Infogol's Jake Osgathorpe uses expected goals (xG) to predict the likely winners and scorelines of each match in Gameweek 6...
"Chelsea conceded three at home to Southampton in a draw last weekend, with that result was a fair one (xG: CHE 1.84 – 1.81 SOU), and their performances on the whole this season really haven’t impressed me. They have created an average of 1.28 non-pen xGF per game through five matches, which is a disappointing tally."
Villa's 100% start to end
Aston Villa vs Leeds
Aston Villa are in dreamland after a late Ross Barkley goal gave them a narrow 1-0 at Leicester last weekend (xG: LEI 0.63 - 0.92 AVL). Dean Smith's side have been backed into 60.059/1 on the Exchange to win the title, which seems ridiculous, but they have improved markedly since the restart, especially in defence. Only four times in 14 games since the restart have Villa allowed more than 1.0 xGA, which is an astonishing feat. Leeds suffered their first home defeat of the season at the hands of Wolves, as their clinical touch deserted them (xG: LEE 1.29 - 0.48 WOL). Their mid-table position is a fair reflection of their performances, but I feel as though we haven't yet seen the best of them. We make Villa narrow 36% favourites, but I can see this one being a low-scoring (60% U2.5) draw - 1-1.
Back the 1-1 @ 7.87/1
A more entertaining City win
West Ham vs Manchester City
West Ham did the unthinkable last weekend, as after looking down and out after 80 minutes against Tottenham, they rescued a point thanks to a late rally and a stunner from Manuel Lanzini (3%). At the end of the game, a draw was a fair reflection of the chances in the game (xG: TOT 1.73 - 1.57 WHU), as the Hammers continue to impress, sitting fourth in our xG table based on xPoints. Manchester City did just enough to squeeze past Arsenal, though they were far from convincing (xG: MCI 1.54 - 0.99 ARS). They did look more solid defensively, but I think that was down to their approach to that particular opponent, and I think they will take more chances in this game. City are strongly fancied to get another win (62% MCI), but this should be more goal-laden (67% O2.5, 62% BTTS) given how impressive West Ham have been in attack so far (1.96 xGF pg) - 1-3.
Back the 1-3 @ 11.5
Palace to edge dull encounter
Fulham vs Crystal Palace
Fulham looked marginally better in attack against Sheffield United last weekend, but again failed to create a non-penalty big chance (>0.35 xG). The Cottagers have averaged 0.82 non-pen xGF per game this season, which is worryingly low. Crystal Palace's non-pen xGF per game is even lower (0.70), not helped by the fact they failed to take a single non-pen shot at home to Brighton last Sunday. That speaks volumes for the pragmatic approach that the Eagles play with, so we shouldn't expect many goals here at all (64% U2.5, 42% BTTS). Palace are marginally better defensively, and the Infogol model thinks they are a better team than Fulham even with home field advantage, giving them a 44% chance of edging a win - 0-1.
Back the 0-1 @ 8.415/2
United to punish Chelsea
Manchester United vs Chelsea
Manchester United got some much-needed confidence thanks to a 4-1 win at Newcastle last weekend, though they were again far from convincing. It was a much better performance nonetheless though, as they aim to get their process to the levels we saw post-break last season. Their midweek win in Paris was much more like it, with United looking very impressive. Chelsea conceded three at home to Southampton in a draw last weekend, with that result was a fair one (xG: CHE 1.84 - 1.81 SOU), and their performances on the whole this season really haven't impressed me. They have created an average of 1.28 non-pen xGF per game through five matches, which is a disappointing tally, and suggests they are yet to click in attack. I think we could be in for an open encounter with goals at Old Trafford (54% O2.5, 57% BTTS), with the model suggesting a 43% chance of a home win - 2-1.
Back the 2-1 @ 11.010/1
Reds to add to Blades woes
Liverpool vs Sheffield United
Liverpool were extremely unfortunate not to win the Merseyside derby last weekend, creating by far the better chances and having a late goal dubiously ruled out by VAR (xG: EVE 1.35 - 2.85 LIV). It was an excellent response to the 7-2 drubbing, though the injury to Virgil van Dijk was a huge blow. Sheffield United are still winless after sharing a point with Fulham at Bramall Lane, a result that was very fair (xG: SHU 1.83 - 1.82 FUL). They have been a team that don't see many chances at either end (1.14 xGF, 1.32 xGA pg), but they will likely struggle to keep the Reds out. Liverpool have a 65% chance of winning on Saturday, in a high-scoring game (59% O2.5) that should see both score (53% BTTS) - 3-1.
Back the 3-1 @ 11.010/1
Toffees to be held again
Southampton vs Everton
After a slow start, Southampton appear to have found their groove, picking up seven points from their last three games, including a 3-3 draw at Stamford Bridge last time out. Their process through five games has been very solid (1.47 xGF, 1.27 xGA pg), and they shouldn't be at all underestimated against the league leaders. Everton were fortunate to get a point against Liverpool, but remain unbeaten nonetheless after five games. They have really impressed so far, with their process especially strong in attack (2.14 xGF pg), but they do look vulnerable defensively, conceding seven in their last four games. We think Southampton can avoid defeat here (66%), with the model siding with under 2.5 goals (51%) in a game which both teams net (53%) - 1-1.
Back the 1-1 @ 7.87/1
Another Wolves win with a shutout
Wolves vs Newcastle
After a poor few games, Wolves have once again found their stubborn defensive best, keeping clean sheets in back-to-back wins. Based on xG, they were fortunate to beat Leeds on Monday (xG: LEE 1.29 - 0.48 WOL), but again looked well organised with a solid game plan. Newcastle were thumped by Manchester United, as they once again struggled at both ends of the pitch. They have generated an average of 0.88 non-pen xGF per game, while allowing 1.52 non-pen xGA per game. Those are some really poor underlying numbers that need to be improved upon, and suggest they may struggle to trouble Wolves here. Nuno's side are fancied to win (58%), though goals don't usually follow the Old Gold (56% U2.5, 43% BTTS) - 2-0.
Back the 2-0 @ 8.07/1
Out of sorts Leicester to grab a point
Arsenal vs Leicester
Arsenal weren't blown away by Manchester City last weekend, but they never really looked like getting anything out of the game, again looking toothless in attack (xG: MCI 1.54 - 0.99 ARS). So far this season, the Gunners have generated an average of just 1.28 xGF per game, with only seven teams ranking worst in that metric. Their lack of attacking guile is highlighted by the fact that their captain and talisman, Pierre-Emerick Aubameyang, is averaging 0.07 xG/avg match... Leicester come into this on the back of successive defeats without scoring, and they too are struggling to create chances this season. They have averaged just 1.04 non-pen xGF per game, with 37% of their xG total coming from penalties this season. Improvements are needed, but I feel as though they can set up in a manner that will stifle Arsenal. The model calculates a 60% chance of the Foxes avoiding defeat, with a 58% chance of BTTS - 1-1.
Back the 1-1 @ 6.86/1
Seagulls to get the win
Brighton vs West Brom
Brighton were left frustrated once again last time out, as they drew with Crystal Palace 1-1 despite conceding no shots other than the penalty, which is an incredible feat. To me, they have been one of the most impressive teams in the league so far, with underlying numbers extremely impressive given the schedule they have faced (1.85 xGF, 1.33 xGA pg). West Brom got their second point of the season against Burnley last Monday, a very dull but even game of football (xG: WBA 1.06 - 1.14 BUR). It was their best defensive display of the season, but overall, the Baggies have been worryingly poor (0.57 xGF, 2.31 xGA pg). Brighton are a much better attacking team than Burnley, so West Brom will be tested more severely this Monday, and the Seagulls are taken to win (47% BHA) a low-scoring affair (54% U2.5) - 2-0.
Back the 2-0 @ 9.417/2
Spurs to hang on this time
Burnley vs Tottenham
Burnley picked up their first point of the season against West Brom last week, but it was another sub-par performance by Sean Dyche's side. While they haven't yet go going from a results perspective, their process isn't as bad as those results would suggest (0.88 xGF, 1.21 xGA pg). Tottenham have started to impress me in attack, looking lethal and creative going forward, but defensively I still have doubts. The way in which they crumbled against West Ham last weekend was hugely concerning, conceding three goals from the 80th minute onwards, seemingly switching off mentally. That can't continue to happen, not when they are getting themselves in good positions. Their process is strong through five games (2.49 xGF, 1.48 xGA pg), and I see them getting the win at Turf Moor (40%) in a high-scoring game (53% O2.5, 57% BTTS) - 1-2.
Back the 1-2 @ 8.27/1
Infogol's Correct Score P+L
Aston Villa vs Leeds: Back the 1-1 @ 7.87/1
West Ham vs Manchester City: Back the 1-3 @ 11.5
Fulham vs Crystal Palace: Back the 0-1 @ 8.415/2
Manchester United vs Chelsea: Back the 2-1 @ 11.010/1
Liverpool vs Sheffield United: Back the 3-1 @ 11.010/1
Southampton vs Everton: Back the 1-1 @ 7.87/1
Wolves vs Newcastle: Back the 2-0 @ 8.07/1
Arsenal vs Leicester: Back the 1-1 @ 6.86/1
Brighton vs West Brom: Back the 2-0 @ 9.417/2
Burnley vs Tottenham: Back the 1-2 @ 8.27/1