Ashes Squad Selection: Sentiment Analysis

Gaurav Kumar
4 min readNov 17, 2017

The 13 man squad for first two test matches of Ashes 2017–18 was confirmed by CA early Friday morning.

Paine’s selection was one of the greatest surprises who was given preference over his out of form Tasmania teammate, Matthew Wade, and the NSW’s Peter Nevill as glovesman of the team. Paine scored his lone first-class hundred in 2006 (Aussie coach Darren Lehmann scored a more recent one in 2007) and wasn’t even in Tasmania’s best XI to start the Sheffield Shield season.

However, chairman of selection committee Trevor Hohns defended the selection saying “Tim was identified as an international player a long time ago and has always been renowned as a very good gloveman, also performing well for us whenever he has represented Australia in any format.”

Apart from Paine, Shaun Marsh was selected ahead of Maxwell, and Hilton Cartwright for №6 position. Consistent Cameron Bancroft earned his place to open alongside David Warner scoring a double-century earlier this week and is all set to debut in Gabba test. Mat Renshaw was left out of the squad due to his recent form. However, Steve Smith backed Renshaw to make comeback in recent future.

The squad

“Steve Smith (captain), David Warner (vice-captain), Cameron Bancroft, Usman Khawaja, Peter Handscomb, Shaun Marsh, Tim Paine (wicket-keeper), Mitchell Starc, Pat Cummins, Nathan Lyon, Josh Hazlewood, Jackson Bird and Chadd Sayers”

However, the selection of Marsh and Paine did not go too well with everyone. Here is how Twitter reacted to the selection.

However, Adam Gilchrist feels that it was a tough call for some players but the Australian team is balanced and ready to take on England.

So, I did sentiment analysis of the public tweets to get a general perception about the selcted squad for the first two tests.

Setting up the authorization

Further, I scrapped the tweets containing the keywords #Ashes , squad and save it in a R data-frame.

Preprocessing of the tweet text is required to do further analysis. In preprocessing I removed RT, any Punctutations, Alphanumeric characters and hyperlinks.

Above chunk of Code made a corpus out of the tweet text and makes a word cloud for the most used words in the tweets. Words like Ashes, Australia, Test, England were removed from the corpus as they were frequently occurring but had not much relevance for the sentiment analysis.

Wordcloud of the Tweet

Clearly, the most controversial selection of Tim Paine was most talked about followed by Shaun Marsh’s selection. However, some tweets were also about opener Cameron Bancroft which is all set to earn his baggy green.

Further, Sentiment analysis of the tweets shown in the figure below suggested that not many people were impressed by the selection of the squad. The sentiment of most of the tweets was negative and sad. However, some people were positive about the selection including legendary Adam Gilchrist.

So, let the Ashes begin and see how the selection impact the side.

--

--