Some interesting Twitter stats
Igamania, official accounts shitposting about Big4 and why we should post about Schwartzman and Podoroska a lot more
The summer has been hectic and we’ve barely had time, so we could not get the newsletter out regularly. We took last week off as tennis_gifs in preparation for the North American HC swing but we thought we’d send you some interesting stats we have been working on.
We looked at a few different things some of which are the engagement numbers for different players by accounts.
Analysis
Player Popularity
To assess player popularity I chose the years 2019-2021 for WTA and ATP to include the young players fairly but also because that’s when our account started growing. Players are sorted based on the rankings of 2nd August 2021.
WTA Players
Each player is assigned a colour and the one blue dot indicates the median engagement number on Twitter if they are mentioned in a tweet. (Averages are affected more by outliers like viral tweets). Every other dot is a tweet and the engagements it got for the player. For example, you can see that the WTA tweets a lot more about Hsieh than Podoroska but that Nadia has a higher median of engagements.
For WTA I am showing the figures including all official accounts (WTA, slams, masters tournaments), as well as for WTA and tennis_gifs.
Some of the things we noticed:
If you want engagements, post about Iga Swiatek. Despite only breaking through at the end of 2020 her numbers are incredible. (also interestingly it seems to be just an Iga thing and not a Polish thing as Hurkacz does not generate similar numbers.)
Considering her ranking Podoroska outperforms most people on the list. (Argentinians love their players, wait for Diego in ATP players)
When we combined all accounts Iga is still on top followed by Serena of course.
Despite tennis_gifs being a tiny account, our engagement numbers are quite close to the WTA accounts for these players. (also we need to post more)
ATP Players
I use the same visualization and concept to look at the engagement numbers for ATP accounts. This time I consider tennistv, a lot of official accounts and our account.
What we notice:
What you immediately notice is that the Big 3 are clearly above the whole field which is not entirely surprising.
Thiem is higher than the old guy tier consisting of Wawrinka, Murray, Goffin and Monfils.
For the Next Gen players, it kind of seems like they’re all on the same level but doing better already than LostGen (Goffin, RBA, Monfils).
Hello Argentinians! We see you with those Diego engagements. Only the big 3 have better numbers. (note to self, post more about Diego!)
The Federer effect is clear in our tweets. Nadal and Djokovic are at the same level on our account. We don’t tweet about Berrettini as much so I would take that with a grain of salt.
Murray and Federer barely played in 2019-2021 but you wouldn’t know that based on the way these official accounts tweet.
tennistv engagements are much higher than wta’s. See more below.
Engagement peaks
We also wanted to look at when most official accounts tweet, so we looked at the tweeting frequency per hour. It seems like these accounts know that most of their followers are awake around afternoon European time as that is where most accounts tweet the most, even USO.
I looked at this for several official accounts. The only one that looks different is Australia. (I was very surprised the USO peak was at 4/5 pm).
I looked at how well these tweets do based on weekdays and hours of the day for tennistv and us open by generating performance heatmaps. Both heatmaps make a lot of sense. Darker parts of the heatmap mean less engagements.
USO
the highest peaks are late at night European time, so probably American timezones afternoons and what looks like Europeans waking up and checking tweets.
tennistv
Sunday at 4 pm or Monday at 1 am are the peak. Obviously, those are finals in a European or American timezone. The heatmap also shows that the performance gets better as the week wears on which is also very understandable.
The WTA heatmap looks very similar to tennistv’s. What we notice more on the tour account is how much more engagements SFs, Fs get in comparison to majors where there isn’t such a huge difference between weekdays.
WTA
Total Engagements
To see the growth for some accounts it might make sense to look at total engagements over the years.
Let’s start with our account. You can clearly see that our account has had a huge growth in 2021.
The tennistv account started really growing in 2017. In December they started posting highlights of the whole year. The clay season and US hard court swing is very popular. The WTF usually in November have interesting numbers. You can also clearly see the pandemic dip which most official accounts experienced.
WTA engagements in the last 6 years per month. (The new people they’ve hired are getting a lot of engagements already.)
Final Notes
Ever wonder what the ideal tweet length should be? Well, it looks like the longer the tweet the more engagements it gets for most tennis official accounts.
I scraped the data for all slam accounts, masters events and they post more about the men in these figures than the women. This could be a reflection of the players we chose or a more general trend. More on this next week. It could also be an effect of the amount of times account posts about big 4. Interestingly, USOpen is the only major with a female name in their word cloud.
Thanks to Amelia, Ben Rothenberg, Matthew Willis for their suggestions. Also, the ranking data is from tennis_abstract.
This is only a tiny amount of the data I scraped and looked at. so if you have any suggestions or ideas, dm us or leave a comment.
A quick game of Guess who
I made a word cloud including all 250k+ tweets of someone we all know on TT and unsurprisingly oh, lol, wow, Ana and Kei are featured.
Which official account gets the most engagements despite tweeting the least?
Data and Code
I’ve made all data and code available if anyone’s interested to do more with what I’ve shown you so far.
1. Jose Morgado?
2. Wimbledon?
Maria Sharapova?