The table tells you who has the points. It rarely tells you who has played the best — and the gap between those two things is where most of the interesting analysis lives.
Football analytics has gone from fringe curiosity to mainstream vocabulary in barely a decade. Expected goals appears on Match of the Day. Pundits who once sneered at “the laptop boys” now reference xG without flinching. But for all that the language has spread, a surprising number of the underlying truths it reveals are still routinely ignored — by fans, by media, and sometimes by clubs themselves.
What follows are five statistical patterns that survive the test of multiple seasons. None of them are season-specific hot takes. They are structural features of the Premier League that show up year after year, and understanding them changes how you read a table, a result, and a run of form.
1. Teams that overperform their xG regress — almost without exception
This is the single most reliable finding in football analytics, and it is still the most widely resisted. Over a meaningful sample, teams that score significantly more than their expected goals tend to come back towards the mean, and teams that score fewer than their xG tend to improve.
The mechanism is straightforward. Expected goals measures the quality of chances a team creates and concedes. Goals are a noisy, low-frequency outcome layered on top of that underlying chance quality. A side converting at an unusually high rate is either enjoying a hot streak from its finishers or benefiting from variance — and variance, by definition, does not last. Over a full season the noise washes out and results drift towards what the chance quality predicted all along.
The practical implication is that a team riding high on a string of one-goal wins built on modest underlying numbers is fragile, even when the table flatters them. Conversely, a side stuck in mid-table despite generating strong chances and limiting good ones against is usually a few weeks of normal finishing away from climbing. The table shows you where teams are. xG shows you where they are heading.
2. The league table systematically lies about team quality
Building directly on the first point: the final league standings are an imperfect representation of how well teams actually played. Researchers studying this have given it a name — ranking uncertainty — and the gap between a team’s xG-derived “deserved” position and its actual finish can be several places in either direction.
A defensive error in the 90th minute costs the same points as one in the 5th. A wonder goal from 30 yards counts identically to a tap-in. A team can dominate a match by every underlying measure, lose 1-0 to a deflected counter-attack, and the table records only the three points lost. Over 38 games these distortions partially cancel out, but they never fully disappear, and in any given season at least two or three sides finish meaningfully above or below where their performances warranted.
This is why a smart analyst never treats the table as the final word on quality. It is the scoreboard. The performance lives underneath it, in the chance data, and the two only loosely agree.
3. Possession is a tactic, not a virtue
Few myths are more durable than the idea that controlling the ball controls the game. The data has been awkward about this for years. The correlation between possession share and points is real but weak — far weaker than fans assume, and weaker in the Premier League than in most other major leagues.
The reason is that possession is a means, not an end. Pep Guardiola’s Manchester City use possession to suffocate and create. But plenty of mid-table sides have posted high possession numbers while doing nothing dangerous with the ball — passing sideways in front of a compact defence, generating volume without threat. Meanwhile, several of the most effective teams of the past decade have won by ceding the ball deliberately and attacking the spaces left behind.
What actually correlates with winning is not how much you have the ball, but the quality of what you do with it — and crucially, the quality of chances you concede when you don’t. A team obsessing over possession percentage is measuring its process, not its output.
4. Set pieces are worth roughly a third of everything
Here is a number that should reshape how clubs allocate coaching time: set pieces account for around 30% of all goals scored in the Premier League. Corners, free-kicks and throw-ins that lead directly to a goal make up a far larger share of scoring than their share of attention.
The clubs that have understood this — Arsenal’s recent investment in a dedicated set-piece coach being the most visible example — have turned dead-ball situations into a repeatable, low-variance source of goals. It is one of the few areas of the game where preparation reliably converts into points, because unlike open play, a set piece is a controlled restart that can be drilled and rehearsed.
For the analyst, set-piece dependency cuts both ways. A team scoring heavily from set plays has a sustainable edge if the underlying routines are good. But a team whose open-play numbers are poor and whose goals come almost entirely from corners is more fragile than its goal tally suggests — strip out a couple of set-piece specialists and the attack can collapse.
5. Home advantage is real, shrinking, and unevenly distributed
Home advantage in the Premier League has been a measurable phenomenon since the league’s inception — historically home sides have taken comfortably more than half of all available points. But two things complicate the simple version of the story.
First, it has been gradually declining. Improved away-day logistics, better pitches, and the professionalisation of preparation have all narrowed the gap that crowd, familiarity and travel once opened. The behind-closed-doors period during the pandemic offered a natural experiment, and home advantage demonstrably weakened without supporters present — strong evidence that the crowd effect, particularly its influence on refereeing decisions, is a real component rather than folklore.
Second, it is not evenly shared. Clubs with large, intense home supports and distinctive grounds extract more of it than clubs playing in half-full modern bowls. Treating “home advantage” as a single league-wide constant, the way most casual analysis does, misses that the size of the edge varies considerably from one fixture to the next.
Reading the league with these in mind
Put these five truths together and a different way of watching the Premier League emerges. You stop reacting to single results and start looking at the run of underlying performance. You treat a team’s league position as a hypothesis to be tested against its chance data, not as a settled fact. You discount possession dominance and pay attention to chance quality. You watch how a side scores, not just how often.
The good news is that the data required to do this is now widely and freely available. Detailed Premier League statistics — expected goals and xGA, home and away splits, clean-sheet records, scoring trends and form guides — can be found on free platforms covering the Premier League in depth, the kind of breakdown that was the preserve of professional analysts not so long ago. Anyone willing to spend twenty minutes a week looking past the scoreline can build a far more accurate picture of the league than the table alone provides.
None of this strips the romance out of football. The deflected winner, the underdog who defies their xG for a glorious month, the wonder goal that no model saw coming — those are still what make the game worth watching. The statistics simply tell you which of those moments are likely to last, and which are borrowing against a future correction. The table tells you the score. The numbers tell you the story.
