Infogol’s Jake Osgathorpe uses expected goals (xG) to analyse the end of the 19/20 Premier League season and pick out some value for the upcoming 20/21 campaign.
Don't get too giddy about Man United
The signing of Bruno Fernandes has been hailed as the catalyst for Manchester United’s strong second half to the season, but a close inspection of their xG numbers before and after the Portuguese star arrived makes for interesting reading.
Prior to Fernandes’ arrival, United sat fifth in the league table but fourth in our xG table according to xPoints. Over the first 24 games of the season they averaged 1.80 xGF and 1.17 xGA per game (+0.63 xGD per game).
In their last 14 games, after Fernandes joined, United picked up the most points of any Premier League team (32) and went unbeaten down the stretch to jump into third.
However, they were only the third best team on xPoints in that time – they benefited from Liverpool coasting having secured the title – and their expected goal difference remained the same on +0.63 per game as they created slightly less in front of goal (1.69 xGF) but tightened up defensively (1.06 xGA).
United deserved to finish in the top four according to xPoints, but their underlying process still needs a lot of improvement if they are to reach the levels of Liverpool and Manchester City and compete for the title.
Basically, United had a lot of positive variance down the stretch, and post-break had the easiest schedule of any team in the league, so we can’t get carried away about their chances this season.
Chelsea look strong but major keeper issues
Chelsea performed at a higher level than United in 19/20, sitting third in our xG table and finishing just 0.7 xPoints behind Liverpool.
Their process (2.06 xGF, 1.21 xGA per game) was only marginally worse than Liverpool’s (2.01 xGF, 1.13 xGA per game) and they actually created better opportunities than the champions in attack.
Frank Lampard deserves a lot of credit as the Blues improved their expected goal difference per game by 0.22 from the previous season under Maurizio Sarri. That is a hugely positive sign, and they have already strengthened in attack in the transfer window, bringing in Timo Werner (28 goals, 20.8 xG) and Hakim Ziyech, and Kai Havertz could yet follow.
Ben Chilwell and Thiago Silva have been added defensively, and both look good additions, but the issue they have heading into the new campaign is in goal.
We calculate that, based on the on-target attempts faced during the 19/20 season, an average goalkeeper would have performed better than Kepa Arrizabalaga 99% of the time. He was the worst performing goalkeeper in the entire Premier League according to this metric.
Willy Caballero was brought in towards the back end of the campaign, and he fared only marginally better - an average keeper would have performed better than the Argentine 74% of the time.
That’s a damning stat, and suggests that a goalkeeper should be top of Chelsea’s shopping list if they are to make a more serious challenge; it clearly hasn't been given their significant outlay elsewhere already.
They could pose a bigger threat to the top two than United if they improve in this area.
Both United and Chelsea are rated well clear of their pursuers according to the Infogol model, so backing them as the Premier League winner without Manchester City and Liverpool at 2/1 looks a decent bet.
Results papers over North London cracks
Tottenham grabbed sixth spot thanks to a post-break surge that saw them collect 18 points from a possible 27, while Arsenal rallied under Mikel Arteta to finish eighth and win the FA Cup - they've since added the Community Shield too.
However, based on the underlying numbers, results are flattering the pair.
Jose Mourinho has earned a lot of plaudits for dragging Tottenham from 14th up to sixth but Spurs sit 10th in our xG table on xPoints since he took over in November, showing a negative expected goal difference (1.41 xGF, 1.54 xGA per game).
See below: Tottenham's 10-game rolling xG average 19/20
As for Arsenal, they finished the season 12th in our xG table according to xPoints, below the likes of Everton (7th), Southampton (8th), Sheffield United (9th), Burnley (10th) and Brighton (11th) – not the company you expect the Gunners to be keeping.
Arteta’s first game was on Boxing Day against Bournemouth, and since then the Gunners have ranked as the 13th best team according to xPoints. They were hugely fortunate to collect the sixth-most points in that time as they posted the sixth-worst xG process (1.31 xGF, 1.69 xGA per game).
The Gunners were 18 goals better off than what could have been expected based on the chances created and conceded (+11 GD, -7.6 xGD), and that is unsustainable over a long period of time.
Like Spurs, they have plenty of improvements to make if they are to make a top four challenge.
Wolves are rated as the fifth best team in the league by Infogol, ahead of Spurs and Arsenal. Also, Wolves have no European football this term, so they look well set to finish in the top six and are value to do so.
You can get 5/2 for Wolves to finish in the top six.
Leicester free-fall a worry but Saints charge impressive
Brendan Rodgers’ Leicester sat in the top four for 92% of the season, but crucially slipped out when it mattered, finishing the campaign fifth after winning just six of their final 22 league games (they won 12 of the first 16).
Ultimately, they deserved to finish outside of the top four, sitting sixth in our xG table according to xPoints, with their level of performance and results notably dropping after the halfway stage of the season, suggesting that opponents may have figured them out.
See below: Leicester's 10-game rolling xG average 19/20
They came back after the break in third with a five-point cushion over Chelsea and an eight-point lead over Manchester United with only nine games remaining. Infogol calculated that, based on the Foxes schedule, they had an 89.5% chance of a top-four finish. Leicester well and truly blew it.
Their performance levels deteriorated and if they continue in that manner in the upcoming season, it will likely be mid-table mediocrity.
Southampton were one of the form teams at the back end of the 19/20 campaign, and put up impressive underlying numbers, ranking fourth in our post-break xG table based on xPoints.
Expect the Saints to keep marching on in 20/21. They should steer well clear of the relegation zone as a result, and will instead be targeting a top-half finish.
You can back Southampton at 2/1 to finish above Leicester, which looks a big price given how those teams finished last season and the fact that the Foxes have the Europa League to contend with.
Villa deserve credit; Newcastle heading for drop
Aston Villa were written off by many last season when the league was halted, but the turnaround in results and performances was extraordinary.
Prior to the break, Dean Smith’s side had allowed a league high 2.41 xGA per game, but post-hiatus that was reduced to just 1.04 xGA per game - an incredible turnaround for a team who had obviously worked hard on defensive shape.
Former Villa boss Steve Bruce guided Newcastle to safety, and they were comfortably clear of the drop zone in 13th. However, they have to be very careful this season as, according to expected points, they ranked as the worst team in the Premier League in 19/20 – below even Norwich.
They finished the league season with an expected goal difference (xGD) of -29.6, and to put that into perspective Huddersfield’s xGD when they were relegated in 18/19 was -31.1.
Newcastle really need to improve if they are to survive again, and their level of over-performance is unsustainable.
You can back Newcastle at 5/2 for relegation, which is a value play.
Odds correct at 1700 BST (03/09/20)
Follow Sporting Life on social - find us on Facebook here or tweet @SportingLifeFC
Related football content
- Transfer Market: Latest updates
- Transfer Window: Done deals
- Premier League: Team-by-team fixtures
- 10 fixtures to look out for in new season
- How do promoted teams get on in the Premier League?
- Premier League: Historical points averages
- Latest football previews and tips
- Sporting Life App: Android & iOS
Responsible gambling
We are committed in our support of responsible gambling. Recommended bets are advised to over-18s and we strongly encourage readers to wager only what they can afford to lose.
If you are concerned about your gambling, please call the National Gambling Helpline on 0808 8020 133, or visit begambleaware.org.
Further support and information can be found at GamCare and gamblingtherapy.org.