Saturday, August 10, 2013

Tony Abbott is paying people to like him.

Not too long ago Tony Abbott's staff uploaded this image to their Facebook page - 

100,000 likes is very impressive!  It's a pity that they are clearly being made by bots.  

Some enterprising souls noticed that Mr Abbott's likes had been growing very quickly of late.  Growing out of line with any expectation.  

This is what Mr Abbott's increase in likes looks like -

For comparison this is Kevin Rudd's -

And the Liberal Party of Australia's -
Edit - Added the Liberal Party page for another comparison.

That alone looks suspicious.  Notice the sharp increase in the grey line - new likes - and the almost direct correlation between new likes and "people talking about this".  

Something is off.  

Luckily, as I said, some enterprising souls noticed this yesterday, and have been tracking the growth in likes ever since -

22 hours of data collected, once per minute. Sampling likes and "mentions" (mentions are cached and only update once every 24 hours).
Raw data is here:
This is what Abbott's graph looks like:

This is Rudd's:

There are four distinct periods: 
"Normal": 8am-11pm EST
"Rampdown": 11pm-12:15pm EST
"Low": 12:15pm-6:45am EST
"Rampup": 6:45am-8am EST

During these periods, the differences were pretty stark:

First Normal
Rudd: Mean 1, Stdev 1.18
Abbott: Mean 18.22, Stdev 4.31

Rudd: Mean 1.19, Stdev 1.07
Abbott: Mean 12.32, Stdev 4.43

Rudd: Mean 0.22, Stdev 0.53
Abbott: Mean 2.89, Stdev 2.38

Rudd: Mean 0.39, Stdev 0.69
Abbott: Mean 10.15, Stdev 3.06

Second Normal
Rudd: Mean 0.6, Stdev 0.82
Abbott: Mean 18.47, Stdev 4.96

Some notes:
Not only are the total numbers of likes way too high to account merely for likebait saturation, Abbott's variation is far too low for to be entirely human-based. There are also no major spikes that we'd tend to see during periods of policy releases (for instance, his release of indigenous policy this morning). The means of each time period tend to line up too perfectly and again, lack in variation. Overnight, the deviation returns to what we'd tend to expect from this kind of data - indicating that the bot/net are probably turned off overnight after they ramp down/up (so as to not see an immediate jump from a mean of ~16-20 straight to 2-3). In short, this is exactly how I'd code a bot to be difficult to detect (if I were actually ridiculous enough to do so).
There's some other notes: the number of mentions cached line up almost exactly with the number of likes added in the same time period - 6727 likes from collection start to cache refresh, 8010 new mentions since previous cache refresh. There are several hours unaccounted after the previous cache refresh, which would likely make up the missing number there. Note that in a similar time period, Rudd gained approximately 350 likes and 0 new mentions. As people tend not to reference people they don't know on facebook except through page likes (not mentions), you can pretty safely assume that each bogus account is also mentioning the page 0-1 times.
Since collection, Rudd has gained 582 likes and 0 mentions, and Abbott has gained 15867 likes and 8010 mentions. It's worth noting that Abbott averages to almost exactly 1000 likes per hour yesterday, and 1200 today. Rudd's is all over the place, by comparison - 20-60 yesterday to 35-50 today.
tl;dr it's a bot, but written exactly how I would do so if I were to write a bot to spam likes. I'd make a couple of modifications - randomised ramping start/end times (within a tolerance of 2 hours), and much greater variation in the number of likes per minute.
Another kind soul decided to compare the increase in likes to people outside of the Australian political spectrum, and found that Tony Abbott is gaining popularity at a greater rate than One Direction, Justin Bieber, and even Facebook itself.  He's around 50% more popular than Game of Thrones, even.

 It's just too perfect an increase in likeability, particularly when combined with the people talking about it, and the curve over time.

What it adds up to is somebody paying a bot to mass-like Mr Abbott's page.  Alas for the image at the top of this post, and for Mr Abbott's ego, the likes are illusory.

And against Facebook's terms of service.

Update August 11 (Morning):
Tony A-bot-t update: 
Rampdown / Rampup occurred at exactly the same time today, and at extremely similar rates. Normal operation has returned to very similar means as yesterday.  
For comparison:
First rampdown: Mean 12.32, Stdev 4.43Last night's: Mean 13.21, Stdev 3.6
First rampup: Mean 10.15, Stdev 3.06This morning's: Mean 10.26, Stdev 3.41 
And now we're back to normal operation of ~16-20 likes per minute.
From the last two days worth of data (sans most normal operation)(from normal -> rampdown -> low -> rampup -> normal)1094, 833, 510, 250, 128, 88, 85, 142, 356, 681, 1092 1123, 847, 595, 260, 168, 90, 112, 131, 401, 724, 1011

Update August 11 (Evening):
The graph for the two days of activity that were monitored.  Almost perfectly mirroring.  

Edit - 10th of August: Clarity.
Edit - 11th of August: Added the Liberal Party of Australia Likes graph. Updated with last night's data.  


  1. every time I open facebook I see an ad asking me to like Abbott

    I have been banned and blocked from his page for several months now but they still pay to advertise to me :-)

    1. Yeah, I mentioned this odd behaviour on his page, and I've been banned from posting comments now.

    2. You are full of it. I am on the page all the time and would like to point out we ignore you and other Labor stooges some of your comments become personally insulting to individuals on the site, I suggest that perhaps might be more the reason if you were banned. Why would you go on a site if you aren't a fan. I wouldn't bother going on the Labor site and have no interest in it.

    3. I'm glad you are a fan of my blog Supa. :)

    4. If you take a look at the "people talking about this" you will see that it's a significant number. Bots don't talk, or like posts, or comment, and his page is very active with likes and comments. Coincidentally, this week is not his most popular week, so your "facts" might need checking. You might also want to take a peek at this

    5. Zoe, Facebook counts a 'like' as one of the interactions which contributes to 'talking about this'
      So you can take all the paid new likes off the count to get to the real number.

    6. Zoe quoting from Menzies House is a long way from unbiased and/ balanced opinion. A true believer?

  2. ...I am curious if the system to measure 'mentions' can differentiate between mentions of positive and/or negative sentiments ?(the latter may well explain the spike)

    1. Such tools exist, but you tend to have to pay for them.

  3. The mentions will not differentiate, this is the line that will naturally spike when something newsworthy is happening - such as an election announcement, or in the case of Rudd, when he regained the leadership. The problem lies in the number of "likes" being generated in a very short amount of time. See that instant "jump"? You have a natural rise in Kevin Rudds page and a big jump in mentions (because of the election presumably) however, Abbott's has a massive jump in likes, where do tens of thousands of people suddenly "like" a politician? But you are right, the big jump is both positive and negative comments for Rudd and Abbott.

    1. Yeah - the mentions are positive and negative mentions.

      The issue is that mentions don't correlate with likes anywhere else. What's happening on Abbott's page is distinctly aberrant.

  4. Cool. You seem to have done an impressive work up there. However,I want to point out that in order to determine whether the likes of a FB are genuine or fake you have to assess three categories: First, if the most popular city is a foreign city, then the likes are most likely to be bought. In this case we have Sydney as a the most popular city which is pretty normal in this context. Second, if the main age group does not match the target demographic, then the likes might not be real. However, in this case we find that the main age group is 25-34 and those are more likely to be concerned with politics than others as they're legal voters and, also, new to the job market, and suffer from unemployment problems more than any other age group. Third, the most popular week started when Kevin Rudd called election which is again PRETTY NORMAL. Most people would start making research to decide who would they vote for after the election is being called and that's why the likes increased significantly since that day. Moreover, it is VERY COMMON in the pages where likes are being bought that the gap between likes and 'people talking about this' is very huge as the bought likes wouldn't bother to comment or participate in the page. In this case we have 71,000 people 'talking about this' which is a pretty impressive number. BOTTOM LINE, the likes of this page are GENUINE. Cheers. Jared

    1. A) VPNs exist.
      B) Fake accounts exist.
      C) Ramp up started before the election was called. And even if we accept that it's an event that precipitated an influx of likes you'd expect to see other upticks for the Greens and/or Rudd.

      Bonus round: 18 likes per minute, like clockwork.

    2. Alrighty, I see your point. It's a good one. I know fake accounts exist. However, look at the engagement in Abbott's page. It shows the genuinity of the likes. shares, likes, and comments of every post are just about significant.

    3. I also noticed the jump in likes today, but I also run a page about Kevin Rudd which got several hundred likes in a couple of days, because I posted a pic that everyone liked and shared. I think you're jumping to the wrong conclusion.

    4. There is a significant difference between your page receiving several hundred likes within a couple of days and another page receiving tens of thousands of likes in the same time period, while related pages receive no such increase.

    5. There is a huge difference if my page only had 500 likes to start with, and the other already had close to 100,000. You think the other page would be reaching a lot more people?

      So my little page reaching 200 more people is quite ok. His big page reaching 10000 more is a scandal obviously.

    6. This comment has been removed by the author.

    7. Going by the likes per week shown on the page one week ago he was at --- likes, (likes per week listed on his page for the last week are: 42333). So it's not like your page going from 500 to 700 over a week, it's like your page going 500 to 1000 in a week.

      His page over the last week about the same amount of new likes as it had likes before. it's absolutely ridiculous.

      Also, that 10000 he has listed there on his table is over 15 and a half hours, not a week, and Facebook which has 90 million likes only got 7000 over the same time period.

      *-the deletion is because I misread how the like history works, I thought it was showing likes in week long blocks, not likes over a week day by day, sorry-*

  5. As Doug mentioned above, they are advertising on Facebook - that's why likes are going up - no bots are required. It's quite common to see large growth in likes when you are running a targeted Facebook campaign. The number of my friends who like his page have tripled in the last week and they are definitely all real.

    1. I'm not saying that advertising won't see an uptick - they've certainly poured a lot of money into their Facebook presence.

      What doesn't add up is the smooth regularity of the 'Like' behaviour. They were chugging along at almost exactly 18 likes per minute last night. Without fail.

    2. I've just had a look at the raw data file and there seems to be quite a bit of variation minute to minute. Certainly not 18 likes every minute - although no doubt that's the average. Again it's not surprising that there would be some consistency if they are advertising - they would be buying a specific number of ad impressions every minute and you would expect a similar percentage of people to respond.

    3. While the standard deviation is still uncharacteristically low when comparing minute-by-minute data (which would already indicate a problem), comparing the hourly results indicates a deviation of less than 10%, even during what we'd typically call peak periods - policy announcements and peak hour usage. The lack of variation, lack of spikes during peak, consistent voting over a period of days leading to an altogether ridiculous rise in followers (and altogether related increase in mentions) indicates that somebody is gaming the system. In addition, CPM is not broken down by minute - ads are shown where relevant over the course of a whole day, not set for a certain number of displays per minute. We'd also expect to see dropoffs when Ad funding for the day is depleted (unless the LNP has allocated infinite funds towards page advertising, which I find extremely difficult to believe).

      This, plus the tendency of political parties to buy social media penetration such as this in the past, would definitely lend credence to the theory.

  6. Get your hand off it, you wanker. If anybody is the fudger of figures, it's you! Time to admit that Rudd is a dud.

    1. I can assure you if I had fudge the only thing getting fudged would be my mouth.

      (I would eat the fudge)

    2. Fudge is delicious.

      Jim you assume Llew supports Rudd by default. He's not said that. He's just pointed out the discrepancies in the massive upswing of likes versus the more natural likes of other political based pages.

  7. I think you're forgetting that you'd struggle to find 100 000 people willing to actually advertise to their friends that they like him....

  8. Paying to be liked on Facebook?

    Only Labor people would do that - wasting money is in your guys's DNA.

    1. When the Coalition came to power in Western Australia, government debt was $3.6 billion. Forward estimates now have it at $28 billion.

      Would you like me to go through the story of the other LNP State governments as well?

      It's probably not a very good idea to bring who is and isn't wasting money, because you'll be in a world of butt hurt if you try to debate this with me.

    2. "which party manage economy better?" - coalition 58%, Labor 38% (News poll).
      Poll after poll the answers are the same.

      Where is your surplus? Ops, I better not go on to bore other bloggers here.

    3. That's really funny. Asking a Coalition newspaper who manages the economy better and still 38% of them say Labor.

    4. and its not about what people think, 100% of people could have said katter would be the best manager, doesn't make it true, go look up real facts and figures and you'll find howard was the most wasteful leader we've ever seen and since that point labor has been reeling in the ridiculous spending of the libs.... and before you argue with me, check the REAL FACTS, not shit vomitted up by the MSM.... I think you'll be surprised...

    5. dXie

      The Liberal Party have simply been lucky. Lucky to win in 2001 when gone in the polls due to September 11 and Tampa (Although the crucial Children Overboard lies weren't so much luck as their characteristic manipulative sneakiness.)

      The calm economic waters after that and the revenue lottery win on Mining Boom Mk I, plus flogging assets, made their job look good, but they still they produced a structural deficit by underfunding infrastructure, training etc by blowing the money on electoral bribes. And waste, such as the staggering $1.2 billion for 11 helicopters that never worked properly and were given back.

      If the GFC had been in 2006, they would have had deficit and debt blown out. Again they were simply lucky they avoided it.

      As Lev says, the Liberal Party in WA has generated a debt blowout despite being the strongest state economy. They even admit they have a debt problem. Other Liberal states have debt and deficits and are constantly putting back their plans for a balanced budget.

      The same as the Federal govt, but blinded Liberal supporters can't see the obvious. There is no magic wand and the aggressive razor gang massacre Abbott plans might jolly the budget, but at cost to the economy, as 'austerity' is being shown to fail elsewhere.

      Abbott is a danger. Hockey has no magic wand economically (and is frequently a buffoon), but will just whine about how everything is Labor's fault for as long as people will believe him.

  9. Again, I appreciate your work. Please consider my criticism to be a constructive one. I really think you should consider the context. We're in an election time. Even the ramp up started before the election is being called, it didn't start long time before that. Everyone knows that the election is coming in September. Look at Obama's likes during the 2008 and 2012 election years; they increased at a faster pace than Abbott's. Of course, America has more population than Australia, however, what concerns us is the average. Another thing is that the likelihood of Abbott becoming a PM is what mainly boosts his FB likes; indeed, lots of the likes come from non-liberal supporters or independents who want to get updates from the upcoming PM. Bottom line, there's no evidence in your data shows that Abbott is buying likes, because in all likes-buying processes, most of the likes aren't likely to participate. In this case we have a significant daily participation from a huge number of profiles. You can analyse them and find out whether they're phoney or genuine.


    2. if you're going to talk likelihood of abbott winning the election, well it was a lot more likely a couple of months ago, why wasn't there a surge at anytime during last 3 years?

      also you can say analyse the profiles but who has the time to do that? I don't doubt the page is quite active but the jump is ridiculous and can only be seen as a response to the claims they're not in touch, especially after this weeks Q and A, I'm sure many centre & left voters would disagree that social media will have no influence over the outcome of this election. the lies told by the MSM in the states forced people online for more accurate news and despite obama's negatives the repulsiveness of the rights ideals were exposed. murdoch's empire is attempting the same here and this is a further attempt at encouraging a heard mentality......

  10. There is some interesting stats-based criticism here, but I enjoy that the empty criticism presumes that you are a Rudd supporter. Typical black-and-white viewpoint of the true believer.

    1. If you're not {A} you must be {B}, after all.

  11. 2 possibilities.
    1. The libs have set up a scheme to give the impression that Abbott's likes are increasing using a program to deliver likes on the site at predetermined times linked to appearances and announcements.
    2. Some one is gaming twitter like various online opinion poles have been gamed by hackers in the past.

  12. What about Rudds twitter account? Surely you analyzed jumps in support there??

    Jumping by hundreds in minutes! We await your analysis.

    1. I had a quick look at this. Rudd's follower count has grown 4.56% in the last 7 days (approximately since the campaign started), while Abbott's has grown 6.5% in the same time period. You can also see both the Twitter and Facebook account comparison here:,216342268645&twitter=KRuddMP,TonyAbbottMHR

      As you can see, The 171.16% rise in Abbott's followers on Facebook is uncharacteristic, while the trends in both accounts' Twitter followers is about what we'd expect from increased interest during an election cycle (about 5%).

      Honestly, it would be somewhat pointless to look at both lists of twitter followers and the increases there - fake Twitter followers are extremely prolific, as you can see by these links:

      It's interesting to note that even though Rudd has a much higher follower count, a higher percentage of his followers are also real (which is a surprise).

      Either way, the environment of Twitter v Facebook differs significantly and Twitter is typically rife with bots, so we can pretty much assume that everybody is being dodgy there. Facebook is a much less common medium for this, though, which is why it's noteworthy.

  13. That is really really sad. Also sad are people instantly ignoring this and attacking Rudd because of blind political faith to a party. Like him or not this is still really concerning.

  14. Two days ago I saw an ad for Tony Abbott on my FB page and at that time he had around 78000 likes and now it is almost 120000. So in two days he has somehow gained more than 40000 new likes by running an ad campaign? Some part of me definitely thinks no, because that's a lot of likes to achieve from ads that you generally only see on the browser and that most people tend to ignore/overlook. Add to that the fact that a lot of people would probably be more likely to like the liberals page than that of Tony Abbott's and it definitely looks suspicious to me.

  15. I currently run a somewhat popular facebook page for my business (1500 likes). My analytics tell me that in the week that I get a like, I get an additional "person talking about this". This is because it appears on their friends newsfeeds that they have liked my page, as if they are talking about the page.

  16. Are you suggesting he's paying people to like him or *for* people to like him? Paying for ads is very different to paying a company that runs fake users (bots) to like him and increase the numbers. You seem to mention both and prove neither.

    And if you change words 'ramp-up' and 'ramp-down' to 'wake-up' and 'go-to-bed' things might be a little less confusing/misleading.

  17. All you've really proven is that people are more likely to have "liked" Tony Abbott's page when they were awake than when they were asleep. Not particularly surprising. You only have to look at the engagement rate on Abbott's page compared to Rudd's page to see that the likes are real.

  18. These are obviously fake likes and to the party afficianados writing comments, you like your political candies are living in a world of delusion. Spin it, write long winded comments all you like but anyone on any social media platforms can see that sort of 'like hike' doesnt happen overnight. Its a bunch of kids in India working for peanuts. Whats more unbelievable is a man with the charisma of a boiled peanut gaining such popularity in such a short space of time.

  19. Nice analysis. Pity it takes so long to explain basic statistics to motivated reasoners.

  20. Hah! Nice stats work, especially the comparisons. Though to be fair, he isn't paying *people* to like him...he's paying a person to generate a lot of likes on his page.

  21. There is a lot of hype about buying facebook fans where everyone seems to be so interested in buying the fans. Although, the fans are paramount in your business: helping you to build a brand and increasing your sales, you should not buy any facebok fans. To be on the safe side you should buy real facebook fans. If you want to buy facebook fans, you should note that there are two types of fans that you can buy: real and unreal facebook fans. Real fans are those obtained from people’s real accounts. On the other hand unreal fans are usually bot generated. Here the fans are computer generated and the accounts of these fans are usually inexistent. Although, unreal facebook fans will add more fans in your account, these fans are usually of no value to your business. This is because they are from unreal people thus the fans can’t buy or make your product popular. To ensure that you buy facebook fans that will be of value to your business, you should ensure that you buy them from reputable sites such as . In addition to this, you should avoid buying facebook fans selling at extremely low prices.


  22. Facebook Likes are a powerful way to get website visitors to socially engage with your brand.
    There are two types of likes you can solicit with Facebook’s developer code – likes for website pages and l
    ikes for your fan page.
    The best part is users can like both of these directly on your website without having to go anywhere else!

    social media services
    buy facebook likes
    buy facebook fans

  23. Many Facebook fans are good as they make you famous. The fans are also excellent people to market your products to; therefore, by having many facebook fans you will have many people to market your products to thus you have very high chances of increasing your profits. Although, facebook fans have the following benefits, you should not buy just any fans; you should only buy real facebook fans. These are fans that can be able to see, comment, and even share your posts with other people. To buy real facebook fans is easy. This is because all you need to do is to identify a reputable site and place your order. Once you have placed your order, it will take you just a few days for the order to be delivered.

  24. Thank you for sharing this information with us. I hope this information will help peoples,. Please take a look my blog.
    I am social media marketing expert. I am providing facebook likes, twitter followers, instagram followers and much more...
    If you want to buy Facebook Likes, Twitter Followers, Instagram Followers. Please Take a look.
    Best Social Service
    Best Social Service