Jake Osgathorpe uses Infogol's expected goals (xG) model to help predict the Premier League correct scores for Gameweek 6, as well as rounding up the latest kick-off and TV information.
Scroll to bottom for this week's Premier League TV schedule in full
What a huge game this is, particularly for Manchester City who have a tough week including this game, facing Paris St-Germain in midweek before a trip to Anfield next weekend. As for Chelsea, they have already played three of last season's top eight and have the opportunity with a win here to pull six points clear of City.
Thomas Tuchel's side were excellent at Spurs last weekend, blowing their hosts away in the second half to record an easy and deserved victory (xG: TOT 0.78 - 2.49 CHE). Defensively they are immense, allowing over 1.0 xGA (Expected Goals Against) only twice this term - at Anfield and in a 3-0 win over Aston Villa - picking up from where they left off last season. We have seen improvements in attack too, with the addition of Romelu Lukaku having the desired effect.
Manchester City were surprisingly held to a goalless draw by Southampton last weekend, though they did create enough chances to warrant the victory (xG: MCI 1.47 - 0.42 SOU). Key for City heading to Stamford Bridge though is their form defensively, where they boast the best defensive process in the Premier League, allowing just 0.56 xGA per game.
Matches between these two teams since Tuchel arrived in England have been very tight, with Chelsea winning all three by a single-goal margin. Given this is a must-not-lose match for City, a draw looks likely, and despite these teams being the best defensive sides in the league, a goal apiece wouldn't be a surprise.
Manchester United's Carabao Cup run was ended before it started on Wednesday by West Ham, but they did beat the Hammers 2-1 in the Premier League last weekend and deservedly so despite Mark Noble's last-gasp penalty miss (xG: WHU 1.67 - 2.13 MUN). In fact, United have impressed on underlying numbers this term, winning the xG (Expected Goals) battle in four of five matches and allowing 1.17 non-pen xGA per game while racking up 1.86 xGF non-pen per game.
Aston Villa have been steady thus far, winning two and losing two of their first five, but they have been going about things with a defence first attitude as their attack is yet to click into gear. Only Manchester City (2.8 xGA) and Brentford (4.3 xGA) have allowed fewer xG than Villa (4.9 xGA), so they will prove a tough nut to crack at Old Trafford.
Don't be alarmed Everton fans. A 3-0 defeat to Aston Villa was an undeserved loss based on expected goals (xG: AVL 0.97 - 0.90 EVE), and while Dominic Calvert-Lewin, Richarlison, Seamus Coleman and Jordan Pickford are set to miss this game, the Toffees' strength in depth and underlying process means they have a great chances of getting the win.
Norwich have lost 15 straight Premier League games over two seasons following their 3-1 loss to Watford. That was the most disappointing defeat of all - having lost their first four against teams who finished in the top eight last season, a home game against a fellow promoted side was their easiest match to date. The manner of the defeat was hugely concerning (xG: NOR 0.75 - 2.92 WAT), with Daniel Farke's continuing to look hopeless at the back (2.38 xGA per game).
Leeds don't look like themselves this season. Last term their vibrant attack consistently bailed out a vulnerable defence, averaging 1.63 xGF per game, but so far this season they aren't hitting those same heights (1.15). This means that results are hard to come by, with Marcelo Bielsa's side shipping chances and, unless something changes, expect them to continue to struggle.
West Ham will be kicking themselves for missing a late penalty against Manchester United, but a defeat was a deserved result, with David Moyes's side struggling in attack in the absence of Michail Antonio, mustering just 0.89 non-pen xGF. Antonio is back for this match, and the Hammers' attacking process should prove too strong as they get back to winning ways.
Leicester simply haven't got going this season, with three defeats in five league games, but importantly Brendan Rodgers's side have lost the xG battle in all five. The Foxes were beaten 2-1 by Brighton last weekend, again conceding good chances, with their season average at 2.06 xGA per game. Worryingly, only Spurs and Norwich have a worse attacking process than the Foxes (1.02 xGF per game).
Burnley have fewer points on the board than their hosts, but their performances have been better, winning the xG battle in three of five games, with four matches being extremely tight. They were unfortunate to lose 1-0 to Arsenal, and have been creating more chances than their goal tally suggests (1.49 xGF per game). The Clarets can get a result at the KP.
Watford got their second win of the season at Norwich last weekend, a really good performance, though it did come against a team that looks like the worst side in the Premier League (xG: NOR 0.75 - 2.92 WAT). The Hornets' underlying process has been steady thus far, but defensively they have looked vulnerable (1.50 xGA per game).
Newcastle remain winless through five league games after a 1-1 draw with Leeds, another open game with chances at both ends. Steve Bruce's side have been very open at the back, and not as potent up front as they were at the back end of last season which is a concern, but their counter-attacking capabilities should help them avoid defeat here.
Brentford have done what we expected so far this season, and that is for them to look like a solid Premier League team. They are the best of the promoted teams, and victory at Wolves last weekend was another statement of intent (xG: WOL 1.45 - 2.11 BRE). Defensively, they have been excellent this season, allowing just 0.85 xGA per game, but this is a stiff test, and their lack of creativity is a worry heading into this match (1.17 xGF per game).
Liverpool look excellent. Virgil van Dijk has made a huge impact - as expected - since returning, but the fact that Sadio Mane and Mohamed Salah got a summer off thanks to having no international commitments has meant the Reds attack has looked as sharp as their title-winning campaign. In fact, their average of 3.20 xGF per game is by far and away the best attacking process in the Premier League. They will take some stopping here.
Southampton have drawn their last two league games 0-0 against West Ham and Manchester City, two teams who finished last season in the top six. Saints have struggled to create chances in both games (0.78 and 0.42 xGF), but defensively have looked much better. Whether their approach changes for this game against a 'lesser' team remains to be seen, and if it does, we could see a more open contest.
Wolves were comfortably beaten by Brentford last weekend, in the first game where they looked vulnerable defensively under Bruno Lage (xG: WOL 1.45 - 2.11 BRE). Their xG process over the season so far makes for pleasant reading, averaging 2.00 xGF and 1.07 xGA per game. However, a failure to convert chances has seemingly been their downfall thus far.
After a poor start to the season, losing their first three, Arsenal have bounced back by winning successive games 1-0. A return to fitness of key starters has coincided with an upturn, as has an easier fixture schedule, but the signs are promising that the Gunners are now on the right track. Mikel Arteta's side allowed just 0.64 xGA to Norwich and 1.11 xGA to Burnley, suggesting they are tightening up defensively, and that bodes well heading into this north London derby.
Tottenham are in a mess. They looked competitive in the first half against Chelsea, but the second half xG total tells the story of their implosion (2nd half xG: TOT 0.05 - 2.45 CHE). That was their second successive 3-0 loss, and Spurs now sit 17th based on expected points (xPoints), showing how fortunate they are to have accumulated nine points already. Their xG process is horrendous (0.99 xGF, 2.06 xGA per game), and means there is little confidence about them getting a result on Sunday.
After beating Spurs 3-0, Palace were on the end of a 3-0 thrashing against Liverpool last weekend, getting brushed aside with ease (xG: LIV 3.30 - 0.87 CRY). Nonetheless, it has been a solid start to the campaign for Patrick Vieira's side, with their xG process drastically improved on last season (1.28 xGF, 1.37 xGA per game). They aren't creating many good chances regularly, but they aren't allowing many either.
That is the same for rivals Brighton, whose process is down on last year's but still positive. Graham Potter's side have won four of five league games following a 2-1 win over Leicester, and they have had the rub of the green, coming out on top in tight matches. Their xG process shows how close their games have been, averaging 1.20 xGF and 1.17 xGA per game. This will be another tight match.
Saturday 25th September
Sunday 26th September
Monday 27th September
We are committed in our support of safer 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 / GamCare on 0808 8020 133.