Chamisa, listen to the numbers!

Over the past few months, Nelson Chamisa has been full of bombastic and grandiose claims about the upcoming election.

He began by claiming that "if we don't get 70% of the vote, then they would have rigged the election," without elaborating on where he got this figure from. A few weeks later, his confidence (arrogance?) increased further, with his now infamous claim that ""If Mnangagwa wins 5% in a free election, I will give him my sister. I am betting on this because I know it won't happen." And just this weekend, he claimed that "The elections are a walkover. I will beat Mnangagwa hands down. I know if we have a free and fair election, I will win."

While any right thinking Zimbabwean has always been sceptical of these claims, until now we have had no way of proving our suspicions. It was Chamisa's word and unscientific and largely subjective claims of crowd sizes, versus logic and experience, and many chose to believe Chamisa.

Now fortunately, we have a measure by which to assess Chamisa's claims: scientific opinion polls. Not the voodoo opinion polls from questionable Kenyan organisations published in the past few weeks, but a reputable poll from the highly credible pan-African research organisation, Afrobarometer. Afrobarometer's credentials for delivering impartial and high-quality research are impeccable - A non-partisan, independent and non-profit research organisation, it has conducted seven rounds of surveys in 36 African countries over the past two decades, providing highly accurate results. It would be a fool to claim either the body is biased, or that its research methodology is flawed. And so, when Afrobarometer speaks, we should listen!

Their poll of 2400 Zimbabwean citizens of voting age, conducted in early May with a 2% margin of error, showed that Chamisa's claims are way off. Contrary to his boasts, Emmerson Mnangagwa currently leads Chamisa by 42% to 30%, with other parties taking 2%, and 26% still up for grabs.

In other metrics, it is also advantage Mnangagwa. 49% of Zimbabweans think he is doing a good job as president, as opposed to 33% who think he is doing badly, giving him a net approval of +16% - a solid and impressive, if unspectacular number (by way of contrast, an average of recent surveys give President Donald Trump a -10% net approval among Americans). Further, 46% of Zimbabweans like ZANU-PF, while only 36% like Chamisa's MDC.

There will of course be those who dispute these findings, pointing out that polling has wrong in recent elections. But this would be a mistake. Take the case of Brexit. Critics claim that the polls were "way-off" with Brexit, while in reality the final polling averages showed a dead-heat. So the result of 52-48% in favour of leave represented a 4% deviation from the polls. In this case, even if this Afrobarometer poll were also off by 4%, and say that swing is in favour of Chamisa (far from a certainty), ED still leads by 8%.

Or take Donald Trump's victory. The final polling averages showed Hillary Clinton with a 3.5% lead in the popular vote. She ultimately won the popular vote by 2%, meaning the national polls were off by just 1.5% (the fact she lost the election based on the Electoral College formula is largely irrelevant for this assessment of national polls). If this error were replicated here, ED would be leading by 10%.

Closer to home, in the 2014 South African election, pre-election polling showed the ANC winning 61-64% of the vote and the DA winning 20-24%. The eventual result was 62% to 22%.

This is not to say that polling is perfect or foolproof. Like any science it is imperfect. But it should certainly be taken as a better representation of the will of the electorate then the hyperbolic claims of a boisterous opposition leader.

The purpose of this article is not to claim that victory for ED is certain – that is not what the numbers say. It is clearly closer than 2013, and Chamisa definitely has a chance (albeit a small one). Instead, what this poll proves is that any claim that Chamisa is winning by a landslide is nonsense, and should be treated as such. Not an easy thing for his devoted supporters to accept, but the truth.

And so, with ED currently leading by 12%, and a reasonable margin of error to be around 4%, were he to win by up to 16%, or as little as 8%, that would be in line with the science. Were that to happen, the opposition's inevitable cries of foul play based on their expectation of a 70% win should be treated with caution. Maybe in the bubbles of the big cities and Twitter Chamisa is heading for a landslide, but in the country at large, ED looks likely to score a solid victory.

So for now, with 50 days to go until the elections, let us ignore the bombastic claims of interested parties, and listen to the numbers!