2020 may be an unconventional year on polling given the highly irregular nature of voting this year…namely how many people are voting by mail or dropping off their ballots rather than showing up in person to the polls on Election Day. This year will also tell us how real the “shy Trump voter” phenomenon is and, most importantly, whether it makes a difference. Republicans are generally behind in the polls and it may indeed be true that the Trump Administration’s handling of COVID-19 is the defining issue of the election, as the Biden campaign has been betting on. Incumbents have lost before due to factors that were even less in their control than this…in 1956 Missouri Republican Congressman Dewey Short, who had represented his Springfield based district since 1935, lost reelection because the state was suffering a major drought and the state’s voters took it out on Republicans in the polls that year. It is possible that the economy’s stronger rate of recovery than predicted will help the president, but it may be too late. I suspect that the subject of protests and “cancel culture” will not have the impact the president desires, but the latter may turn into a prime issue in a future election. There are many Senate seats in the balance this year, and for most the GOP is on defense. Although in the last three national elections, average polls have been biased to Democrats, they grew less so in 2018 as opposed to 2016. What the poll bias will be this year is up in the air given certain additional factors. There are two possible things that potentially harm the Democrats this year despite polling edges: their late entry into door-to-door campaigning (which is the most effective campaigning, I know from experience) and the aforementioned “shy Trump voter” phenomenon. I know to a certain extent that this is real (I have talked to a few self-reported “shy Trump voters”) but how real or relevant is yet to be known. The GOP has an additional buffer seat with Democrat Doug Jones almost certainly losing reelection there. Democrats may have an edge, however, with apparently high turnout. As of today, Biden is up 2.3 points on RCP poll averages in battleground states. This is outperforming Clinton’s 1.1 advantage in 2016. Let’s look at the poll averages now versus when Trump won in 2016 in current battleground states.

For Florida, the polling momentum seems to be breaking for Trump. Possibly undecideds are moving in his direction.

Biden wins in Michigan and Nevada.

Trump wins in Georgia, although the victory is narrow.

Trump is ahead in Iowa, just like last time in the polls. He wins there.

Trump is ahead in Ohio, just like last time in the polls. I think he prevails there.

Trump wins in Pennsylvania, polling momentum is moving his way.

Trump wins a narrow victory in Texas.

For the Senate:

Although the race seems to be narrowing a little bit in Arizona, Democrat Mark Kelly has had, however, a consistent polling advantage over Republican Martha McSally. There is only one poll in this entire election season that has put McSally ahead. Arizona seems to be moving away from the GOP, at least for the time being. Kelly wins the seat.

Colorado seems to have been the race in which it was a foregone conclusion that the incumbent would lose on the Republican side. Cory Gardner’s profile has fallen since he was elected in 2014 and the state will certainly not be voting for Trump in 2020. It is possible that he runs ahead of the president, but that will not be enough to save his keister, and this is despite Democrat John Hickenlooper’s missteps during the campaign. Although it is true that the “doomed” incumbent in 2016 appeared to be Ron Johnson of Wisconsin and he pulled off a victory, Hickenlooper has repeatedly been considerably outside the margin of error in recent polling of the race while Johnson’s numbers had moved into that margin by Election Day. Gardner goes down.

Georgia is looking like a nail-biter for the GOP. Perdue seems to be faltering in the polls lately to Jon Ossoff. Right now, it is unknown with Georgia’s other Senate seat whether it will be Kelly Loeffler or Doug Collins who faces off against Democrat Raphael Warnock in the runoff. The Democrats have their best shot with Ossoff and they could pull off a surprise if allegations of insider trading haunt Loeffler sufficiently should she best Collins in the runoff. My gut tells me the GOP holds both seats, with the former being very close.

Iowa seems to be moving in the president’s and Senator Joni Ernst’s direction lately. This seat is vital to keep if the Republicans want a shot at maintaining a Senate majority in the event of a Biden presidency. I think she narrowly pulls off winning another term.

Susan Collins is certainly in trouble in Maine. She hasn’t led in a single poll in facing against Democrat Sara Gideon since February, and it seems likely that Gideon takes the seat in a state that is not going to vote for Trump. Although Collins voted against Barrett in an effort to save her seat, its probably not enough.

Unfortunately for John James in Michigan, incumbent Gary Peters has led in nearly all polling for this race and averages over five points ahead. James may lose by less than the poll average here, but it would require a major upset for him to win this one. Major upsets are quite rare…but Trump did score one when he won Wisconsin in 2016. Don’t bet on it for this race. Peters wins another term.

Minnesota may indeed be the sleeper race for the GOP. Jason Lewis has been performing far stronger than was expected at the start of the campaign season and if he wins, he may be the one to save the GOP majority. The race has been narrowing in October, and the last two polls were within the margin of error. I am granting myself one wild prediction for a Senate race: Lewis wins.

Montana is rather iffy for Steve Daines as he is facing the only candidate the Democrats could field that has a chance of winning the seat in former Governor Steve Bullock. However, I give Daines a slight edge as Trump is leading in Montana and if he has made it clear he is an ally of the president and Bullock as someone who would have voted to impeach the president as did all Democratic senators, he will win another term.

Although Democratic challenger Cal Cunningham got hit with an extramarital affair scandal, Republican incumbent Thom Tillis is still endangered as he is widely regarded as mediocre and is still on average behind in polls, but within the margin of error. If North Carolina tips to Trump, which it seems to be doing lately in the polls, Tillis may just survive. Indeed, Tillis won in the 2014 election despite the Democrats’ valiant efforts to help Kay Hagan survive reelection and her leading in nearly every poll. Perhaps he will pull it off again. However, he is running behind the president, and President Trump is only slightly up. I think Tillis pulls through by the skin of his teeth.

Lindsey Graham gets an unusually close race against challenger Jaime Harrison, but he pulls through…enough voters mark their ballots for Trump and Graham for him to survive.

Thus, my overall prediction is the GOP has 52 seats in the Senate at the end of this.

I call the following:

• Incumbents Perdue, Ernst, Peters, Daines, Tillis, and Graham win reelection. Incumbents Jones, McSally, Gardner, Collins, and Smith are sent packing.
• Trump wins Florida, Iowa and Ohio.
• Biden wins Michigan and Nevada.
• At least two upsets occur in Senate races.
• The race is closer than the polls have it.