Gunners to win Premier League. Probably.

Will Wagstaff • 1 February 2023

Premier League title is Arsenal's to lose

With the Premier League transfer window now closed thoughts turn to who will win the title, which teams will qualify for European football and which will be relegated.

We have developed a predictive model which helps us answer these questions.

Our predicted final Premier League standings is based upon probabilities. To generate the probabilities we use results from this season and previous ones to model the number of goals scored in a match as a Poisson distribution (the probability for a team to score one goal, two goals, three goals and so on in a match). We then identify parameters to represent a team’s attacking strength, defensive strength and home advantage. Applying this model to the remaining Premier League fixtures we can predict the final standings probabilities.

The probabilities suggest that Arsenal will win the league from Manchester City. We give the Gunners a 56% probability of winning with City on 43%. These are decent odds in the Gunners' favour and the title really is theirs to lose. 

We predict that Manchester United and Newcastle will join them next season in the Champions League. We predict that Southampton, Everton and Bournemouth will go down. 


We have also drawn up a heatmap showing the probabilities for the position each team is likely to finish in. This shows that Manchester United has a 37% probability of finishing 3rd ahead of Newcastle. The Magpies are predicted to finish 4th but the model is close as to whether they finish 3rd, 4th or 5th.


Usual Champion League stalwarts Liverpool and Chelsea are not predicted to finish in the top 4. Of the two, Liverpool has the highest probability of qualifying at 39%. Chelsea is predicted to finish in 7th or 8th place. At this point it should be pointed out that the model does not take into account the impact of transfers in or out of the squad during January's window. We saw last year that Newcastle's spending helped improve their league position and with Chelsea having spent heavily in the window, the same could happen this year. Injuries, managerial changes and players coming in and out of form will also have an impact on where teams finish.


Premier League Probabilities Heatmap 


At the bottom, although the model predicts Southampton, Everton and Bournemouth will be relegated, it will be very close. Nottingham Forest, Wolves, Leeds, Leicester and West Ham are all in the relegation mix.


We will update the model in April and publish the results then. Last season our model successfully predicted the winner, Champions League placings, Europa League and Europa Conference qualifiers plus the bottom 3 to be relegated. All in the correct order. 

Back of the net. 



by Michael Wagstaff 2 June 2025
With the insights industry creaking under poor quality data, generative AI adds to the mix by creating convincing synthetic respondents. How should we treat these fascinating fakers?
by Michael Wagstaff 26 May 2025
Do tech people know more about market research than social scientists? In this article we discuss whether big tech is taking over the insights industry and what role is there for human researchers.
by Michael Wagstaff 12 May 2025
Does curating a personal brand make a difference? In this article, I consider my own lack of brand image and reflect on what might have been.
by Michael Wagstaff 9 May 2025
If surveys are ill-suited for addressing the questions that remain genuinely unanswered and unnecessary for those that are readily answered, then what exactly is their future?
by Michael Wagstaff 8 May 2025
In an effort to improve quality, the market research industry's current focus is on survey length and question engagement. This overlooks a more fundamental issue: respondent authenticity.
by Michael Wagstaff 7 May 2025
The true value of review sites lies in going beyond the stars and analysing what reviewers are actually saying.
by Michael Wagstaff 28 April 2025
Predictive analytics is reshaping brand insights by moving beyond descriptive dashboards to proactive decision-making
by Michael Wagstaff 28 April 2025
The uncovering of large scale survey fraud has shaken the market research industry. In this article, we discuss how a much more rigorous panel sign up procedure is needed to reduce the impact of fraudulent respondents.
by Michael Wagstaff 8 April 2025
The huge volume of data available through consumer comments, reviews and surveys can make cutting through the noise difficult. In this article we discuss how text analytics combined with human expertise can uncover the insight.
by Michael Wagstaff 10 March 2025
Market research agencies are going all in on AI based models to generate next level consumer insight. But are these just more illusion than substance?
Show More