By Opta Analyst | Created at 2026-09-22 12:21:27 | Updated at 2026-09-22 15:23:17
3 hours ago
We’re five games into the 2026-27 Premier League season. That gives us some data to work with, but it’s still a small enough sample for luck and variance to have played a significant role in the results. So, which teams have been fortunate, and which might feel hard done by?
Is the early-season Premier League table lying to us?
With five wins from five, Manchester City sit top as the only team with a 100% winning record. Opta’s pre-season favourites, Arsenal, are second, three points behind. Brighton and Brentford, who have both made hugely impressive starts, round out the top four.
Tottenham prop up the table, sitting bottom of the league with just two points. Fulham and Coventry City join them in the relegation zone.
But is that how the table ought to look?
In many ways, absolutely. The table is a perfect reflection of how matches have gone so far this season. The aim of the game is to win football matches, and there is no more hard-hitting proof of success than the league table.
In other ways, though, it might be slightly skewed. Some teams might feel they have performed better than their results suggest. For others, the opposite might be true.
This is where Opta’s expected points model can add a little more nuance.
The expected points model simulates the number of goals scored by each side in each match based on the expected goals (xG) value of every shot taken. It then uses those simulated goal totals to determine the outcome of the match, whether that is a win, draw or defeat. Each match is simulated 10,000 times.
The expected points for each team can then be calculated based on the proportion of simulations they win, draw or lose.
We’re aware this is not an exact science. Expected goals data doesn’t account for every factor at play during a football match, including game state and dangerous periods of possession that don’t result in shots. We’re also very much aware that real points matter infinitely more than expected ones.
Nevertheless, the expected points table can give us some interesting additional context on what we have seen across the opening five matchdays.
The Title Race
The first major talking point is that Arsenal sit top of the expected points table.
They remain favourites to win the title according to the Opta supercomputer, and their underlying numbers are still strong. They have the highest expected goals difference (+4.44) of any side in the Premier League.
Man City are second according to the expected points model. They have made a hugely impressive start under new manager Enzo Maresca, but the model suggests they have collected 5.8 more points than the underlying data from the chances they have created and conceded would normally warrant.
Sunderland generated 3.57 expected goals against them on MD5, the third-highest figure ever recorded (since 2013-14) by a losing team in a Premier League game, while the chances in City’s game against Manchester United (won 1-0) were essentially even.
There is important context to add here, though. City have spent a greater proportion of their game time leading (52.5%) than any other Premier League side this season. That means they have had less need to chase matches and push numbers forward, which could naturally reduce their overall xG output.
That’s something we need to bear in mind when analysing the expected points table, and one of the model’s limitations. A team protecting a lead doesn’t necessarily need to generate chances at the same rate as one searching for a goal.
The Biggest Underperformers
So, which teams might feel they should be higher up the league?
Michael Carrick has come in for plenty of criticism following Manchester United’s poor start to the season. United have taken just five points from their opening five league games, their joint-fewest at this stage of a Premier League campaign alongside 2014-15.
Their underlying numbers, however, are far more encouraging. United lead the Premier League for total shots (113) and touches in the opposition box (182). Their total xG of 9.89 is the fourth-highest in the division.
The expected points model has them fourth overall, with 8.4 expected points. In reality, they sit 12th. That eight-place underperformance is the largest of any team in the division.
Their problem has been converting those promising attacking numbers into goals. United have converted just 7.1% of their shots, the fifth-lowest rate in the division. The chances are there, but they have not been clinical enough.
Next are Tottenham, who head into the international break in dead last. They are bottom after five matches for the first time since the 2008-09 season.
The underlying data paints a slightly more favourable picture of their predicament. Spurs sit 12th in the expected points table, eight places higher than their actual position.
Roberto De Zerbi’s team were dire in their 3-0 defeat to Brentford and disappointing in a goalless draw with Everton, but arguably deserved more from home defeats to Newcastle (2-0) and Aston Villa (3-2).
Across those two games, Spurs generated 2.68 xG while conceding chances worth only 1.85 xG. They failed to take early opportunities in both matches and were punished.
Now, 12th isn’t exactly a position Spurs would be happy with, but it would be a lot more palatable than the position they currently find themselves in.
Bournemouth and Sunderland are both six places lower in the actual table than they are according to expected points.
For Bournemouth, the issue has not been generating good chances. Instead, it has been their failure to hold on to leads.
The Cherries have led in four of their five matches this season but have dropped points in every one of them. They have already surrendered nine points from winning positions and really ought to have won more, particularly in their 2-2 home draw with Brentford, when Kevin Schade scored two superb but highly improbable finishes for the visiting Bees.
After a season in which they rode their luck at times, Sunderland are now feeling the sting of variance in the opposite direction.
Last season’s expected points table had them closer to relegation than Europe. This time around, they sit 14th in the actual table but are six places higher, in eighth, according to expected points.
The Black Cats have scored just six times from 9.77 xG. That underperformance of -3.77 goals is the second-largest in the league, behind only Coventry City (-3.88).
They had the better of the chances against Brentford in a 1-1 draw, missed a penalty while the scores were level in their 2-0 defeat to Arsenal, and produced an excellent attacking performance despite losing 5-3 to Manchester City.
The Biggest Overperformers
Plug your ears and close your eyes, Newcastle fans.
At this early stage of the season, the expected points model has the Magpies in 20th place.
In reality, they are ninth. That 11-place gap is the largest difference between a team’s actual and expected position in the division.
It’s worth saying that the difference between Newcastle’s actual and expected points totals is 3.99, which is smaller than Manchester City’s 5.82-point overperformance.
The reason the positional difference is so dramatic is simply that teams in the lower half of the table are much more tightly bunched together. At this stage of the season, a swing of four points can lead to a lot of positions changing hands,
Still, there is some cause for caution in Newcastle’s underlying numbers. They have the second-worst expected goals difference in the division (-3.76), ahead of only Aston Villa (-4.35).
And, while nothing irritates football fans quite like this statement, Newcastle have ‘lost’ the xG battle in all five of their Premier League matches so far. Despite that, they have emerged with a W2 D2 L1 record.
Again, though, we can’t ignore game state. Newcastle have spent 45.5% of their playing time in the lead this season, the joint-second-highest proportion of any side. That has undoubtedly influenced their attacking numbers, potentially curbing their need to push forward and generate chances once ahead.
Hull City, meanwhile, continue to defy the xG gods. They have an identical record to Newcastle (W2 D2 L1) and sit one place above them in eighth. In the expected points table, however, they are 16th, with a projected tally of 5.18 points.
In fairness, that would still represent a respectable return for Hull, who many had written off before the season began and who have faced the sixth-hardest start to the campaign according to our analysis before the start of the season. They have already played Manchester United and Chelsea, taking four points from those two matches with a 2-0 win and 2-2 draw, respectively.
Once again, we aren’t suggesting expected points should be treated as a substitute for the real Premier League table. It certainly should not.
But what it can do is add context. Five games into the season, expected points can give be revealing in comparing performances on the pitch with the points on the board.
Enjoy this? Add Opta Analyst as a preferred source by clicking here.
Enjoy this? Subscribe to our football newsletter to receive exclusive weekly content. You should also follow our social accounts over on X, Instagram, TikTok and Facebook.
