Real-time football analysis company Squawka has carried out an unprecedented experiment into how perceptions on social media are influenced by specific events in a football match.

Working in partnership with social analytics experts DataSift, they tracked online sentiment using 380,000 Tweets over the course of Chelsea's home defeat to Manchester United on 28 October.

The results were certainly more nuanced than some may have expected, with Juan Mata's goal sparking a flurry of online activity far beyond that seen at any other point of the match.

The data also shows a pattern of considerable lag after major incidents, with Twitter chatter remaining high in the minutes after Mata's free kick. The number of match-related Tweets per minute didn't drop below the 90 minute average until after half time was over.

The longest period of high-volume tweeting came in the final quarter of the match, coinciding with a frantic twelve minutes in which first Ivanovic and then Torres were sent off, and then Hernandez scored the winner.

In addition to monitoring overall activity levels, Squawka drilled down further and examined Tweets concerning ome the game's central protagonists, Fernando Torres.

For every player on the pitch, Squawka generate a Performance Score, calculated by running an algorithm on detailed performance data provided by Opta Sports.

In their experiment, Squawka decided to see how closely correlated Twitter sentiment was with this score.

In the early stages of the match Torres' contributions were generally well received. A successful aerial duel and shot on target were followed by thousands of positive Tweets mentioning the forward, and even after he collected his first yellow card, tweeting fans remained on his side.

The predominantly positive tone of Tweets over half time about Torres suggest that fans were generally happy with his first half performance.

The most interesting activity came following his sending off for a perceived dive. Unsurprisingly, the immediate aftermath was dominated by negative online reaction, which intensified when Hernandez restored United's lead and remained as an undercurrent for the rest of the match.

After full time, however, the mood changed, spurred by comments by the likes of Gary Neville, whose judgement that Torres had been perfectly entitled to go to ground was picked up and repeated on Twitter by Chelsea fans who felt hard-done-by.

Squawka's co-founder and CEO Sanjit Atwal spoke to the Guardian about the experiment and the role he feels big data can play as sport performance analysis reaches a wider audience:

This is a fantastic example of how big data can be used to enhance the experience of watching a football match. It adds a whole new layer of drama.

We've got the full text of every Tweet, which lets us take a closer look when we see an interesting trend in the aggregated data and find out exactly what is happening. Instead of wondering what triggered the upturn in perceptions of Torres, we can find specific references to watching replays.

There are so many ways we can use this sort of data, and we're certainly planning to do more in this area. For example, does a bit of ultimately unnecessary skill boost online reaction more than a simple yet effective pass? Do foreign players get more abuse for diving than their English counterparts? It's all in the data.

Atwal was speaking in London at this week's Big Data World Congress, and added that Squawka is exploring the possibility of widening the scope of its performance analysis scores, with a view to adding services for cricket and formula one.

NEW! Buy our book

• Facts are Sacred: the power of data (on Kindle)

More open data

Data journalism and data visualisations from the Guardian

World government data

Search the world's government data with our gateway

Development and aid data

Search the world's global development data with our gateway

Can you do something with this data?

Flickr Please post your visualisations and mash-ups on our Flickr group
• Contact us at data@guardian.co.uk

Get the A-Z of data
More at the Datastore directory

Follow us on Twitter
Like us on Facebook