What is the most common Premier League scoreline? What should you expect to see more of in a game, fouls or shots at goal? Which team is more likely to have a player sent off, the home side or the visitors?
The football fans among you will no doubt have their answers to these questions ready by the end of this sentence. In fact, I reckon that even some of the casual fans will be giving these questions a moment’s thought. If you want to know the answers to these questions – along with many other Premier League (PL) stats – then the accompanying infographic is for you. If, however, you’re slightly curious as to how we got these numbers in the first place, read on.
Still, before we get to the good stuff, we need to go through the proper introductions.I work as an analyst for Optalitix, a company that delivers cutting edge, data based services, including database management and data analytics. I’m also a long-time supporter of a certain PL football team (name withheld for feeble reasons) and if there’s one thing I’ve learnt, it’s that there’s no quicker way than to win a tribal argument lively debate with a rival fan than with cold hard stats.
Why did we carry out this analysis into PL data? There are a number of answers. Football, like all sports, is awash with randomness and probabilities – being the data nerds that we are, the idea of being able to explain away the randomness, and quantify the unquantifiable, piqued our curiosity. We’d also recently finished work on an API modelling tool and were keen to see its application on something other than client data. Football data also has the advantages of being plentiful (if you’ve ever watched a football game on TV, you’ll have seen how frequently statistics flash up on the screen), not confidential and easily understandable by almost anyone who doesn’t live under a rock (or in the USA).
Besides, with the new Premier League season kicking off, could you really blame us for getting a little carried away with the whole football theme?
In order to carry out our research, we obtained historic data for the past 8 PL seasons (2010/11 – 2017/18). We didn’t go any further back – to, for instance, the inception of the PL in 1992 – because we assumed the relevance of the data decreases the further into the past you go (much to the chagrin of all Liverpool/Arsenal/Manchester United supporters).
Once we had compiled the data, we were able to look for key trends and patterns in the data (and settle some age-old debates with various colleagues). The fabled “home advantage” seemed like a good place to start - 57% of all goals are scored by the home side, and victory for the home team is the outcome of 46% of all matches. Needless to say, we soon found ourselves pretty far down the rabbit hole – did you know that Crystal Palace would have been relegated last season if all matches ended at half-time? And that they would have been just 1 point away from European football if only the second half of games counted?
The most common PL scoreline, by the way, is 1-1. I’ll leave it up to you to decide why that is.
Yoni Greis
What Industries Could AI Impact the Most?
From virtual health care assistants to driverless taxis, AI is transforming the way the world operates, making processes more productive and efficient. Read more.
How AI Is Used in Our Everyday Lives
AI permeates several aspects of our everyday lives, from our from smart phones to shopping and much more. Learn more about the impact of AI here.
How Is AI is Changing The World?
Artificial Intelligence (AI) has the power to change the world across multiple sectors including business, healthcare, transport and education. Learn more here.
How Could AI Change Future Workplaces
AI has the power to change the future of working globally. Read more about how AI can assist with staff collaboration, hiring employees, and remote working here.
How Can Machine Learning Help My Business?
Machine learning can help your business thrive by processes such as automating decisions, mundane tasks, and security. Find out more here.
Hot Models
During Covid-19, AI or non-human solutions were being used to model potential outcomes and solutions. Read how this was done and the findings that came from AI.
Optalitix selected to participate in Lloyds Lab
The Optalitix team are delighted to have been selected to participate in cohort 4 of the Lloyds Lab. Read more about their application and success in this guide.
How do machine learning models work?
There are a variety of regression and classification models within both the supervised and unsupervised categories of machine learning, a type of AI.
Will machine learning replace humans?
Machine learning technology has already started to replace humans, yet it has potential to create many opportunities and enable people to do their job better.
Is AI taking jobs?
Developments in AI led to machines taking predominantly unskilled jobs, and this looks set to continue. Learn more about how AI is impacting the workforce here.
How is AI being used in healthcare?
New AI technology from Microsoft and Google is able to detect diseases such as prostate cancer and diabetic retinopathy. Learn more the NHS's use of AI.
How is AI used in financial services?
The financial services industry is being transformed by AI. Learn how AI is used as robo-advisory services, chatbots, and algorithmic trading in finance.
What is the AI industry worth?
The AI industry is estimated to be worth somewhere in the region of $5 trillion dollars, and this number is only set to increase in the future. Learn more here.
How does machine learning work?
Machine learning algorithms are a type of AI that can learn patterns and processes with minimal to no human intervention. Discover more about how they work here.
How is machine learning used in finance?
The financial sector is increasingly using machine learning in order to be more accessible, efficient and provide a better customer experience overall.