Home » Worst prediction ever: Arsenal to win Premier League from Liverpool and Man City

Worst prediction ever: Arsenal to win Premier League from Liverpool and Man City

If your supercomputer says that Manchester City have only a 25% chance of winning the Premier League, then get yourself on Facebook Marketplace.


Supercomputer says yes to Arsenal
Mediawatch knows there is little fun to be had in predicting that Manchester City will win the Premier League title. Treble winners Manchester City with their three consecutive title wins, six successive Premier League victories and Premier League top scorer are a dull prediction. And a particularly poor prediction because there are a teeny-tiny number of City fans compared to Arsenal, Liverpool or Manchester United.

But do you know who are overwhelming favourites to win the Premier League? Manchester City.

And do you know why? Because they have won it three times in a row and are quite clearly the best placed to win it again. So best placed that the shortest odds you can get on it happening are 7/10. That means of every 17 times this particular title race is won, the bookies expect City to win 10 times. Or just less than 60% of the time.

We know some of you struggle with maths so to put it really simply: It is rather more likely to happen than not.

So when we see in The Sun that ‘ARSENAL will lift the Premier League title in May according to a supercomputer’, we can only suggest that somebody gets a new supercomputer. Can you get them on Facebook Marketplace? From a smoke-free home?

The artificial intelligence has Arsenal first, followed closely by Liverpool in second and last season’s champions Manchester City slumping to third.

Then the ‘artificial intelligence’ is f***ed. Now we can see how a machine might log the fact that Arsenal have beaten Manchester City and Liverpool in recent months and recorded a 6-0 win over a theoretically good West Ham side and conclude that the Gunners are an excellent football team but if the machine says yes, sometimes the person running the machine has to say no. And look again at any algorithm that puts City – who have not finished as low as third since 2017 – in actual third.

The data was crunched by BettingExpert’s algorithm named BETSiE, which simulates the Premier League season 100,0000 times.

The data also takes into account results from last season, pre-season results, the current season and xG throughout the year.

BETSiE was clearly impressed by Arsenal’s 6-0 rout of West Ham on Sunday and tipped the Gunners to beat Liverpool to the title by a point and giving them a 39.3 per cent chance of their first crown since 2004.

Funnily enough, the actual betting experts who come up with actual betting odds, give Arsenal a 14% chance.

Liverpool meanwhile are given a 35.8 per cent probability and Manchester City just 24.8 per cent.

Where can we sign to put every penny in our child’s cash ISA on a 3/1 shot for City to win the Premier League?

So to be clear, Manchester City – who won the Premier League last season and have the most points-per-game this season – are somehow only going to finish third?

Something has gone wrong with that algorithm and we suspect it must be the weighting on xG. Because the one area in which City lag behind both Liverpool and Arsenal is xG. Not actual goals (they score plenty of those) but the expected goals.

So City – who have the world’s best goalscorer as well as a gaggle of excellent finishers in Phil Foden, Julian Alvarez, Bernardo Silva and more – are deemed by this ‘supercomputer’ to be far, far less likely to win the Premier League than Liverpool, ostensibly because they have a striker who is not as much fun as Darwin Nunez.

Now Mediawatch likes a bit of xG (here’s a Premier League table based on xG) but it’s not a statistic that should carry more weight than actual football results.

And you should spot the problem as soon as Arsenal become favourites on the back of a 6-0 win over a really sh*t West Ham side.


Turn the beat around
You know what’s even better than predicting that Arsenal will win the Premier League? Pretending that everything is rosy at Manchester United, obviously.

And for that we turn to MailOnline:

How Erik ten Hag has turned Man United’s season around: Patience paying off with Andre Onana, a balanced front three is unlocking Rasmus Hojlund’s potential… and super-sub Scott McTominay is thriving in a new role

Pesky fact: They are still sixth and 13 points behind Liverpool.

We are then walked through ‘five things Ten Hag has done to help turn United’s ailing season on its head…’ and when they start with ‘sticking with Onana when it would have been so easy to bow to fan pressure and drop the Cameroonian shot-stopper’ then you immediately know it’s horsesh*t.

After a disastrous 3-3 draw at Galatasaray in the Champions League in December – where Onana was blamed for all three goals – calls were deafening for Ten Hag to accept he had made a mistake signing Onana and to instead turn to Altay Bayindir or Tom Heaton.

Bollocks they were. Any sensible person would have noted that Manchester United had literally won their last three Premier League games to nil. Oh and at that point United were sixth and just four points behind Liverpool. So actually in a better position than they are now.

We are happy to give Ten Hag credit for settling on a front three that features Alejandro Garnacho on the right, and for trusting Kobbie Mainoo alongside Casemiro (though this is the first time they have both been fit this season so it’s a bit of a no-brainer).

But Scott McTominay ‘thriving in a new role’? Three of his first five appearances of the season were off the bench. He was only given a run of starts because of injury problems in that United midfield. He absolutely should be on the bench.

And let’s not forget that the sainted Ten Hag wanted to sell him in the summer.