By Larry Schweikart
It’s safe to say those on the right were shocked with the election returns by Wednesday Nov. 4th. In many cases, the results not only confirmed the pollster’s doomsday predictions for Donald Trump, but exceeded them. This came as a double blow because polling has rightly been beaten up since 2016, though some of the pollsters came close in 2018. The dismal polling that showed a Hillary Clinton quasi-landslide four years ago was utterly horrific in many cases at the state level, where the average error in Wisconsin was 6 points, the average error in Ohio was four, and not a single one of the major pollsters had Trump winning the electoral college.
Now, four years later, many of the pollsters are gleeful, savoring the fact that Donald Trump underperformed in some of the suburban communities he carried in 2016, especially in some Philadelphia and Phoenix suburbs. Their national predictions will be held up as vindication that, once again, you can trust polls. When an entire data set is corrupted, however, can you trust anything? Was the entire data set corrupted? Were any of the polls right?
I had an opportunity to discuss this with America’s Pollster, Richard Baris of Big Data Poll/People’s Pundit Daily. Baris made a name for himself in 2016 when he alone accurately predicted a national victory for Hillary Clinton by about 1.5 points, but predicted Trump would carry the key states of Florida, North Carolina, Ohio, Pennsylvania, and Michigan. His only miss in 2016 was Wisconsin. (A story in itself, as his polling showed Trump winning Wisconsin but admitted that he didn’t believe his own polling and predicted Wisconsin as a narrow victory for Clinton.)
We discussed the future of polling, given the rampant and undeniable fraud in many parts of the United States.
LS: I see where Dave Wasserman, for example, is touting certain New York districts where Trump under performed by about six points. How in the world could Trump under perform in districts he won in 2016 but still win over 11 million more votes in the popular vote? What confidence do we have that any analysis or polling was right, given this fraud?
Baris: Wasserman is cherry-picking. He does that all the time. You have to look at places where the integrity of the ballot process. In those places where we have that, where the chain of custody of the vote was not compromised, where the secretaries of state had a final election night count of how many ballots were outstanding, then we had good polling. For example, Texas, Ohio, parts of Arizona were very good. They were done correctly.
LS: You and I both had Florida dead on, not just in Trump and Biden numbers, but also in black and Hispanic percentages.
Baris: Florida was the gold standard of how to conduct an election. Not one state supervisor missed a deadline. The state had great precision. In the past, you know, we would have these ballots from Broward coming in late and always be wondering, “When is Broward done?” Not this time. The integrity deadline is crucial. That’s when there is a final tally of how many votes still remain out there to be counted. So, you get to some of these districts that Wasserman and others cherry-pick, and yeah, Trump did slip a little bit there, but he did better in other districts in New York, for example, Kings County. Even in Indiana, there were some areas where Trump could have done better.
LS: So where did Trump’s additional 11 plus million voters come from?
Baris: Trump actually did better in urban areas. He still lost these, but did better, for example, in Cook County, in Miami
LS: Still, we have these states where you and others had Trump slightly ahead—Michigan and Pennsylvania—where Trump supposedly lost if these counts hold up.
Baris: I will take it to my grave that I got Pennsylvania right, with Trump winning by just under a point. I will also admit I was off in Arizona by two. Pennsylvania is full of fraud. I wouldn’t be able to tell any client in Michigan or Wisconsin that polling is accurate.
LS: We’ll get back to some of this fraud later. But I want to get back to the national pollsters. Were they right?
Baris: Let’s look at some states, OK? Nate Cohn was off by 11 in Iowa, which Trump won. These guys were off in Ohio, with Cohn having Biden plus five and Trump winning it by eight. That’s an error of 13. Quinnipiac had Biden up four in Ohio, so they were still off by 12.
LS: You and I talked about this in 2016, and it appears it happened again: these errors were always all in the same direction.
Baris: Exactly right. You might have some confidence in them if they were occasionally wrong the other way. Where are their 5-point overestimates for Trump? Anywhere? There aren’t any.
LS: At least this time around, you had some of those going in both directions, which seems to validate your polling as not being ideologically driven.
Baris: Right. I already said, I missed Arizona by two. I missed Minnesota by about two or three in Biden’s direction, though. He did a few points worse than I thought he’d do. The point is, it’s one thing to miss—we all do on occasion—but when you’re always missing in the same direction it’s more than just an accident, or a bad call.
LS: Something else is clearly driving that. So how do we know when polling in the future is likely to be on? Even yours?
Baris: The best we can do is weigh data that is more trustworthy in those states that have validated, reliable election procedures. Nationally in this election, the data has been corrupted, but there are pockets where we were dead on. And those happened to correspond to the states and counties that had reliable, verifiable procedures. So, I had Trump at 12 percent black and that’s exactly where he came in.
LS: That’s where I had him too. How did he do with Hispanics?
Baris: This is coming from exit polls, and those are even more unreliable than regular polling, but it looks like he got 32 percent with Hispanics—up a lot from 2016, but not up near the 38 percent that some were predicting. The exits had the white vote at 58 percent. I think that’s low by 2 percent.
LS: In Arizona, the claims of Dominion machines adding certain percentages of votes to Biden have a problem: there is a paper trail of ballots. So, for example, if you saw that on Oct. 27 Biden got 324 more votes in Cochise county, and want to claim those are from a machine, it’s hard. Because you have the same number of paper ballots there. Someone would have to be in every single precinct and know exactly how many votes were being added every single day and then inject exactly that many paper ballots into the system. But in Michigan and Georgia and elsewhere, we have ballots just appearing that don’t match the computer counts.
Baris: What Sidney Powell has from her whistleblowers are sworn testimonies in some of these other places that the votes don’t comport with how the votes are entered and worked. So it’s a question.
LS: The Biden people say, “Look in Michigan some of the elections people found the errors themselves and fixed them.”
Baris: The only reason these people in Michigan caught the last batch of errors is that they were in Trump-friendly districts. They were willing to go through the process of re-calibration. This hasn’t happened in non-Trump districts, even those where Republicans are in charge. They just let it happen. A recount just records the same ballots. You need an audit to take out the ballots, review the whole system, with both sides present. That still hasn’t happened in Georgia, Wisconsin, Michigan, or Pennsylvania.
LS: What then, could have been done about this fraud?
Baris: Trump didn’t have time to get his machine of re-counters and faithful people in place before a lot of this happened. It’s going to come down to Georgia, Michigan, Wisconsin, and Pennsylvania
Larry Schweikart is the co-author with Michael Allen of the New York Times No. 1 bestseller, A Patriot’s History of the United States; author of Reagan: The American President, and founder of the Wild World of History, an online history curriculum with complete U.S. and World curricula for grades 9-12 that includes teacher’s guide, student workbooks, maps/graphs/charts, tests and answer keys, and video lessons for all units.