The motivation behind this project is to capture user activity on Twitter during a major television broadcast and observe how well the program is resonating with their audience. The analysis can identify popular topics and hashtags, locate key users, and provide additional insights on how content is being shared. I wanted to apply this to AEW Dynamite which airs on TNT every Wednesday at 7pm CST for its 2 hour broadcast of wrestling matches and promos. The latest episode, on Wednesday, March 18th, 2020, was especially interesting for several reasons.
For an interactive demo, check out this Twitter Activity Dashboard. It contains a subset of the total dataset (9,466 tweets) covering the last ten minutes of the show and five minutes after it ends (8:50pm-9:05pm CST).
This is an on-going project and updates will be made regularly. Feedback and comments are always welcomed.
Find me on Twitter @vizwrestling
or send me an email at visualizingtheelite at gmail dot com.
The tweets were collected on Mar 18, 2020 from 05:59PM CST to 10:04PM CST, for approximately 4 hours and 5 mins. This line graph shows the number of tweets over the span of time. The vertical black dotted lines are reference lines for when the show started 07:00PM CST (or 19:00) and when it concluded 09:00PM CST (or 21:00).
| Summary Statistics | |
|---|---|
| Total number of tweets | 77474 |
| Number of unique users | 19282 |
| Average number of Tweets Per User | 4.02 |
| Average Number of Tweets Per Minute | 317.52 |
This map shows the locations of the tweets based on the Twitter user's location
as declared on their profiles, if provided. It only includes valid states in the United States.
For example, “Parts Unknown” is not a valid US location.
From the map, we can see that largest number of the tweets
originated from users in New York (3,633), Texas (2,591), California (2,424),
Florida (1,941), Pennslyvania (1,658), and Illinois (1,638), respectively.
The users that are tweeting about AEW are mainly identified as being
from the eastern United States with the exception of California.
This pie chart shows the breakdown of the different types of tweets.
Retweets (RT) are tweets that repost or forward a message posted by another user.
Responses are tweets that are directed to another user.
Mentions are tweets that mention another user but is not a retweet or a response.
All other tweets are grouped under text tweets.
More than half of the tweets in the dataset were retweets,
implying that there is a lot of content sharing and broadcasting.
Let's look at who and what they are retweeting.
This table shows the 10 most retweeted users in the data set
along with its frequency.
The official AEW accounts @AEWonTNT and
@AEWrestling are the top
two most retweeted accounts in the dataset. This may suggest that they are quite
effective at reaching their audience with proactive content that encourages them
to engage and spread (i.e. retweet). tde_wrestling,
a popular account that creates and shares wrestling GIFs, was a close third.
| Username | Frequency |
|---|---|
| @AEWonTNT | 9943 |
| @AEWrestling | 2991 |
| @tde_wrestling | 2741 |
| @JDfromNY206 | 967 |
| @CodyRhodes | 924 |
| @PhoenixAEW | 829 |
| @TonyKhan | 631 |
| @smarktodeath | 584 |
| @EIiteAEW | 547 |
| @nodqdotcom | 496 |
This table shows the 10 retweeted tweets found in the data set
along with its frequency.
What was the top three tweets?
| Tweet ID | Frequency |
|---|---|
| @1240458505867288577 | 2008 |
| @1240446453991923712 | 1427 |
| @1240429357782294528 | 1078 |
| @1240441999473422336 | 631 |
| @1240451016698388480 | 616 |
| @1240459668058750978 | 590 |
| @1240448408264949760 | 567 |
| @1240430552911695877 | 546 |
| @1240458434799173632 | 543 |
| @1240441616491581443 | 513 |
More to come on the detailed process in putting together this project, but here's a summary
of the technologies used.
© 2020 Visualizing the Elite