Every election cycle, we hear the same question from voters: how do we actually know the count was right? It’s a fair question, and it deserves a better answer than “trust the process.” That’s where post-election audits come in — the quiet, unglamorous work that happens after the headlines move on, when election officials go back and check their own math.
At its simplest, a post-election audit is a comparison. Officials pull a sample of paper ballots, count them by hand, and check whether that hand count matches what the voting machines reported. If the numbers line up, that’s strong evidence the tabulation worked as intended. If they don’t, it’s a signal to look closer before anyone certifies a result. We think of audits less as an accusation and more as a routine safety check, the electoral equivalent of a pilot’s pre-flight checklist.
Not all audits are created equal, though. Some states still rely on fixed-percentage audits — always checking, say, 1% of precincts regardless of how close the race was. That approach is easy to explain, but it doesn’t scale well: a 1% sample might catch problems in a landslide election and miss them entirely in a race decided by a few hundred votes. A newer method, the risk-limiting audit, adjusts the sample size to the actual margin of victory, which means closer races get a harder look. Research from the Brennan Center explains why this approach is gaining ground among election administrators.
We’d argue the real value of audits isn’t just catching errors — it’s what happens when nothing goes wrong. A clean audit result, conducted in public, with observers present and the methodology explained ahead of time, is one of the most persuasive tools available for building confidence in an outcome. It’s much harder to dismiss a result as illegitimate when the paper trail has already been checked by hand, in the open, before certification.
Of course, audits only work if the underlying infrastructure supports them. That means paper ballots or paper backups, secure chain-of-custody procedures, and enough trained staff to do the counting correctly. None of that happens by accident — it requires sustained attention from election officials, lawmakers, and the public between election cycles, not just in the anxious weeks after polls close.
We don’t think audits are a cure-all. They can’t fix a broken registration system or resolve a genuinely contested legal dispute. But as a baseline check on whether the count matched the ballots, they’re one of the most concrete, verifiable tools we have. The more routine and unremarkable audits become — the more they’re just something that happens after every election, win or lose — the more they can do their real job: giving everyone, regardless of who won, a reason to trust the number on the board.