When it comes to election irregularities in Georgia, smoke and fire are now revolving around the unprecedented number of mail-in ballots. Given the evidence we have already seen and the well-known opportunity for fraud with mail-ins, why would Republicans not use their full control over state government to prohibit this practice headed into the critical twin runoffs that will determine control of the Senate? And how can we trust the bizarre and incongruent unofficial results in so many other states?

On Tuesday, Georgia officials announced a second discovery of lost ballots that weighed heavily for Trump. A spokesperson for the secretary of state announced that Fayette County officials failed to upload a memory card that recorded 2,755 votes into the total county tally. Those votes just happened to break down as follows: 1,577 for Trump, 1,128 for Biden, 43 for Libertarian Jo Jorgensen, and seven write-ins.

The news out of Fayette County came just one day after Floyd County officials uncovered over 2,600 votes that went missing during the initial count on election night. That batch of lost and found ballots just happened to contain a net 800-vote advantage for Trump over Biden. The secretary of state is now opening an investigation into the election counting in both counties and has called for the elections director in Floyd County to step down.

In defending Al Gore's contesting of the election results for 35 days in 2000, then-Sen. Joe Biden said that he had every right to question things that just didn't pass the "smell test." Well, what is going on here just doesn't pass the smell test. We are to believe in an election with unprecedented mail-in ballots, the following sequence of events occurred:

• That Trump could be on pace to win the state by 4 points (on par with the final WSB poll), according to the New York Times algorithm after well over half the vote was counted;
• Then, suddenly, late at night, the counting stops just in the most Democratic areas;
• When the counting resumes, the votes come in so lopsided for Biden, over and beyond the typical urban Democratic margins, that Biden blows out Obama's turnout in these heavily minority counties;
• That every single batch of found ballots in the final tallying of the initial results, without exception, heavily favors Biden;
• But then, after pressure mounts for an investigation, all of the votes that were found in the audit as not having been included in the initial count weigh heavily toward Trump.

Does this really pass the smell test?

Also, if just one discovered batch of votes each in two of Georgia's 159 counties made up 8.7% of the gap between Trump and Biden, what else is missing? And that doesn't even begin to factor in potential fraudulent votes in Biden's tally from the near-impossible number of mail-ins in some counties.

In a closely divided county, with elections increasingly being decided by small margins up and down the ballot, how can we have confidence in the results, given what we know?

The number of votes cast in total, but particularly Biden's margins, in the Atlanta metro counties are very hard to understand, even if we account for the population growth in the region. Take Gwinnett County, for example. Yes, this is a quintessential suburb that has trended away from the GOP during the Trump era, but the numbers that Biden racked up are extremely anomalous to say the least.

The population increased by 4.6% since the 2016 election, but Biden crushed Clinton's vote share in the county by 46%. The population grew by 42,000, but Biden improved the Democrat vote share by 66,000. You might suggest that this county has simply soured on Trump, like many suburban neighborhoods around large cities. However, Trump himself actually improved on his 2016 performance and even did better than Romney in 2012. So this is not a matter of a mass number of crossover voters going for Biden. Somehow, Biden magically managed to turn out some untapped reservoir of voters nobody ever imagined and dramatically expanded the pie.

We see the same trend in the heavily African-American urban counties of Fulton and DeKalb counties. Here is the trend line of vote share by presidential choice since 2008 in Fulton County, along with the population size by election year.

We see that Trump's share of the vote grew modestly, as we would expect with him generally performing better among black voters than four years ago. But Biden's growth is unfathomable. The entire county grew by roughly 50,000 people (4.8%), but Biden managed to expand the Democrat vote share by 98,000 (35%).

Moreover, Obama, with his historic blowout election in 2008 that produced record black turnout with a highly sophisticated ground game, achieved a vote tally corresponding to 30% of the county's population. We are to believe that Biden, with no ground game and less enthusiasm from black voters than even regular Democrat candidates, somehow got 35% of the population to vote for him?

Here is the similar trend in DeKalb County, the other urban center of Atlanta:

The population grew by just 2.6% since 2016. Yet Biden's vote total was 28.5% higher than Hillary Clinton's, even though Trump also improved slightly. In this county where nearly half the population is black, Obama racked up votes of 36% of the total county population. Biden topped it at 40%!

It's true that turnout was higher nationwide, but not that much higher. It just seems that Biden achieved nearly impossible turnout levels just in the areas of the four or so states that he needed to do so. It's very hard to see how just in these places, he crushed Obama's turnout numbers that seem nearly impossible as a share of the population in an election where Republicans wound up winning all 27 of the House races rated as toss-ups by political analysts.

Democrats will contend that Biden simply lucked out by running a great mail-in ballot campaign. But given the complexities of properly filling out those ballots, especially for those filling them out for the first time, it's hard to imagine how Democrats could achieve a 1:1 ratio of retaining all of their traditional votes through mail-ins, then expanding the pie with even greater turnout of legal and properly filled-out ballots. It's hard to imagine this wasn't achieved through a mix of ballot-harvesting fraud and invalid votes, signatures, and other incomplete information.

Even liberals worried about this occurring prior to the election. Nate Silver's Five Thirty Eight blog warned on Oct. 13 that "rejected ballots could be a big problem in 2020."

According to the Election Administration and Voting Survey, less than 1 percent of the 33.4 million absentee ballots submitted in the 2016 general election across the 50 states and Washington, D.C., were rejected. 1 This year, though, rejection rates could be much higher because so many people are voting by mail for the first time and may not know the rules. According to research by David Cottrell, Michael C. Herron and Daniel A. Smith, voters without experience voting by mail are up to three times more likely to have their ballots rejected.

Yet, somehow Democrats managed to miraculously achieve the impossible, especially in the states where it mattered most. The New York Times referred to it as the "November surprise." In 2016, 6.4% of mail-in ballots in Georgia were rejected. This year, despite the rushed volume of first-time absentee voters, just 0.2% were rejected. In North Carolina, where Trump underperformed his showing in 2016, the rate of rejection dropped threefold, from about 2.7% in 2016 to 0.8% this year.

It doesn't take a rocket scientist to understand the fraud endemic to mail-in ballots even in a normal year, much less a chaotic election like this where a significant number of Democrats votes were processed this way. The 2005 report of the Commission on Federal Election Reform, chaired by former President Jimmy Carter and former Secretary of State James Baker III, concluded that "absentee ballots remain the largest source of potential voter fraud." Even the Washington Post admits that mail-in fraud goes as far back as the 1864 election. In that vein, it's truly shocking that Republicans in Georgia haven't immediately banned mail-in votes for the runoffs.

Could it be that all the glitches, ballot dumps, pre-final (but not recount) lost and found votes, and successful mail-in campaigns all worked in Biden's favor in the right places at the right time? Unlikely, but possible. However, as Joe Biden said following Al Gore's contesting of Bush's win in 2000, "At the end of the day they want to know – is it fair? Does it pass the smell test?"