Like most of you I follow many baseball writers on Twitter so I can get their views on the latest news and read their new articles. Their aren’t many sites I check every day to see what new content is there – I just wait for a tweet from the author and read it then.
But when viewing these tweets there is always someone in the comments who is upset and you can usually find one, if not multiple, of the following responses in every tweet by a national level MLB writer/content producer.
- You hate my team
- West coast media bias
- East coast media bias
- Anti-New York media bias
- Pro-New York media bias
Most of them are ignored because they are absurd. If you follow Buster Olney you will see that each week he gets the ‘you hate my team’ reply from a fan of every team.
If any of this was actually true then we should be able to analyse these tweets and find out, so let’s do that. Twitter has an API which allows you to get the last 3.2k tweets someone has made, so let’s start with Buster and see what we get. This does include retweets and replies.
When I ran the data grab for Buster I got tweets from 9th June to 11th November. I then searched the tweets to see if they contained a team name or city in them – for cities with more than one team I used team names only e.g. “Yankees” and “Mets” and not New York – but “Houston” or “Astros” for 1 team cities.
This gave me a list of tweets which Buster had made about each team, which gives us some detail into how much he thinks about each team but the real question is what is he saying in those tweets.
For that I used a basic form of sentiment analysis which analysed the text of each tweet and classified the tweets as either negative, positive or neutral. Now this isn’t perfect, a tweet with negative language doesn’t necessarily mean the tweet is negative about the team but that the content is negative. It is the same for positive language.
E.g. A tweet saying ‘Team fired/hired manager X’ would be categorised as a negative/positive tweet for that team.
For Buster that gave me the following.
We can pick out a couple of interesting things here. The two teams which Buster talked about the most were the Yankees (150 tweets) and the Mets (185 tweets), whilst the teams with the lowest number of tweets were the Royals and Rockies with just 5 tweets.
You can see that most tweets don’t get given a sentiment so let’s look at the average sentiment for each team. Sentiment ranges from -1 (fully negative) to 1 (fully positive).
So on average, Buster tweeted positively about 27 of 30 teams with just the Rangers, Brewers and Pirates averaging negative sentiment. The top and bottom teams are ones which he tweeted about the least so they will be skewed by an individual tweet more than the other teams. To analyse this properly we should really look at teams with larger volumes of tweets and compare them.
If we do this we can clearly see that the Houston Astros have a much lower average sentiment that the other most tweeted about teams. That is probably to be expected following the Brandon Taubman controversy involving their now fired Assistant General Manager and the quagmire of incompetence that came from the Astros during the whole incident.
I wanted to see if there were any trends across the MLB writers/content producers as a whole so I repeated the exercise above but for 60 people and combined their data together. These were all national level coverage and not someone who is specific to one team/area. That got me 35k tweets to analyse.
It’s not that surprising that the Yankees come out as the most tweeted about team. The White Sox are the least, they may be hurt by not having Chicago as a search term but the Cubs have been tweeted about way more often. So I don’t think the impact is large.
If we look at the number of Twitter followers each team has, a good proxy for popularity, you can see a small trend that the teams who have more followers are tweeted about more.
There are a few outliers though. The time frame for these tweets includes the playoffs which means that some teams are going to get more coverage. The four teams who made it to the Championship Series have been tweeted about much more than expected. If we ignore those four teams for that reason we are left with two teams as outliers: the Mets and the Blue Jays. With the Mets getting way more coverage and the Blue Jays way less than expected.
One might be inclined to suggest the managerial change may have increased the volumes for the Mets but we don’t see that for other teams who have changed manager recently. It may be the writers trying to counter balance talking about the Yankees so much but they don’t do the same for the White Sox or Angels.
Let’s look at the overall sentiment.
Now this graph is a lot more surprising, the Tigers being top and the Red Sox being bottom isn’t what I was expecting. As before, the teams with fewer tweets are generally at the top or bottom but the Red Sox have quite the quantity of tweets and are quite significantly different than the other teams.
If we look at a more granular level, the Red Sox are hurt by having no one writer tweet about them very positively. As well as a the 8th lowest sentiment from any writer to any team from Bob Nightengale.
Speaking of worst sentiment here are the bottom five scores and the top five scores, for people with more than 20 tweets about a team.
What can we conclude from all of this? Not much really, but in general, no, writers don’t hate your team. They might not talk about them but they don’t hate them. The other takeaway? I think there is room for more content about the Blue Jays as there should be the audience for it.