Is there a “Bots for Trump” Takeover?
Is Donald Trump mustering an army of Twitter Bots to take over the social media landscape? Probably not.
It was May 30, 2017. The bot-pocalypse. The day the Twitterverse stood still.
B-Day. Well, that sounds a little over the top, doesn’t it?
So what really happened?
An unverified and false claim—trustingly perpetuated even by people like Malcom Nance—was made on Twitter that Donald Trump had gained millions of followers in just a few days. It made for pretty compelling conversation: users were literally refreshing Trump’s Twitter profile page, watching his follower count increase and reporting on the number.
But nobody was watching Justin Bieber!
We’ll come back to him shortly.
What were they doing?
You see, someone was bulk-creating a bunch of accounts on Twitter. They probably weren’t made for Donald Trump—in fact, it’s quite unlikely, though not out of the realm of possibility.
Well-schooled Twitter spammers and evil ne’er-do-wells won’t bulk-create accounts as rapidly—and obviously — as someone (or some group) did today; at one point, I was watching 100+ new accounts per minute being created! That’s one red flag that it’s not a Trump digital operation.
Another tip-off? They all followed the default group of Twitter users that would normally be added automatically if you didn’t choose any “interests” when you signed up. The specific accounts varied slightly—based on the fake location of the fake accounts. But they were all from some region’s top 20–25 Twitter users.
Twitter’s default algorithm at work: amplify the already-loud.
Someone using a bunch of new accounts for secret purposes isn’t going to shout, “HEY EVERYONE, WATCH ME CREATE A BUNCH OF NEW ACCOUNTS FOR SECRET PURPOSES, ALL THE LIVE LONG DAY!”
Who would? Spammers would. They DGAF, if you KWIM.
“IDGAF” — Spammers
Why do they do this?
Spammers gonna spam. They need lots and lots of accounts that run automated scripts to like/follow/retweet popular users or trending hashtags, in hopes that they’ll catch your eye and you’ll check out their profile. You’ve seen them, right? The profiles that have a scantily-clad young woman as their pinned tweet? Sex sells. And porn sites kick back enough to spammers to make it worth their while. Or illegal prescriptions. Or some other affiliate link that kicks back to the spammer.
There’s a more nefarious type of fake-account army…One that likes/follows/retweets to get your attention, and then offers a tantalizing link in their pinned tweet. You click it, and if your computer isn’t protected, you might be tricked into installing malware—literally “bad” software. This malware might be one of those bitcoin ransom schemes that encrypt all your data, or it might cause your computer to run real slow while it mines for Bitcoin (and other “cryptocurrency”) for the spammer’s benefit.
Ever seen those sites where you can buy followers? Pretty much every major celebrity uses them. Some of the more sleazy operations are selling these newbie accounts. Sure, you get followers, but none of them are humans.
There are higher-end sellers of social media followers, whose accounts are run by humans, which puts them firmly in the realm of “totally allowed by the Twitter Terms of Service”. These accounts made today? They’re not for the discriminating Social Media Marketing service.
My money’s on spam.
Okay, back to The Bot Rising
So, they all followed similar bunches of popular users, which caused the follower accounts of those users to increase by as much as 100 users per minute. Trump wasn’t even in all of the auto-follow lists.
But Justin Bieber was.
Let’s have a look at Justin Bieber’s follower growth during the merry, merry month of May.
Here, the green shaded area shows the total number of followers — note that the graph’s Y-axis doesn’t start at zero—and the red bars indicate the number of new followers each day. I’ve also added a light green trend line for the number of new followers each day, using a moving average of the past two days.
May 24 and May 26 were big days for the Beebs! You’d better Belieb it.
Okay, I’ll stop.
I wonder how Hillary was doing during this same period. She’s often one of those pre-configured “people to follow” for new accounts. Not always—she leads the trends in fewer regions overall—but often enough.
Looks similar. How about Donald Trump?
You have to scroll a lot to see these growth trends side-by-side. Because I’m such a nice guy, here’s a simple line graph with Trump, Clinton and some Canadian pop singer, showing the number of new followers they each amassed during each day of May:
Some folks stopped Beliebing in early May, and Clinton took a bath around May 17th, but mostly these lines trend together.
In fact, not only did Donald Trump not amass a significant number of followers on May 30th, he had bigger growth days earlier in the month, outpacing even Justin Bieber for most of May.
So why didn’t anyone notice?
Somebody did. I did. I monitor Trump and other political figures each and every day, and I’ve watched his follower account go up and down. Since the inauguration, he’s had some interesting days:
Aside from a glitch where there was no data (not Twitter’s fault) in mid-May, Donald Trump had several bursts of new followers. In fact, there were many days where he beat the “Botpocalypse 2017” growth handily. Around March 21, he added over a quarter of a million new accounts. And who can forget “The Great Force Follow” after the inauguration? From January 21–23, 2017, Trump gained over a million new followers.
That’s not all. Let’s see how he did between the election and inauguration.
Dang, easy D. Over 800,000 new followers on November 10th. Over 7 million in new followers from Election Day to Inauguration Day. That’s an average of over 100,000 new followers per day.
But on The Day the Bots Stood Still: He earned less than that.
Finally, for shiggles, here’s the whole enchilada. Donald Trump’s rise to Twitter (fame?) since June 6, 2015. Oh, and in case you’re wondering what the bump was on October 14, 2015?
The first Democratic debate was held that night.
Oh, and this happened.
Ouch.
On with the graph.
Thanks for hanging in there. That’s all I have for you tonight. Until the next analysis of some crazy happenstance on Twitter!
