Premier League 2025/26: Goals From Open Play

There’s been changes a-doing

In my last article I shared some data around goals scored from Set Pieces in the latest Premier League season. While it is interesting that Arsenal topped the Set-Piece table, feeding the narrative among some that Arsenal have somehow cheated their way to the title, what is more revealing for me is what the Set-Piece data told us about how football is being played now.

With total Premier League goals being at their lowest since the 2020/21 campaign, it’s evident there has been a more fundamental shift in gameplay across all teams, not only Arsenal.

Rise of the Set Piece

We’ve established that in the 2025/26 Premier League season, goals were more scarce across the board. In theory, this should mean that each goal scored carries more significance. Why? Because it doesn’t require as many goals to win football matches.

Look at the Set-Piece data for the last six seasons. I picked six seasons because it brings in the 2020/21 campaign, the last time total goals scored was lower than the most recent campaign.

Set piece goals didn’t shoot up in the latest season only because of Arsenal. Everybody was at it.

We can also reasonably say that set piece goals were not responsible for the low number of total goals back in 2020/21. This again suggests, to me, that something more fundamental has happened across Premier League coaching in the latest campaign.

What it means for goals from Open Play

I’m always thinking of my predictive model when I look at these kind of stats. This week I looked at whether or not it is worthwhile building in a Set-Piece factor into the model. The long and short of it is no, I don’t think it is (see below for a little more on that).

But it did lead me down another path. Look at that chart above. 26% of all goals were from set pieces in the 2025/26 Premier League campaign. That means 74% weren’t, and this happened in the lowest scoring season in the last six.

This itself got me thinking that goals from Open Play are more valuable than previously, the logic being that it takes fewer to win games and thus climb the league table. Here is the total number of Open Play goals placed alongside the set pieces data. I’ve also added Counter Attack, Penalty, and Own Goal data too (with thanks to whoscored.com).

This is a profound shift, isn’t it?

The impact on predictive modelling

Taking this a step further, I asked a couple of obvious questions. Is there a correlation between higher Set Piece goals and Premier League position? And is there a similar correlation between Open Play goals and Premier League position?

Thanks to the magic of Excel, there a very weak correlation between Set Piece goals and final league position. This holds true for the last several seasons. In other words, scoring a high or low number of set-piece goals doesn’t appear to correlate with higher or lower league position.

However: there is a strong correlation between goals scored from Open Play and final league position. Correlation doesn’t necessarily mean causation, of course, but in this instance it does feel like there must be causation. It logically makes sense. Goals win matches. Teams that can score more goals tend to finish higher in the league table. These are established facts.

From the modelling perspective, I have tinkered around with the idea and introduced a new factor value to represent goals scored from Open Play by a team versus the league average. It has made a small but consequential impact, based on my early experimentation.

What next?

As will be the case for a few months now, my eyes are focused on the start of the new football season and how I can improve my predictive capability. I will continue to experiment with the goals from Open Play metric on past seasons to confirm if it does indeed make the initial impact I think it does.

On top of that, I’m very intrigued by the dramatic shift in Set Piece vs Open Play goals in the most recent Premier League season. This will inform the nature of my match analysis in the next campaign, while also being mindful that we might see some law changes to counteract a more rustic style of football. If fewer goals are being scored from Open Play, it must mean one of two things: 1) players have become less good at creating goal scoring opportunities; 2) teams are better organised defensively. It might even be a combo of the two.

I think I’ll look at the Set Piece data for other countries too, especially in La Liga which has arguably underperformed in European competitions this season compared to usual. Are they on the same Set Piece track as the Premier League, or will they start rough-housing from next season too?

Plenty to chew on.