It’s probably not fashionable to say this now, but we really need better answers for those who mistrust election results. Yes, yes, the president has been deliberately sowing mistrust about the results, and his allegations are baseless. No, there is no evidence of widespread fraud. Yes, a lot of the conspiracies floating out there are wild and absurd. No, I don’t know how we can convince people that the election results are correct.

Still, there is a genuine problem, evinced by widespread public skepticism about election results. And it’s not just a Republican problem. In every presidential election since 2008, those on the losing side doubt if the election was free and fair.

Early on, when the election results were unclear, I fielded inquiries from people who opposed Donald Trump asking whether the voting machines or Republican secretaries of states (who oversee the electoral process in their respective states) could alter the results—to rig the results in his favor. Regarding some states, like Georgia, which have risk-limiting audits on their books, I had good answers. For others, I actually did not have reassuring words. Because, frankly, our process for conducting and auditing elections, while it has gotten better, still leaves much to be desired. 

Now that it’s clearer that the president has lost, we’re seeing widespread questions and claims about election reliability from his supporters. It’s not enough for the rest of us to say that those people are prone to conspiracies and hopeless. The way to inspire trust and to fight deliberate misinformation has to involve procedures and processes that work, and that provide better answers than merely repeating, “trust us.”

Hence my piece for the New York Times on electoral process reforms that I think are both doable and plausible—and urgent. 

Since 2008, partisan distrust of presidential election results has been substantial. In 2016, only 43 percent of Democrats believed that the election was free and fair; now, only 30 percent of Republicans do. Each party’s supporters are more likely to believe that the vote was free and fair if they won, and those on the losing side are becoming more suspicious of the results. With a defeated president trying for weeks to overturn an election he has falsely called fraudulent, our partisan breach will be hard to repair.

But electoral reform can still provide a better foundation of trust. Two decades since the 2000 Florida recount debacle revealed the shoddiness of how America votes, we should be able to provide a straightforward, sensible answer to anyone who asks, “How do you know the results are correct?”

Yet, we still do not have nationwide standards and procedures to assure Americans that results are reliable. Claims of widespread fraud are false, but we can do much more to provide stronger answers to those who might want to question the process or the results.

The true scandal is that we know what we need to do and have even begun to implement reforms in many states, but we have not instituted the changes nationwide.

What if the election was much closer in some of the key states? What if the voting machines that remain in many states appeared to play deciding roles? What if a sizable glitch was discovered in their software? What if an optical-scan machine had malfunctioned and the errors were discovered too late? 

In Georgia, a post-election audit (yay!) discovered thousands of ballots that had been overlooked—there had been a human error in uploading the ballots, which really calls into question a process that allows for such errors. The ballots were on memory cards that had been forgotten because of, apparently, bad user interface design. 

The memory cards are a component of Georgia’s new voting system, which uses a combination of touchscreens and printers to generate paper ballots. When voters insert their paper ballots into scanning machines, their votes are stored on memory cards.

But unlike the state’s previous voting system, the new computers don’t raise as large of a red flag if a memory card hasn’t been accounted for.

County election workers were accustomed to “obnoxious” color-coding on election computers to help them account for all memory cards, but the new system uses more subtle checkboxes, Mr. Sterling said.

The missing memory cards were found in three counties: Douglas, Fayette and Walton. The Douglas votes favored Mr. Biden. The Fayette and Walton votes favored Mr. Trump.

What if there wasn’t a risk-limiting audit in Georgia? What if the error had been large enough or the difference small enough to call the results into question? 

There’s a tendency to assume that widespread public mistrust is unfixable, that those people are not reality-based and so there’s nothing to be done. But mistrust—including the kind that’s deliberately fostered—does not thrive in a vacuum. Fixing what we can, and creating processes that are designed to resist efforts to cast doubt, is the right thing to do. There is no other way out of this crisis. Moreover, it’s wrong to assume that taking vital steps to secure elections are done just for their benefit. If this election had been closer, I have no doubt that the conspiracies and mistrust would sweep through people in both parties. That’s no way to run elections, or a country. It may not be possible to fix everything, but we can—we should at least try—to fix what we can.