = more corners.
To get a base line, irregardless of specific teams, I’ve been going over past Premier League games and analysing them. I’ve only done 60 games so far (the 10 games played this season, plus 50 from the 2017/18 season), but these stats should get more accurate as I do more research.
To boil things down to their absolute basics, here are my findings so far:
- A team gets 0.41 corners per shot on goal
- A team gets 0.93 corners per shot on target
Teams that hit the target more are more likely to get corners. Not exactly rocket science, but it does confirm what you’d have guessed. Which is nice!
How do these findings actually help us?
Well, in theory, I can look at how many shots a team normally has, multiply it by the above number, and predict how many corners they should get. Ditto for shots on target.
Of course, it’s not really that simple. More stats are needed. More, I tell you!
Team X’s History
The above stats are useful for giving us a wider, overarching idea of how many corners a Premier League team should get per game. I’d be a fool, however, if I didn’t also encompass each team’s individual stats.
To be clear, I am not a stats expert. I do know, however, that stats-based predictions are part-science, part-art. Human input – in this case, actually watching and knowing about football – is required.
I’ve gotten Corners per Game data for all 20 Premier League teams from the past five seasons. I won’t use all of it, however, because – again – I actually watch football.
I know, as any other Premier League fan does, how much the league has changed in the past five years. Defending, in particular, has become an afterthought. Player turnover is so high now, too, that a Man United team in 2018 barely resembles that of a Man United team in 2013.
So, I’m only going to use data from the past three seasons. That helps with the player turnover part. I’m also heavily favouring the more recent data (more details below), to account for the more attacking nature of the Prem nowadays.
This is the biggest area in which I hope to build going forward.
You’d obviously expect a team with more possession to get more corners.
Right now, all I’ve done is proven that to be the case. In the 60 games I’ve looked at, the team with more possession has had more corners 71.67% of the time.
At a basic level, this could help us with another bet – Team With Most Corners – which you can also get decent odds for. It doesn’t help us at all, however, to predict how many corners there will actually be.
That’s something I’m going to work on going forward. I want to more clearly establish, if it’s possible to do so, the link between a team’s possession stats and their corner stats.
Watch this space!
Right now, there are three main stats I’m using to predict corners:
- Shots on Goal
- Shots on Target
- Teams’ Historical Corners per Game
The first two give us a baseline of what should happen in a Premier League game. The third builds in what an individual team has done in the past.
As mentioned above, I’ve used data from the past three seasons, and weighted the values to favour more recent ones. Right now, I’m generally using a 10%, 30%, 60% split.
For teams that have experienced massive changes in their fortunes, however – like Wolves – I’ve weighted recent seasons more heavily. In that particular case, I’ve gone 10%, 20%, 70%.
Finally, I’ve also attempted to tackle the problem of promoted teams; i.e. teams who’ve played in the Championship in the last three seasons.
I’ve given their shooting stats a 0.95 modifier, because… well, it’s easy for the Championship big boys to pile up shots against the poor saps who haven’t been taken over by enormous Chinese conglomerates.
Here’s how it currently looks:
S/G = shots per game. SOT/G = shots on target per game. C/G = corners per game.
So, when all’s said and done, Chelsea should get around 5.9 – let’s call it 6 – corners in any given game.
To work out how many corners we should get in a specific game – Chelsea vs Arsenal, for example – I just add up the two averages. Chelsea’s average (using my formula) is 5.9, Arsenal’s average is 5.87, so we should get 11.77 (11 or 12) corners.
In this case, I would massively favour the Over 10 Corners bet.
The odds for that bet? 5/6.
EPL Betting Tips – Week 2
Here are the number of corners I’m predicting for each game, followed by a recommended bet.
(For the record, I’ve favoured safer bets this week, even if the odds aren’t great)
Cardiff vs Newcastle
- Predicted Corners: 9.44
- Suggested Bet: Under 12 Corners, 2/5
Spurs vs Fulham
- Predicted Corners: 11.67
- Bet: Over 10 Corners, 4/5
West Ham vs Bournemouth
- Predicted Corners: 9.02
- Suggested Bet: Under 12 Corners, 4/9
Leicester vs Wolves
- Predicted Corners: 9.15
- Suggested Bet: Under 12 Corners, 4/9
Everton vs Southampton
- Predicted Corners: 9.7
- Suggested Bet: Under 12 Corners, 1/2
***Chelsea vs Arsenal***
- Predicted Corners: 11.77
- Suggested Bet: Over 10 Corners, 5/6
Man City vs Huddersfield
- Predicted Corners: 11.23
- Suggested Bet: Over 10 Corners, 8/13
Burnley vs Watford
- Predicted Corners: 8.17
- Suggested Bet: Under 10 Corners, 11/10
***Brighton vs Man U***
- Predicted Corners: 9.49
- Suggested Bet: Under 12 Corners, 2/5
- Suggested Bet: Man U More Corners, 4/7
***Crystal Palace vs Liverpool***
- Predicted Corners: 11.03
- Suggested Bet: Over 10 Corners, 10/11
- Suggested Bet: Liverpool More Corners, 4/9
Bet of the week = Chelsea vs Arsenal, Over 10 Corners. I’d recommend a double stake.
I also like the Man U and Liverpool games, hence the two bets on each.
Whatever happens this week, I won’t overreact. Any betting strategy needs to play out over many, many weeks before it can be judged a success or a failure.
There are three things I want to build on in the near future:
- Research more games, pure and simple. The more data I have, the more accurate my predictions
- Evaluate possession, and how it relates to corners
- Of the three current factors, look at weighing more heavily towards specific teams’ corners stats
For now though, my work is done, and it’s time to put my feet up and actually watch some football.
If you’d like to put down a bit of money and take some of my betting tips, I highly recommend William Hill for doing so.
Stick a tenner down this week, on my chosen bets or any other football odds, and you can get £30 in free bets going forward. Not bad!
Finally, any feedback genuinely is welcome.