After landing two winners last week, Infogol's Jake Osgathorpe uses expected goals (xG) to predict the likely winners and scorelines of each match in Gameweek 8...
"Usually, goals would be the obvious play in this game, but Manchester City’s defensive improvements could make this a tighter contest, and Liverpool’s ability to control the match with or without the football means I can see them avoiding defeat."
All square at Amex
Brighton vs Burnley
Last weekend at Spurs was Brighton's worst performance of the season to date according to expected goals, as they struggled to create opportunities (xG: TOT 2.0 - 0.4 BHA).
They had been excellent prior to that game, but have won only five points from a possible 21 which is of huge concern. Burnley are one of three winless teams through gameweek seven, and sit bottom of the Premier League after a disappointing 3-0 defeat against Chelsea, though again they conceded few quality chances (xG: BUR 0.5 - 1.2 CHE). That has been the case throughout 20/21 so far, with Sean Dyche's side allowing just 1.2 xGA pg, so they will again be tough to break down here. The Infogol model makes this a much more even game than the market suggests, with Brighton given only a 39% chance of winning. This means we calculate a 61% chance of Burnley avoiding defeat, though we see both teams hitting the net more often than not (51% BTTS) - 1-1.
Saints train to keep rolling
Southampton vs Newcastle
Southampton really have turned their season around after a poor start, picking up 13 points from their last five matches including a clinical display against Villa last time out (xG: AVL 2.5 - 0.9 SOU). Ralph Hasenhüttl's side have been particularly good at home so far this season, boasting a process of 1.5 xGF and 0.9 xGA pg a St Mary's. They will have to play for a month or two without their talisman Danny Ings though (0.37 xG/avg match), which is a blow. Newcastle have made a steady start to the new season, winning three and drawing two of their seven matches, including a deserved win over Everton (xG: NEW 2.8 - 1.4 EVE). On the road though, their underlying process has been tragic (0.9 xGF, 1.8 xGA pg), despite the fact they are unbeaten away from home. Even without Ings, Southampton should prove too strong here (51% SOU), with BTTS likely (52%) - 2-1.
United to get back to winning ways
Everton vs Manchester United
Everton's season has been derailed since the international break, with the Toffees winless through three matches and deservedly so according to expected goals. They have since dropped to fourth in the league table, but have fallen further in our xG table, now residing in 7th. Having looked solid defensively through four games, they have since conceded a worrying amount of good scoring chances (2.1 xGA pg since break), taking their overall process to 1.8 xGF and 1.4 xGA pg. Manchester United looked to be finding their feet, playing some excellent football in the Champions League, but a defeat at home to Arsenal last weekend halted their momentum. In attack, they were woeful against Arsenal (0.4 xGF), but defensively they were resolute once again, allowing just 0.3 non-pen xGA. This means that, since the international break, United have allowed just 0.7 non-pen xGA pg in all comps. Defensively they are improving, but we need to see more fluidity in attack, and Anthony Martial's reintroduction could see that happen. The Infogol model likes United in this game (46%), though goals may not follow (53% U2.5) - 0-2.
Selhurst to host cagey draw
Crystal Palace vs Leeds
Crystal Palace were well-beaten by Wolves last Friday, as they continued a trend of offering little when going behind. Only 14% of the xG they have created this season has come when they have been trailing, showing how much they struggle when going behind. Following an impressive win of Aston Villa, Leeds were thumped 4-1 at home by Leicester in a disappointing defensive display (xG: LEE 1.9 - 3.0 LEE). They do continue to be hit and miss, which could in part be due to the type of opponents they face, but their process overall is mid-table (1.7 xGF, 1.9 xGA pg). This could be another tricky game for Leeds, with Palace happy to play on the counter attack, so we like the Eagles chances of avoiding defeat (66%) in a low-scoring game (43% U2.5) - 1-1.
Chelsea look to have hit their stride
Chelsea vs Sheffield United
Things are looking up now for Chelsea after a questionable start, with the major improvements coming defensively, with the Blues keeping clean-sheets in three of their last four league games. Edouard Mendy's introduction has likely helped boost the confidence of Chelsea's defenders, but a more defensive-minded set-up has also contributed, and had an effect on their attacking numbers (1.6 xGF pg in PL). Clinical finishing is rife though, something that nearly cost them last term (16 goals, 11.1 xGF). Sheffield United remained winless after a narrow loss at home to Manchester City last weekend, and they continue to struggle to score goals, netting just three times in seven games. Chris Wilder's side are averaging 1.1 xGF pg, so creating the chances is an issue, so they could struggle to test Chelsea, who themselves can rack up another big win. The Blues should win here (65%), with goals likely (54% O2.5) - 3-0.
Hammers to bounce back
West Ham vs Fulham
West Ham saw their four game unbeaten run come to an end at Anfield last weekend, as they were rightly beaten according to xG, with their attack struggling to create (xG: LIV 1.5 - 0.3 WHU).
Overall though, it has been a stellar start to the campaign for David Moyes' side, who have picked up eight points from a tough schedule (NEW, ARS, WOL, LEI, TOT, MCI, LIV), with their process impressive also (1.5 xGF, 1.2 xGA pg). Fulham got their first win of the season on Monday, deservedly beating a West Brom side who continue to offer nothing in attack (xG: FUL 1.5 - 0.5 WBA). The Cottagers have been poor on the whole this campaign, especially defensively (1.8 xGA pg). West Ham are fancied to get another win here (51%), with goals likely at the London Stadium (52% O2.5, 54% BTTS) - 2-1.
Easy win for Spurs
West Brom vs Tottenham
As mentioned, West Brom's attack continues to be non-existent in the league this season, averaging a measly 0.5 xGF pg through seven matches. They are on course to be the worst PL attacking team since Infogol started collecting data (2014), with that title currently held by Aston Villa (0.8 xGF pg). They shouldn't threaten an improving Tottenham here, who rightly beat Brighton last weekend thanks to Gareth Bale's first goal back in a Spurs shirt. That win moved them up to third in the table, but they sit second in our xG table, showing it is no fluke they are near the summit at this stage. Only Liverpool (2.5 xGF pg) have been better in attack than Spurs (2.2 xGF pg), so expect Harry Kane and co. to register another win here. We give them a 56% chance of doing so, though interestingly, the model thinks this could be low-scoring (56% U2.5) - 0-2.
Low-scoring draw between two evenly matched teams
Leicester vs Wolves
Since defeat to Aston Villa, Leicester have been excellent in the Premier League, beating both Arsenal and Leeds away from home to jump up to second. Their display at Elland Road was impressive, as they created plenty (xG: LEE 1.9 - 3.0 LEI), but it was a contrast to their performance at the Emirates in which it was their defence that impressed (xG: ARS 1.0 - 0.9 LEI). This shows that this team can play, and win, in different ways, making them a tricky team to assess.
A week after frustration against Newcastle, Wolves dealt with Crystal Palace with ease, scoring a few rare first half goals. Nuno's side continue to be a really tough team to play against, having allowed 0.9 xGA pg in their last four. They are yet to blow anyone away in attack, which would be my only doubt around Wolves, but they hang in games and keep them close. This should be no different. Infogol is doing some good old-fashioned fence sitting here, with Leicester and Wolves given a near identical chance of winning (~35%), so a low-scoring (63% U2.5) draw is the selection - 1-1.
City and Liverpool to play out cagey draw
Manchester City vs Liverpool
What a huge game early on in this season, with the result of this drastically changing the percentage chance of each team winning the Premier League, with both sides currently Infogol's (and the market's) runaway favourites.
Manchester City have looked strong and sturdy in recent weeks, allowing a staggering 0.5 xGA pg in their last six matches across all competitions. That is a monstrous improvement, but it has come at a cost, as in that same period, City have created an average of just 1.6 xGF pg. For perspective, City averaged 2.7 xGF pg in 19/20, 2.4 in 18/19 and 2.3 in 17/18. So, they currently aren't performing at their frightening attacking best just yet, though Gabriel Jesus' return could help with that.
Liverpool were so impressive in midweek in what was arguably their best performance of the season to date, thrashing Atalanta 5-0 in Bergamo (xG: ATA 1.2 - 2.5 LIV). Their deserved win over West Ham put them back at the summit of the Premier League table, and they sit atop of our xG table too, so while results may look unimpressive, they have been thoroughly deserved. Defensively they have looked shaky at times (1.3 xGA pg), and are missing key defensive personnel which is an issue, but they continue to control matches exceptionally well. Their attack also looks to be at their best, averaging 2.5 xGF pg, with Salah and Mane constants in the side, though Roberto Firmino (0.29 xG/avg match) now has serious competition from Diogo Jota (0.5 xG/avg match), who netted a hat-trick in midweek.
Usually, goals would be the obvious play in this game, but Manchester City's defensive improvements could make this a tighter contest, and Liverpool's ability to control the match with or without the football means I can see them avoiding defeat. The Infogol model calculates a 55% chance of Jurgen Klopp's side getting at least a point, with both teams scoring highly likely (64% BTTS) - 1-1.
Arsenal to follow up Old Trafford win
Arsenal vs Aston Villa
Arsenal got the monkey off their back last weekend, as their win at Old Trafford was their first at a 'big six' team in 29 attempts. They deserved the win, limiting Manchester United well defensively, but again they didn't look threatening in attack, with the penalty (0.8 xG) being the only big chance in the game (xG: MUN 0.4 - 1.1 ARS). Mikel Arteta needed that win, but it was another good performance against a top side, a test they have repeatedly passed under his tutelage. Aston Villa's dream start is looking like a distant misery, with back-to-back home wins knocking them down a peg or two.
Their 4-3 defeat to Southampton was a strange one, as Saints raced into a 4-0 win thanks to clinical finishing before Villa rallied late and racked up the xG (xG: AVL 2.5 - 0.9 SOU). They still sit in a lofty 5th place in our xG table, with a game in hand, and with their process they shouldn't be dismissed here (1.9 xGF, 1.3 xGA pg). Villa will cause Arsenal problems, but the Gunners are taken to win (57%) in a high-scoring game (53% O2.5, 53% BTTS) - 2-1.
Infogol's Correct Score P+L
Brighton vs Burnley: Back the 1-1 @ 7.613/2
Southampton vs Newcastle: Back the 2-1 @ 9.617/2
Everton vs Manchester United: Back the 0-2 @ 13.5
Crystal Palace vs Leeds: Back the 1-1 @ 6.25/1
Chelsea vs Sheffield United: Back the 3-0 @ 10.519/2
West Ham vs Fulham: Back the 2-1 @ 9.617/2
West Brom vs Tottenham: Back the 0-2 @ 8.415/2
Leicester vs Wolves: Back the 1-1 @ 7.06/1
Manchester City vs Liverpool: Back the 1-1 @ 8.88/1
Arsenal vs Aston Villa: Back the 2-1 @ 8.27/1