On election night in Alberta, as the numbers from a provincial referendum began to flash across the screens of the official Elections Alberta website, a quieter but more disturbing story was unfolding on social media. People were scrolling through the live results on X, formerly known as Twitter, and seeing patterns that did not make sense to them. Some noticed that one referendum question seemed to have far more votes counted than the others. Others noticed that the total number of votes reported seemed to exceed the number of eligible voters in the province. In a moment of heightened political anxiety, these observations quickly transformed from curiosity into accusation. Within hours, posts were circulating with dramatic headlines, claiming that mail-in ballots had been quietly dumped to rig the outcome and that the election had somehow processed more votes than human beings who were eligible to cast them. The posts received tens of thousands of views, and the language was charged with suspicion and betrayal. “That smells of corruption,” one post declared. But according to Elections Alberta and political experts, these claims were based on a misunderstanding of how the referendum worked. What looked like fraud or manipulation was, in fact, a carefully structured counting process that had been publicly announced months earlier. The real story, as is often the case with election information in the digital age, was not about corruption but about the gap between how democratic systems actually operate and how ordinary people believe they should operate.
The first and most emotional claim centered on what one social media user called a “dump” of mail-in ballots. The person posted screenshots of live preliminary results from the Elections Alberta website, showing that the tenth question on the ballot—the question about whether Alberta should begin negotiations toward independence from Canada—had substantially more votes tallied than the other nine questions. To that user, the only explanation was fraud. Hundreds of thousands of special ballots, the post alleged, had been released all at once to tilt the independence question in favor of Alberta remaining in Canada. It was a dramatic story, and it spread quickly. But Duane Bratt, a professor of political science at Mount Royal University in Calgary, called the claim “ridiculous.” He explained that the sudden appearance of results for the independence question was not a surprise. It was the plan. The Alberta government had announced months in advance that results for the independence question would be released first, and election workers had begun counting advanced polls and mail-in ballots hours before the polls even closed. They started at five o’clock in the afternoon, long before election night’s tense count began. By the time the polls closed and the results were allowed to be posted, thousands of ballots had already been processed and verified. All that was left was to press the button. So when the first results appeared within minutes, showing a strong lead for staying in Canada, it looked sudden and suspicious to people who had not followed the process. But it was simply the culmination of hours of patient, careful work by election officials who were doing exactly what they had been trained to do.
The gap between the number of votes for the independence question and the number of votes for the other questions was also not a sign of secrecy or manipulation. The referendum ballot contained ten separate questions, and voters were not required to answer all of them. They could choose to submit one ballot, five ballots, or all ten. In fact, many voters were actively encouraged to vote only on the independence question, because that was the question they cared about most. Others found the wording of the remaining nine questions confusing and decided to skip them. At polling stations, election workers counted the ballots for the tenth question first, then moved on to questions one through nine. As Bratt explained, “In each polling station, they counted Question 10 first, then they went through one to nine. That is why there was a discrepancy, and there continues to be a discrepancy in the totals that emerged.” In other words, the numbers were not coming from some hidden stash of ballots. They were simply the natural result of a system in which voters had different levels of interest and engagement with different questions. Some people filled out their entire booklet. Others submitted just one ballot, folded it, and dropped it into the box. The same principle applied to special ballots sent through the mail. A person could submit one ballot or ten. There was no rule that every voter had to cast a vote on every question. This is not unusual. In many elections with multiple items on the ballot, such as bond measures, municipal questions, or advisory referendums, voters routinely skip items they do not feel informed about. The result is often a gap in total votes between different items. But when that gap appears in a tense political environment, with social media amplifying every number, it can easily be misinterpreted as evidence of wrongdoing.
The second major claim was even more dramatic. A social media user noted that Elections Alberta had reported 3,148,343 eligible voters in the province. Then they looked at the accumulating vote totals on the official website and noticed that the numbers added up to more than 4.3 million votes. The post concluded that there were more votes than voters, which seemed impossible—and therefore, in the minds of its audience, proof of something deeply wrong. “It is hard to believe 100 per cent of eligible voters took part in the referendum,” the post said. But the confusion lay in a misunderstanding of what the word “vote” means in this context. In a normal election, each voter casts one vote, and the number of votes should be roughly equal to the number of voters. But this referendum was different. Every single voter received ten ballots, one for each question on the sheet. Each of those ten ballots counted as a separate vote in the official results. So if every eligible voter had participated and answered every question, the total number of votes cast would have been more than 31 million. The fact that the reported total was around 4.3 million did not mean there were more voters than people. It simply meant that, on average, the people who took part answered roughly one and a half questions out of the ten. That is a sign of selective participation, not of voter fraud. It is also a sign of how focused many people were on the independence question. They came to vote on that one issue, cast their ballot, and went home. The Elections Alberta website made this distinction clear, separating the number of eligible voters from the number of votes cast across the ten questions. But that distinction was easy to miss for someone scrolling through the live results on a phone, trying to understand what they were seeing in real time.
In response to the spreading rumors, Elections Alberta issued a firm and unusually direct statement. The office said that the allegations of a mail-in ballot dump were false, and that the order in which results were released had been announced on May 29, months before the vote took place. They also noted that they had received 519,082 special ballots on the independence question alone. By Wednesday, the live results page showed more than two million total votes cast on that question, with 95 percent of locations reporting. The office went further, warning that the misinformation was not harmless. “This allegation now is disinformation at best and an attempt to deliberately undermine faith in the democratic process at worst,” a spokesperson said. That warning was echoed by Bratt, who saw the posts as part of a troubling trend in which election processes are deliberately misrepresented in order to erode public confidence. Before the vote, many people on social media had encouraged others to vote only on the independence question. Some had even urged their followers to submit blank ballots on the other questions. When the results came in, the same people were pointing to the very discrepancy they had helped create as proof of a conspiracy. It was, as Bratt suggested, a cycle of misinformation that feeds on itself. First, people are persuaded to vote in a certain way, or not to vote on certain questions. Then, when the official results reflect that behavior, the same voices cry fraud. The online audience, already primed to distrust institutions, accepts the accusation without checking the details. The real damage, he argued, is not to any particular outcome but to the basic belief that elections can be fair and trustworthy.
At its heart, this entire episode is a story about how human beings make sense of complex events at a time of deep uncertainty. People are not naturally cynical. They look at numbers, they look at patterns, and they try to find meaning. When the patterns seem to contradict what they expected, especially in an area as emotionally charged as the question of whether their province should leave Canada, their minds leap to the worst conclusion. They assume that someone is cheating, that the system is rigged, that their voice is being drowned out. What is missing is context. The context that mail-in ballots were counted in advance. The context that results were released in a specific order. The context that each voter received ten ballots. The context that thousands of voters chose to answer only the question that mattered to them. None of this was secret. It was all printed on the website, announced in advance, and explained by officials. But context cannot compete with emotion. In the minutes and hours after the polls closed, people were not reading fine print. They were watching numbers rise, and their hearts were racing. They wanted to know, immediately, whether their side had won or lost. When the numbers looked wrong, they reached for the darkest explanation available. That is a profoundly human reaction. But it is also a dangerous one, because it allows those who are deliberately spreading disinformation to use ordinary people’s confusion as a weapon. The remedy is not to shame or mock those who fall for false claims. It is to slow down, to breathe, and to remind ourselves that elections are run by human beings who are not perfect but who are, in this case, following a clear and publicly documented process. The final results will come out, not because of a conspiracy, but because hundreds of thousands of Albertans took part in a complicated question of civic life, and because the people who counted those ballots did their jobs with care and integrity. In the meantime, the best thing any of us can do is ask questions, check sources, and remember that a number on a screen is rarely the whole story.