Summary
A series of snap polls show that Trump either won or tied in last night’s debate. The total number of responses was likely in the 350,000 to 400,000 range making these snap polls difficult to ignore. Further, many of the snap polls highlighted here took place on local news focused sites which we believe lends credibility as the likelihood of bots or over-enthusiastic supporters voting multiple times appears lower. Additionally, credibility is further increased by the fact that some of the snap polls were conducted by sites serving liberal-leaning states (Oregon, New Jersey, New Mexico) – in other words a pro-Trump location bias seems rather low.
One brand name debate-focused poll conducted by CBS News showed that Biden handily won the debate by approximately 7 percentage points, which more or less reflects the current national poll average. The poll had slightly over 1,000 respondents. As it so closely mimics national polls, it does not seem to provide significant new insights.
In general terms, the snap polls tend to depict a tight race with Trump having a slight advantage. Reading a little more into the results, we infer that Trump supporters are more enthusiastic on average. Though not a significant amount of work, they took the time to vote on-line in slightly larger numbers than Biden supporters. The enthusiasm gap which is implied here has been highlighted by poll data and by our own alt data analysis.
Betting markets tended to give the win to Biden but not to any great extent. From immediately before the debate to the morning following the debate, the probability of a Biden victory (according to predictit.org) has increased by around 1 percentage point. This is well within the normal daily fluctuation and basically says the consensus believes the debate did not change the outlook to any significant degree.
Our outlook continues the same, namely that the nation is closely divided and that the 2020 election will be far less of an easy win than consensus believes. Biden, according to headline nominal poll figures, clearly has the advantage. However, according to many of those same polls, Trump clearly has the advantage in terms of enthusiasm or intensity of support. Our alt data figures also show Trump having much higher levels of enthusiasm and interest, which should provide him the benefit of higher turnout.
There was no knock-out blow in the debate.
Snap Polls and Polls
Snap polls are haphazard. Basically, something like a website will set up an on-line survey and ask a question to get a quick read on a topic. Registration is normally not required, so responses are anonymous. The advantage of snap polls is that they often attract high numbers of responses, in many cases multiples of what you would receive in a normal poll. The disadvantage is that these responses are not adjusted in any way so they may or may not reflect the electorate. In theory, the higher number of responses should help to decrease distortions but some will likely occur regardless.
Polls are controlled and scientific. They adjust responses by demographic weightings and attempt to replicate how an actual election could play out. Respondents are at times paid or receive some benefit (respondents in the poll listed here received payment for participating). The disadvantage here is that the responses are not truly anonymous (though they could be nominally anonymous), the number of respondents is considerably lower, and non-transparent adjustments are made to the data so that it reflects the expected turnout in the election.
Neither snap polls nor polls are perfect and both have advantages and disadvantages.
Snap Polls from the First Debate
This snap poll was conducted presumably through multiple local news sites, which may be owned by a common parent company. At least three sites carried the same snap poll, nj.com, oregonlive.com, and al.com. These three sites appear to serve New Jersey, Oregon, and Alabama. But, it is unclear how many other sites fed into this snap poll. Regardless, it is the largest in terms of number of responses, with 300,467. Results:
A second snap poll was conducted on tvline.com. They do not provide number of respondents. Results:
A third snap poll appeared on wevv.com which according to Wikipedia:
“WEVV-TV, virtual channel 44, is a CBS/Fox/MyNetworkTV-affiliated television station licensed to Evansville, Indiana, United States, serving the Tri-State area of southwestern Indiana, northwestern Kentucky and southeastern Illinois.”
The snap poll had 9,191 responses. Results:
Another snap poll was conducted by KRQE.com which according to Wikipedia:
“KRQE, virtual and VHF digital channel 13, is a dual CBS/Fox-affiliated television station licensed to Albuquerque, New Mexico, United States and also serving the state capital of Santa Fe.”
With 42,973 responses, the snap poll showed:
CBS Poll
CBS News conducted an actual poll which it describes as follows:
“This CBS News survey is based on 1,039 interviews of likely voters who reported watching the presidential debate on Tuesday, September 29, 2020. An initial survey was conducted by YouGov between September 25-28, 2020 using a nationally representative sample of 7,447 U.S. registered voters, including 4,622 likely voters who planned to watch the debate. Respondents who planned to watch the debate were recontacted for this survey. Only those who watched the debate were included in the analysis.”
Its headline results of who won the debate:
CNN Poll
We should mention the CNN poll conducted by SSRS. According to the poll:
“The study was conducted for CNN via telephone by SSRS, an independent research company. Interviews were conducted on September 29, 2020 among a sample of 568 registered voters who watched the presidential debate.”
This poll appears to be an outlier, producing the following results as to which candidate won:
Not only are these figures considerably different than others but they are unusually similar to CNN’s poll of the first debate in 2016 in which they put Clinton at 67% and Trump at 27%. As an outlier, caution should be used when interpreting the results.
TeleMundo Snap Twitter Poll
Another outlier was the twitter poll put up by Spanish-language Telemundo which showed that 66% of respondents voted in favor of Trump winning the debate whereas 34% stated Biden won. Like the CNN poll, these results are so outside the norm of what we have seen either in other debate-focused polls or in national polls, that we would recommend caution using these results as well.