Streaky Scorers

Measuring Consistency Among Premier League Goal Scorers

Can a player who scores 15 goals be more valuable than a player who scores 20?

Yes.

Football is a game where every win is decided by one goal, and the most common scoreline is 1–0.

If our 15-goal striker scores his 15 goals as single contributions across 15 games, he has given his team a chance to win as many as 45 points (if every game ends 1–0).

If our 20-goal striker netted three hat-tricks and scored twice in another game, he will have made goal contributions in 13 games, 2 fewer than our 15-goal man and therefore potentially worth 6 fewer points in the standings.

In other words, smaller scoring intervals and consistent production have greater value than spread-out clusters of explosive scoring.

Most of the time, more multi-goal games means more droughts between those games.

You can’t score a lot of hat tricks and braces and also score a lot of goals overall, unless you’re Erling Haaland or Mo Salah, and most guys aren’t.

So let’s say you’re a middle-weight Premier League team and you have the opportunity to buy a striker.

The league average Goals-per-Game (GpG) among the top scoring players is 0.38, but this is a misleading metric in this study as it makes no distinction for scoring intervals.

Let’s imagine a player’s “chance of scoring a goal in a game” as CoS. This is the number of games in which the player scored divided by the total games played.

The average CoS among top scoring players is 0.31, or a 31% of scoring in any given league game.

If our example players from earlier played 38 games (a full Premier League season), they would get:

(Where Potential Point Value of Goals = scoring games x 3 points for a win)

Choosing between these two, you might be seduced by that 0.53 GpG from the 20-goal guy, but that’s also the reason he’ll probably cost you a lot more. And you’ll potentially be getting 6 points less value for your extra investment.

Imagine both players had scored the same number of goals but the former 20-goal guy is now a super streaky 15-goal guy who still scored three hat-tricks and a brace. This would mean he only scored in 8 separate games.

With the same overall goal totals, their price might be similar. There’s no argument who you’re buying this time:

The Study

I wanted to see how this concept applies to real world players.

I gathered data on the top scoring player on each of the 20 Premier League teams for the last 5 seasons (2021–2025 inclusive, 68 players total).

Perennial Premier League teams (e.g. Man City) will therefore have more player representation than teams who only featured in the league during one of those seasons (e.g. Luton), but every team has some inclusion.

To qualify, a player has to have featured in at least half the available games for each season (19). This is why Allan Saint-Maximin represents Newcastle for 2022 rather than Callum Wilson, as Wilson only played 15.4 90s that season.

The players are not exclusively strikers. For example, James Ward-Prowse (Southampton, 2022 & 2023) and Philip Billing (Bournemouth, 2023) are midfielders, but were their teams top scoring player in those seasons.

Here’s how all 68 players in this study shape up:

[embed]

The streakiness axis shows the coefficient of variation of scoring intervals.

In other words, it measures how predictable the time between a player’s goals is — higher values mean goals come in bursts with long droughts, lower values mean goals arrive more evenly.

The extremes of this study are 0.30 (steady end) to 1.45 (streaky end).

The quadrants show:

Low Streakiness + High Chance of Scoring = Consistent Goal Threats

This is where the elite strikers live.

Low Streakiness + Low Chance of Scoring = Steady Contributors

These guys aren’t scoring big numbers but they can be relied upon to contribute.

High Streakiness + Low Chance of Scoring = Streaky Scorers

These guys are a crapshoot. Neither prolific nor predictable.

High Streakiness + High Chance of Scoring = Erratic Goal Threats

These guys know where the net is, but they’ll ruin your fantasy team.

The closer the player to the middle of the chart, the less pronounced the effect.

Insights: Consistent Goal Threats

The standout names here should be no surprise: Kane, Haaland, Isak, Salah.

• Solanke, Barnes, Cunha and Mbuemo won their big-money moves on merit.
• Tottenham really miss Son, although Solanke’s return from injury could dry those tears.
• Patrick Bamford could be about to destroy the Championship.
• Jarrod Bowen gives you plenty of reliable bang for your buck.

Let’s look at Bowen. His 0.36 chance of scoring is very close to his 0.39 GpG, which means he routinely spreads his goals over as many games as possible.

This means the potential point value of his 41 goals is 111 from a maximum available of 123 (since he scored 41 goals in 37 games, not 41 games).

Scaled to a single season, that’s a potential point value of 42 (avg. 14 scoring games) from a maximum available of 45 (avg. 15 goals scored). That’s very efficient, a “potential points per goal” figure of 2.7 (maximum of 3).

For context, in 2023 Erling Haaland scored 36 goals in 35 appearances, which is a phenomenal output. However, that includes four hat-tricks and one brace. He only actually scored in 23 of those 35 games.

His potential point value of 69 is way short of the maximum available of 108. That’s a “potential points per goal” figure of 1.9. Multiple goals scored in the same game become “cheaper” as they will have less effect on the result.

When a player can score as many as Haaland few are going to care about this nuance, but it illustrates Bowen’s high-value as a scorer.

Am I saying Jarrod Bowen is better than Erling Haaland? No. But if you’re shopping on a mid-level budget for a mid-level goal scorer, you could do far worse than buy Jarrod Bowen.

He gets very close to maximising the point potential of his goal contributions, and that becomes extremely important assuming your team can’t find or afford an over-powered player like Haaland (and few can).

Insights: Steady Contributors

• McGoldrick is the McGold-Standard of steadiness. He only scored 8 goals, but they came almost like clockwork every 5 games or so.
• Chelsea took a risk buying Joao Pedro for £55M. He’s quite streaky and doesn’t score enough to justify that price tag.

Insights: Streaky Scorers

• Calvert-Lewin looks like the most frustrating player in the league, but his status as the streakiest scorer of all is the result of an 18-game-long scoring drought in 2024.
• This quad is home to the “just below bang-average striker stack” — that tight grouping of five players arranged in a column. Richarlison, Dennis, Delap, Awoniyi and Pukki all share a 30% chance of scoring in a given game, with an overall GpG of between 0.30–0.37. Side eyes towards Chelsea again for dropping another £30M on Delap.

Insights: Erratic Goal Threats

• Mitrovic would be in the Consistent Goal Threats quad if he hadn’t endured a 6 game drought at the end of the season. He scored 11 of his 14 goals in 10 separate games over the first 15 matches — elite performance as far as this study goes.
• Palmer slips into the pack when looking at his combined totals, but looking at separate seasons we see two totally different players: 2024 Palmer the Consistent Goal Threat, and 2025 Palmer the Streaky Scorer, his streakiness of 1.07 second only to Calvert-Lewin’s chart-breaking 1.45 after suffering a 12 game scoring drought.

Takeaways

Next time your team is considering buying a striker, look at the games in which he doesn’t score, how far apart they are, and how regular the tempo is. Small intervals, regular tempo? He’ll help your team more than you might imagine. Big intervals, erratic tempo? Get ready to smash your tv set.