Archive for the ‘prediction’ tag
Update: Twitter was right. It’s a Romney-Ryan ticket.
It seems very likely that Mitt Romney is going to select Paul Ryan as his running mate for the Republican Presidential nomination. And it’s looking like Twitter predicted this. We hinted that we’d been tracking Republican VP Candidates with the screenshot accompanying the announcement of our new dashboard earlier this week. As you can clearly see from the updated dashboard below, Ryan started to pull away from the potential VP pack three days ago in terms of unique reach on Twitter. Of the pool of likely candidates, Ryan’s seen the greatest increase in reach over the past month, gaining a 65% increase in reach in the past 30 days. In addition, he’s seen the largest gains in both the number of total tweets and unique people talking about him recently.
So, did Twitter predict Romney’s decision correctly? Well, we’ll know soon enough, as Romney is expected to officially announce his vice presidential running mate tomorrow. We find Twitter’s potential to predict (or not) cultural and current events very interesting, so we’ll be following along and will post a more in-depth analysis next week, so more very soon.
When the 2012 United States Republican presidential primaries and caucuses began back in January, we took a look at whether Twitter activity could be used as a predictor of the elections. We started tracking all tweets that mentioned any of a candidate’s Twitter accounts (personal and campaign) and based on the Twitter activity coming out of the Iowa caucuses, we saw that Twitter activity was less an indicator of the outcome, and more a reflection of the overall conversation happening around the candidate. Reach, exposure, and activity were largely driven by mentions by popular news and media accounts, many of which have significant numbers of followers and retweets.
Since January, the Twitter activity on the candidates has been staggering – some of the largest reach and exposure we’ve ever tracked, with over 8 million tweets from hundreds of thousands of contributors. These contributors reached more than 120 million unique Twitter accounts and generated almost 22 billion impressions.
Right before the Super Tuesday primaries in March, we launched the TweetReach Republican Primary Tracker which looked at the relationship between what people say on Twitter and what they do at the polls. In the visualization, we mapped the number of unique Twitter users talking about a candidate to the y-axis, polling results to the x-axis, and tweet volume to the circle radius.
While, based on our previous analysis, we did not believe Twitter conversations could predict winners, we thought it would be interesting to see what tweets can tell us about how potential voters feel about the candidates. The visualizer confirmed that despite a candidate’s tweet volume, reach, and exposure on Twitter, these data were not a good predictor of election results. They are, however, a great way to understand how popular dialogue about a candidate translates into Twitter conversation.
Today, with Rick Santorum bowing out of the race, we took another look and found that Twitter conversation about Santorum had been relatively quiet since Super Tuesday but, as expected, spiked with today’s news as people came out of the woodwork to Tweet about the candidate.
In fact, a full 21% of Rick Santorum’s exposure since Super Tuesday (over 368 million impressions) occurred today after the announcement. When viewed with the TweetReach Republican Primary Tracker, the impact of the conversation around Santorum’s departure is even more pronounced.
We look forward to tracking the upcoming full election. In the meantime, we’d love to know what you think!
Here in the United States, we’re right in the middle of the Republican primaries as the country tries to decide who the GOP nominee for President will be in our election later this year. One of the more interesting conversations around the 2012 Presidential election is the relationship between what people say on Twitter and what they do at the polls. Can we use Twitter conversations to predict election winners? Or, if they can’t predict results, what can tweets tell us about how potential voters feel about the candidates?
With Super Tuesday approaching and the GOP candidate field still wide open, we’ve been tracking tweets about the six top candidates for the Republican Presidential nomination since January 1 – Newt Gingrich, Jon Huntsman, Ron Paul, Rick Perry, Mitt Romney, and Rick Santorum. From those tweets, we built an interactive visualization of how Twitter talks about the GOP candidates, and how that relates to poll numbers over time.
Check out our interactive Republican primary Twitter tracker here or click on the screenshot below.
To create this visualization, we’re using a set of TweetReach Pro Trackers to track Twitter conversation about each of the candidates, along with our API to update the visualization daily. In the visualization, we’ve mapped the number of unique Twitter users talking about a candidate to the y-axis, polling results to the x-axis, and tweet volume to the circle radius. Polling data is from RealClearPolitics.
We often use TweetReach to track the success rates of TV shows and other major media events. We thought it would be interesting to analyze the tweets during last night’s Iowa Caucuses for the Republican nomination for President. As you know by now, former Massachusetts Governor Mitt Romney beat Rick Santorum by just a few votes, and Ron Paul came in third. Could Twitter activity have been used to predict the winner?
Last Friday, we started tracking all tweets that mentioned any of a candidate’s Twitter accounts (personal and campaign), the major news networks’ coverage of the caucuses, and hashtags such as #iacaucus that were used by the major news media and others in their tweets.
Interestingly, the overall Twitter volume about the caucuses was pretty low. In fact, we often track more tweets in an hour about a single TV show than we have in five days about all nine candidates. Nevertheless, early on in the evening we predicted a win by Mitt Romney or Ron Paul based on early Twitter activity and retweets.
Overall tweet volume, the number of unique contributors (people who have tweeted about a topic), reach, exposure, and the retweet rate (average number of retweets per tweet) can be useful indicators for deciding what topics are most popular on Twitter. But can they help predict results in Iowa? Here’s how the data shook out for the six major candidates:
Based on overall reach, Romney, Santorum, and Paul came in as the top three candidates, mapping directly to the final caucus results. Based on this analysis, reach seems to be a good indicator of success. But, since much of this reach can be attributed to mentions by major news media accounts, it’s more likely that Twitter activity is merely descriptive of what is happening. Nevertheless, the percentage of total reach from the major candidates ended up being very close to the actual caucus results:
Also noteworthy, despite having over 2.5x the tweet activity of Romney or Santorum, Ron Paul only had the third highest reach. Paul also had over 1.5x the contributors and the highest retweet rate of the candidates, more likely an indication of his support among younger voters and their engagement on Twitter. But, a larger follower count and more activity on Twitter don’t necessarily help predict a winner.
Other fun facts, the most retweeted tweet in our analysis came from Ron Paul’s account, and mentions Jon Huntsman who didn’t actively campaign in Iowa:
And, the second-most retweeted tweet came from Robert Reich, professor at University of California at Berkeley and former United States Secretary of Labor:
Studies have shown that Americans use social media to follow politics. As the primary season unfolds, we’ll continue to analyze the Twitter activity of the major candidates and report back on what we find. In the mean time, we’d love your feedback!
By now you’ve probably seen one of our posts about this season’s new fall TV shows. For a few weeks, we’ve been using TweetReach to track tweets about all 25 new shows (we’re down to 22 now), and using the tweets to try to predict which ones will be canceled. And we thought it would be fun to bring a guest blogger who knows even more about TV than we do to help make predictions.
So, welcome Adam Rucker to the TweetReach blog! Adam’s been blogging and making videos about TV and pop culture for a long time. He’s even appeared on TV a few times. Here on our blog, Adam will sharing some of his – and Twitter’s – thoughts on new fall shows. And if you like what you see here, you can find Adam on Twitter at @ruckermore, on his YouTube channel, and on his blog.
This week, Adam takes on FOX’s new show, Terra Nova. Will it be canceled? Let’s see what Adam thinks!
One of the biggest bets of the fall season is the one FOX took on its new sci-fi series Terra Nova. The show, which begins in the year 2149, stars Jason O’Mara as the head of a family that travels 75 million years into the past to live amongst the dinosaurs in “Terra Nova.”
The premise of the show is exciting in nature: super director Steven Spielberg produces the time-traveling mix of Lost, Jurassic Park, and Avatar. It’s also the most expensive new show in production this year with a pilot that cost a rumored $20 million to create and subsequent episodes that cost around $4 million each.
Unlike most shows, FOX ordered 13 episodes of Terra Nova when the original pilot was greenlit, meaning it’s unlikely that FOX will pull the plug on the show before it shows all the episodes it’s already paid for. Still, it is the viewer response to these episodes that will determine if FOX decides to continue pouring money into its investment or fill Terra Nova’s valuable Monday night time slot with another spinoff of Hell’s Kitchen starring Gordon Ramsay.
So what is the Twitter world saying about the big budget drama? In its first week on the air, Terra Nova generated nearly 90,000 tweets from more than 50,000 contributors reaching about 18.2 million people, which is nearly 10 million more than reached the recently cancelled Playboy Club (12K tweets from 9K contributors, with a reach of 8.6 million). Interestingly, the several weeks since the premiere haven’t generated much more attention for the show. In total, 111,000 tweets have reached 20.7 million pairs of eyes. But the attention doesn’t mean anything if it’s bad attention.
A look at the top four highest exposure tweets includes three from Entertainment Weekly linking to articles on the show, but number four is a simple review from English television host, Jonathan Ross:
This tweet was retweeted 111 times, reaching even further beyond @wossy’s own 1.2 million followers.
A tweet by E!’s television critic, Kristin Dos Santos, mocking the show’s inferiority to one of her favorites, Lost, reached her 73,000 followers and was retweeted 28 times.
Personally, I wasn’t a huge fan of the show either. Despite the big bucks spent on production, it came off as cheap and even cheesy in some parts.
But to be fair, not all of the Twitterverse had bad things to say about the show. Drew Carey’s positive review went out to his 627,000 followers and gained a spot as one of the highest exposure tweets about the show.
But what does it all mean anyway? For an expensive show like Terra Nova, my guess is a lot. A thumbs up or thumbs down from any one of these influential tweeters could very easily result in the loss or gain of hundreds of thousands (or in some cases, millions) of viewers. While various reports show that Terra Nova’s ratings have been “respectable,” there’s no getting around the fact that it is up against some stiff competition, including ABC’s ratings behemoth, Dancing with the Stars.
It just depends on what FOX executives are looking for. The network recently picked up its new comedy series, New Girl starring Zooey Deschanel, for a full season. While the budget of this good-natured, apartment-based comedy is probably a tenth of Terra Nova’s (and also generated far fewer tweets and tweeters), New Girl has reached nearly two million more Twitter users during its time on the air.
There’s still time for Terra Nova (at least 10 more episodes), but my guess is that, unless it gains a devoted following (quickly), FOX is going to stop paying the bills and its 13th episode will probably be its last.
Do you think Terra Nova is headed for extinction? Leave your thoughts in the comments!
This week, NBC announced that it’s canceling both The Playboy Club and Free Agents. So, our question is, did the tweet numbers predict this? The answer is a resounding yes in the case of Free Agents, but in the case of The Playboy Club, the answer is slightly less obvious.
In its first week on the air, The Playboy Club garnered more than 12K tweets from 9K contributors, generating a reach of more than 8.6 million. And during the last three weeks, there have been more than 36K tweets posted about the show. These are not insignificant numbers; The Playboy Club consistently fell in the middle of our rankings based on volume, reach and contributors. But the picture looks less rosy when we dig into some of the tweets about the show and who’s posting them.
@HughHefner and other Playboy-affiliated accounts drove much of the conversation about the show. Hefner tweeted 30 times in three weeks about it, generating 22% of all tweet impressions about The Playboy Club. Other popular tweets called The Playboy Club a poor imitation of Mad Men and made jokes about not watching it. More than 8K tweets were posted the day the show was canceled, making it the highest volume day so far for the show. That’s probably not a good sign – more people talked about the show being canceled than the show’s premiere.
As for Free Agents, it was one of the three shows we discussed last week as a sure bet for early cancellation. It didn’t receive much attention on Twitter, only generating 8,900 total tweets over three weeks. And the little conversation it did spark was pretty lukewarm – no one seemed to love it and no one seemed to hate it – and even the cast seemed to sense the show would be canceled. Here’s a tweet from the show’s leading actor, @HankAzaria:
So, really, no surprises there. And now we wonder, what show will be canceled next? Will it be FOX’s big and boring Terra Nova? Some of the quieter CBS shows like A Gifted Man or Unforgettable? Or will it be one of CW’s relative duds like H8R or Ringer? We’ll know soon, and we’ll keep you posted.
So far, 20 of this fall’s 25 new network shows have aired. We’ve been tracking tweets about all 25 shows for more than two weeks and today it’s time to check in with our data and see how the shows are doing. I’m going to take my first guesses at which new shows will be cancelled this season, based on a few of our standard quantitative TweetReach metrics. A few caveats before we begin, however…
First, since these shows premiered and air at different times, these initial metrics will be somewhat biased towards the shows that began the earliest in the season. In the next few weeks, this advantage will disappear, but for now, take these numbers with a grain or two of salt. Consider them directional indicators for now. And I’m not including stats for the five shows that haven’t aired at the time of posting. Second, different networks have different audience and revenue expectations from their shows. So just because one show has a smaller reach does not mean it’s less successful than a show with a larger reach on a different network or targeted to a different audience. Finally, this week’s predictions are based on numbers only – we’ll get into the more qualitative and content-based data next week. A terrible show can still generate nearly as much conversation on Twitter as a really good one, and we’ll sort out some of those distinctions in future posts.
So, on to the predictions! I’m going to try a few different models this week to get started. I’ll update these predictions next week when we have more data.
Cancel the lowest performing show on each network, based on reach:
- ABC – Suburgatory* (14.1M)
- CBS – A Gifted Man (3.0M)
- CW – Ringer (6.0M)
- FOX – Terra Nova (16.5M)
- NBC – Free Agents (3.3M)
*ABC still has three shows yet to air, so any of them could take over the lead in this position. Also, Suburgatory premiered just last night, so it could easily catch up to its peers in the next week or two.
Cancel the overall lowest performers, based on tweet volume:
1. CW – H8R (1K tweets)
2. CBS – A Gifted Man (3K tweets)
3 (tie). CW – Ringer, NBC – Free Agents, NBC – Prime Suspect (5K tweets)
4. CBS – Unforgettable (6K tweets)
5. ABC – Suburgatory (8K tweets)
Cancel the shows with the fewest people talking about them, based on unique contributors:
1. CW – H8R (<1K)
2. CBS – A Gifted Man (2K)
3 (tie). CW – Ringer (3K), NBC – Free Agents (3K)
4. NBC – Prime Suspect (4K)
5. ABC – Suburgatory (6K)
6. NBC – Whitney (7K)
Based on these three lists, I’d consider A Gifted Man, Free Agents, and Ringer pretty sure bets for cancellation. They show up in all three categories. (I’m giving Suburgatory one more week before we count it out, since it just started yesterday and the others have had more time to generate conversation.) And it’s not looking good for H8R or Prime Suspect either.
Finally, since I’d hate to end this post without saying something about the shows that are doing well on Twitter, here’s a few stats about some of the top performing shows.
The top five shows that seem safe, based on highest reach:
- X Factor (37.3M)
- New Girl (21.5M)
- Charlie’s Angels (19.6M)
- Revenge (17.3M)
- Terra Nova (16.5)**
**While Terra Nova is FOX’s lowest performing show so far (it’s competing against the X Factor and New Girl), it still has a higher reach than most of the other shows.
The top five shows that seem safe, based on unique contributors:
- FOX – X Factor (73K)
- FOX – New Girl (40K)
- FOX – Terra Nova (37K)
- ABC – Revenge (26K)
- ABC – Pan Am (25K)
We just learned that FOX’s New Girl was picked up for a full season. Based on the tweets, that seems like a good choice. It’s also looking pretty good for NBC’s Up All Night, CW’s The Secret Circle and Hart of Dixie, and CBS’ 2 Broke Girls, but I wouldn’t count my predictive chickens before they hatch. We’ll see what the tweets can tell us after another week of these shows.
So, what do you think? Have you seen any of these shows? Do these lists ring true with your experience? Tell us which shows you think will be cancelled (or picked up) in the comments.
Until next week, happy watching!
At TweetReach, we’ve tracked a lot of tweets about television, from special events like the Oscars, the Golden Globes and the Super Bowl, to regular weekly episodes of many of your favorite shows. That’s because we (and our customers) know that Twitter can tell us a lot about what an audience thinks about a show, from how much viewers tweet about a show, to when they tweet about it, to what they actually tweet about. There’s a lot we can learn about a TV program’s success just by analyzing the tweets about it.
So we thought it would be fun to track this fall’s 25 new shows on the five big broadcast networks (ABC, CBS, CW, FOX, NBC) to see what Twitter thinks about them. But we’re not content to just rank Twitter’s favorite shows. Oh no. Inspired by the New York Times’ Fall TV Season Ratings Pool and based on Twitter chatter, we’re going to predict what new fall shows will be canceled.
Want to play along? Leave your predictions in the comments. Which new shows do you think will be canceled? Or, even better, which new shows do you wish they’d cancel?
We’ll be posting throughout the fall with our updates, predictions and conclusions (and maybe we’ll even feature some special guests along the way!). For reference, here’s a list of the new fall shows, ordered by network.
- Charlie’s Angels
- Last Man Standing
- Man Up
- Once Upon a Time
- Pan Am
- 2 Broke Girls
- A Gifted Man
- How to Be a Gentleman
- Person of Internet
- Hart of Dixie
- The Secret Circle
- New Girl
- Terra Nova
- X Factor
- Free Agents
- Prime Suspect
- The Playboy Club
- Up All Night
Last week, we used a chunk of tweets about the Academy Awards to attempt to predict who would win the Oscars. Our main assumption was that the actors and film that generated the largest reach would be the ones to win. Based on that assumption, our Oscar winner predictions were:
- Best Supporting Actress: Hailee Steinfeld
- Best Supporting Actor: Geoffrey Rush
- Best Actress: Natalie Portman
- Best Actor: Colin Firth
- Best Picture: The King’s Speech
So, how did we do? Was our particular method of tweet analysis a winning one? Here are the actual winners (our predictions are in brackets, correct ones in green, wrong ones in red).
- Best Supporting Actress: Melissa Leo [Hailee Steinfeld]
- Best Supporting Actor: Christian Bale [Geoffrey Rush]
- Best Actress: Natalie Portman [Natalie Portman]
- Best Actor: Colin Firth [Colin Firth]
- Best Picture: The King’s Speech [The King’s Speech]
So, we did pretty well! We got the big three – best actor, actress and film. We missed the supporting actor and actress categories, but we used a different analysis method for those categories, so that might have something to do with. We’ll post a more detailed analysis of our Oscar data later this week, including a review of these predictions and explanation of our methodology. For now, we just wanted to give you the update. More soon!
We’ve been tracking tweets about the Academy Awards for about a month. In that time, 213 thousand people have tweeted more than 417,000 times about the Oscars, reaching 59 million unique Twitter accounts and generating more than a billion impressions.
Over the past month, we’ve used a variety of methods to attempt to predict who will win the 83rd Academy Awards, based on tweets about the nominees. This post includes our third and final round of predictions. (See our first and second rounds here.) So, here are our final Academy Award winner predictions, based on the cumulative unique reach* of the nominees.
- Best Supporting Actress: Hailee Steinfeld (True Grit)
- Best Supporting Actor: Geoffrey Rush (The King’s Speech)
- Best Actress: Natalie Portman (Black Swan)
- Best Actor: Colin Firth (The King’s Speech)
- Best Picture: Black Swan The King’s Speech
Here’s the full list of the ten Best Picture nominees, sorted by reach.
While there are more tweets (and a higher reach) in the Black Swan Tracker than in any other film’s Tracker, this is one of the noisier Trackers we’re running. A number of tweets about Black Swan aren’t actually related to the film (many are related to the ballet or are generic references to the term “black swan”). Because of this, its position at the top of the list of Best Picture nominees is tenuous. As a comparison, a majority of the tweets in The King’s Speech Tracker are related directly to the film. Given that, we believe The King’s Speech to be the true front runner in the Best Picture race.
As a reference, here are our week-by-week comparisons.
We’ll be monitoring tweets as they come in during the awards show on Sunday, so follow @tweetreachapp on Twitter and check back here next week to see what Twitter thought of the Oscars and the award winners.
*Reach is the total number of unique Twitter accounts that received tweets about the Academy Awards. These data reflect cumulative reach since January 24, 2011. To measure reach, tweet volume and other stats, we set up a TweetReach Tracker for each Best Picture, Best Actor and Best Actress nominee.