We take a look at the
expected points data for every Premier
League team in 2025-26 to see if there is more to the current
league table than meets the eye.
It’s often said that no attention should be paid to the league
table until at least 10 games into a season.
Well, now 11 games into 2025-26 in the Premier League, we can
then read into what the current table tells us.
After all, in 11 of the last 20 Premier League seasons, the team
who is top of the table at this stage has gone on to win the title
(55%). Teams often tend to be in roughly the position they will end
the season.
However, there are plenty of exceptions; more than enough to
conclude with absolute certainty that nothing should be taken for
granted just yet.
Last season, for example, Crystal
Palace were 18th after 11 games but finished 12th. Spurs were
10th at this stage, three points off third, but finished 17th, 33
points behind third place.
Perhaps then, underlying data can provide a more accurate
indicator of each team’s prospects for the season. There is more to
the Premier League table than meets the eye.
There are limitations to any table at this stage of the season,
not least because although 11 games is a significant number of
matches, it is only a small portion of the full season. Everyone
has played a different combination of 11 of a possible 19
opponents, so points totals or underlying numbers aren’t directly
comparable. If one team have played the 11 most difficult games of
the season, they could quite reasonably be faring worse in actual
results as well as their underlying data than a team who have
played 11 easiest opponents. Only at the 38-game mark is the table
truly fair (or as fair as it will ever be).
But that said, the underlying data behind performances can still
tell a story. Expected
goals numbers can give a better reflection of how accurately
every team’s points total reflects their displays.
Opta’s expected points model uses expected goals data to do just
that, by calculating the number of goals each team ‘should’ have
scored in a match based on the xG values of their shots. It then
uses that shot data to simulate the match 10,000 times, and each
team’s expected points are taken from the proportion of those
simulations that the match is won, drawn and lost. As real goals
are so rarely scored, expected points is a way of rewarding teams
for creating good chances and dominating games when they might not
actually find the back of the net.
It’s not an exact science, because there is no xG value to
sustained periods of territorial dominance or chances that don’t
include any actual attempt on goal. The model also doesn’t consider
game state (whether the game is being won/drawn/lost at any
particular time), which can affect which of the teams show
attacking intent and create chances.
But over an extended period – such as 11 games – it can be
illuminating, and the Premier
League expected points table for 2025-26 is exactly that.
Arsenal
lead the Premier League and deservedly so, according to the model,
which also has them in pole position. Their total of 24.1 xPts is
slightly lower than their actual total of 26, and so is their lead
over Manchester
City at the top, at 2.7 xPts rather than the four points they
lead by in reality.
What the model says is that Arsenal have deserved to win roughly
as many points as they actually have; when they have won games,
they have created enough chances to deserve to win. Indeed, they
have won eight games and have won the xG battle in nine, but that
includes their draw with City, when they essentially drew the xG
battle with just 0.02 xG more than their opponents. Essentially,
Arsenal have performed almost perfectly in line with their
underlying data.
City, in second, are the only other team who are in the same
position in both the actual table and the xPts table, with their
performances worthy of their position in the league, according to
their underlying data. However, there are big discrepancies
elsewhere.
The teams currently in fourth (Sunderland),
fifth (Tottenham)
and sixth (Aston
Villa) in the Premier League should actually be 15th, 16th and
19th based on the quality and quantity of chances they have created
and conceded. The underlying data behind their performances is far
worse than their results.
For Spurs, whose fans are largely unhappy with the quality of
football they have seen under new manager Thomas Frank, this may
not come as much of a surprise. There have been a few games this
season in which Tottenham have appeared fortunate to get a result.
Aside from a 3-0 win over Burnley in August, they haven’t played
anyone off the park and don’t tend to create much. The model
suggests what they’ve been doing – getting results without creating
a great deal – doesn’t make their league position sustainable and,
concerningly for Spurs fans, that sounds about right.
But for Sunderland, the xPts table feels unfair. Back in the top
flight after promotion last season, Regis Le Bris’ side have been a
revelation. Few would contend that they don’t deserve to be where
they are in the table.
A possible explanation for their lowly position in the expected
points table is that they spend long periods out of possession.
They have averaged 43.2% possession this season and have had as
little as 28.6% vs Aston Villa and 31.7% vs Chelsea.
When you have that little of the ball and ‘suffer’ as much as Le
Bris’ Sunderland, you are more likely to concede chances and ‘lose’
the xG battle.
But playing that way is precisely what makes this his team so
effective; it is almost by design that they have fewer chances than
their opponents, but they are brilliant at making the most of the
ones they do get.
At the other end, they have been able to rely on goalkeeper
Robin Roefs, who has prevented 2.1 goals based on the quality of
the shots on target he has faced. He has been a decent part of the
reason Sunderland have one of the best defensive records in the
Premier League.
Villa’s position is arguably the most shocking of the lot,
probably because although they didn’t have the best start to the
season, their recovery feels like it has been enough to put them
higher than second from bottom.
Their underlying numbers aren’t great, though. They failed to
score in their first four games and then failed to beat 10-man
Sunderland, and even in winning five of their next six matches,
their data isn’t entirely convincing.
Villa have managed better xG numbers than their opponents in
just three of their first 11 games. And that includes only just
outdoing Bournemouth
in last week’s 4-0 win, with 1.7 xG to their opponents’ 1.61,
suggesting they were fortunate to come away with a scoreline that
looked so convincing.
They also posted just 0.32 xG in last month’s 2-1 win at Spurs,
and 0.81 xG to City’s 1.18 when they beat Pep Guardiola’s side,
too. The numbers suggest an improvement is needed for Villa to
maintain their position.
In a more positive light, the xPts model says plenty of teams
deserve to be much higher up the table than they are.
Crystal Palace are the other biggest movers: they are currently
10th but deserve to be third according to their underlying data.
They have underperformed compared to their xG to a greater extent
than any other team in the Premier League this season, scoring 14
goals from 19.3 xG. Given they are just three points off third
place, there’s good reason to believe Oliver Glasner’s side could
soon rise up the table.
Nottingham
Forest (currently 19th but ‘should’ be 13th), Leeds
(16th, should be 10th), Newcastle
(14th, should be ninth), Brentford (12th, should be seventh), and
Brighton (11th, should be sixth) are all also at least five
positions lower than their displays deserve. However, as the below
graphic shows, Brentford and Brighton only deserve slightly more
than they have got in points terms, and move up the xPts table in
part because so many other teams have dropped significantly.
In points terms,
Wolves are the biggest movers: they deserve 11.2 points,
compared to the two they have. Things could turn for them soon
under new manager Rob Edwards.
Of course, there’s no guarantee that the underlying data will
come good. Sunderland, Spurs and Villa could easily carry on
defying the numbers to challenge for a Champions League spot.
Forest and Leeds might get involved a relegation scrap even though
they deserve to be safe. Wolves might not win a game all
season.
Or, the pull of the underlying numbers will be too strong, and
this exercise will prove prophetic. Either way, Arsenal look good
value for top spot.
Subscribe
to our football
newsletter to receive exclusive weekly content. You should
also follow our social accounts over on X, Instagram, TikTok and Facebook.