Goals, Minutes and xG: Reading a Striker

Part One · Three counts

A striker's season can be summarized in three different ways, and the three do not always agree. The first is the count of goals, the second is a rate such as goals per 90 minutes, and the third is expected goals, an estimate of what the chances were worth. Erling Haaland of Manchester City is a particularly useful case because all three are published for him. The first of them, his 36 league goals in 2022-23, was still the Premier League record in early November 2026. Each measure answers a different question, and this book takes them in turn and then dates every record.

Begin with the count, and with what the count is counting. Under Law 10 of the Laws of the Game, a goal is scored when the whole of the ball passes over the goal line, between the posts and under the crossbar, provided the scoring team has not committed an offense. The Premier League reported on May 28, 2023 that Haaland had ended the season on 36 goals, from 35 appearances. Across every competition that season, Opta Analyst counted 52 goals, a figure that a league-only record deliberately leaves out because the Champions League and the cups keep their own separate tables of goals scored.

Part Two · Denominators

A record is only as meaningful as its denominator, the number it is divided by or set against. The earlier league mark of 34 was set in 1993-94 and matched in 1994-95, in seasons of 22 clubs and 42 matches each. The league then shrank to 20 clubs, and in 2017-18 a player scored 32, then the record for a 38-game season. Haaland passed that with his 33rd goal in April 2023 and finished on 36. Those earlier players therefore had four more matches in which to score in a single season than anyone has now. That makes the old totals harder to compare than they first appear to a casual reader.

Rates repair part of that problem. A figure per 90 minutes is a total divided by the minutes played and multiplied by 90. That lets a substitute who plays 20 minutes and a starter who plays them all be set side by side. FotMob's page, through November 4, 2026, showed Haaland with 9 goals in 800 minutes in 2026-27. By this book's arithmetic, 800 divided by 90 is about 8.9 ninety-minute units, which gives roughly 1.01 goals per 90, or a goal every 89 minutes. Nine full games would be 810 minutes, so on average he played about 89 minutes a match, and the rate is a reasonable summary of nine starts, though a small one.

Part Three · Expected goals

Now the estimate. Opta Analyst defines expected goals, or xG, as a measure of the quality of a chance, produced by calculating the probability that it will be scored from information about similar shots in the past. The scale runs from zero, a chance that is impossible to score, to one. The inputs include the distance to the goal, the angle to the goal and the type of shot, such as a header or a volley. Opta's model is powered by nearly one million shots from 40 competitions between 2018-19 and 2021-22. Every penalty receives a constant value of 0.79, which reflects how often penalties have historically been scored.

The method has limits that its own users state. Sky Sports notes that xG struggles to account for where the defenders are standing, and that a model built only from shot data can miss chances that never became shots, such as a cross that no one reached. It adds that some players have beaten their expected goals over several seasons in a row. That suggests the model may not capture every skill of finishing. Companies also build their own models. This book's 2026-27 figures are FotMob's and its earlier seasons are Opta's, so the two sets should not be added together or compared to the decimal.

Part Four · Testing the model

Opta's per-90 table, published on October 15, 2025, lets us test the idea on one player. In 2022-23, Haaland averaged 1.17 goals and 0.92 xG. In 2023-24 the pair was 0.95 and 1.03, and in 2024-25 it was 0.72 and 0.72. Subtracting gives +0.25, -0.08 and 0.00 goals per 90 against expectation. One season sat well above the model, one slightly below it and one exactly level, so the model describes the chances reasonably well while the finishing around it moves from year to year. A gap of 0.25 goals per 90 is a quarter of a goal in each full match, which over 35 matches would add up to about nine goals if he had played every minute.

The same table lets us take xG apart. Expected goals per 90 equals shots per 90 multiplied by xG per shot. For 2022-23 that is 4.0 shots at 0.23 each, which gives 0.92. For 2023-24 it is 4.3 shots at 0.24, which gives 1.03, and for 2024-25 it is 3.6 shots at 0.20, which gives 0.72. A striker can therefore raise his expected goals in two separate ways, by shooting more often or by shooting from better places, and the table shows the two moving independently. At the start of 2025-26 both were at their highest, with 4.4 shots at 0.27. The columns do not multiply exactly, because 4.4 times 0.27 is 1.19 against a published 1.16, a small rounding difference.

Part Five · Samples

Penalties complicate every scoring rate. In 2022-23, Opta lists 1.17 goals per 90 but 0.94 non-penalty goals, so penalties supplied about 0.23 goals in each 90 minutes. In 2023-24 the gap was 0.25, and in 2024-25 it was 0.10. Because each penalty carries a high fixed xG, penalties inflate both the goals column and the expected-goals column, which is why analysts quote non-penalty figures when they want to judge a player's open play. Early in 2025-26, Opta noted, none of Haaland's goals had come from the spot. In his other fast starts, the same analysis added, he had scored at least one penalty, so that start was unusual in this respect too.

What does a striker do between his shots? Opta recorded 24.2 touches per 90 minutes for Haaland at the start of 2025-26, the third-lowest among Premier League players with at least 300 minutes. His shots in that sample were 4.4 per 90, so he took about one shot for every 5.5 touches. Coaches' Voice offers a tactical reading: a lone center-forward holds the ball up, presses the opposing center-backs and runs in behind the defense, and he can be isolated and easily overloaded, so the rare touch has to count. That 24.2 comes from the first weeks of one season, so it is a sample, not a verdict on the full year.

Part Six · Lists and open questions

Small samples are the trap in all of this kind of counting. On October 15, 2025, Opta had Haaland at 1.36 goals per 90 with no penalties, and on September 13, 2025, it had counted his shots as worth 3.86 xG, almost double those of any other player. The season ended with 27 league goals in 35 games and 8 assists, by NBC's table, which is about 0.77 goals a game. Neither of the early figures was wrong. They described a few matches, and rates drift toward the long-run level as the sample grows. The pattern also runs the other way: in 2024-25 his rate fell to 0.72 goals per 90, far below the 1.17 of 2022-23.

Here are the records as the Premier League and the press reported them, with dates. The most league goals in a season is 36, in 2022-23. The fewest matches to reach 20 league goals is 14, on December 28, 2022, against 21 for the previous holder, through November 2026. The fewest games to 100 is 111, on December 2, 2025, against 124 for the previous holder. The most goals by a player at a Premier League club in all competitions in one season is 52. Each is dated to its source and would change only if someone beats the number. Fastest-to records count matches rather than minutes or calendar days, so they measure appearances more than time on the field.

Part Seven · Today

The Golden Boot list records the same story season by season. In 2022-23 Haaland won it with 36. In 2023-24 he won again with 27, five clear. In 2024-25 Haaland scored 22. In 2025-26 he won with 27, announced on May 24, 2026, his third; the record for Golden Boots is four. The winning totals since 2022-23 have been 27, 29 and 27, so 36 has not been challenged, though three seasons are too few to say whether 36 was an outlier or a new level, a question that only more seasons can answer. The 2025-26 total of 27 came in 35 games. Each Golden Boot counts only Premier League goals, so the 12 Champions League goals that made him that competition's top scorer in 2022-23 never enter it.

Even the largest tallies contain disagreements. NBC's career page, dated November 4, 2026, gives 175 goals for Manchester City, yet its own season rows add to 176, and the page does not explain the difference. For 2026-27, FotMob's page, through November 4, 2026, listed 9 goals from 6.57 xG, 34 shots and 16 on target in 9 matches, which is 2.43 goals above expectation over a tiny sample. Whether that gap is skill, luck or both is an open question, and nine games cannot settle it. For Norway, NBC lists 65 goals in 60 caps as of October 11, 2026, and one report said seven came at the 2026 World Cup, where Norway lost 2-1 to England after extra time in the quarter-finals in July 2026. None of those enter the club figures above.