About
Forecasts for September 18th, 2020
Recently, several opinion polls were published in Georgia. Data came from international organizations the National Democratic Institute and the International Republican Institution as well as news media outlets Imedi, Rustavi 2, and Formula. This gave us enough materials to update our forecasts.
All surveys that are used in this forecast, were administered prior to the outbreak of Covid-19 in Georgia in early September. We do not have clear understanding how poll numbers were affected, although how volatile Georgian public opinion can get, we do expect some changes. In sum, the analysis below should be interpreted as something that reflects the situation ante, that is, a “pre-outbreak normal.”
As explained in the methodology section, we only use proportions that reflect decided voters, that is, those who have disclosed their preferences. On the one hand, this is due to the fact that some pollsters (for instance, Survation/Rustavi 2) only publish results that show proportions among decided voters. Next, on what we saw in previous forecasts, proportions that are calculated without undecided respondents are to a large extent accurate reflection of what the public’s voting intentions are. Indeed, we acknowledge that this approach has its perils, for instance, not accounting for “shy opposition” voters, or the uncertainty associated with relatively smaller sample size.
Overall, the Georgian Dream seems to have been leading a league table in our “simplistic” weighted averages model and Bayesian forecasts. Weighted averages predict the ruling party being supported by 53% of those who have declared party preferences; Bayesian model speaks about 49% support. The discrepancies between these models can be explained by the sensitivity of the latter to outliers in raw poll numbers.
Weighted averages model predicts that nine oppositional parties have chances of getting into the parliament, although all of them except the United National Movement (18%), poll between 1% to 5%.
While polls seem to have not shifted drastically, our Bayesian model suggests relative decrease in chances of the Georgian Dream garnering 50% of voters. While our previous forecast showed that the governing party had about 56% of doing so, now chances are about 48%.
Odds of overcoming a 40% “closing” threshold remain stable and high. The Georgian Dream has about 94% chance of obtaining enough votes for single-handedly forming the government.
There are almost negligible chances of the Georgian Dream either getting enough votes through proportional elections to have constitutional majority, or the United National Movement overtaking the ruling party.
Similar to our previous predictions, our models hint on the volatility of support for the Georgian Dream. At a first glance, a Covid-induced boost for the ruling party seems to be plateaued and the Georgian Dream was almost guaranteed an easy ride if elections were held in late August. Nonetheless, it is too early to call the winner, as polls open in almost eight weeks from now. As we have no information how polls shifted due to the second outbreak, and there is still enough time for the opposition (and the government) to campaign, one can argue that the fate of elections will be sealed in the last weeks of October.
The methodology behind predictions is available here. Replication code and dataset used above is available via Github.
Previous forecasts: