A risk-limiting audit (RLA) is a type of post-election audit. It requires manually checking statistical samples of paper ballots to see if official election results interpreted and tallied the ballots correctly. When contests have wide margins, the samples can be quite small, quick and inexpensive to check. If any contest is close, the sample ballots need to be a large fraction of all ballots, to minimize the chance of missing mistakes. Audits depend on checking paper ballots, which therefore need to be transported, handled and stored securely from election day until the audit, in a manner such that even insiders cannot alter or replace ballots. If the audit finds that an election had flaws in computer processing, the audit will recover correct results by using a larger sample or by hand-counting all the paper ballots. If examining sampled ballots shows flaws in ballot storage, the usual approach cannot recover correct results, and researchers recommend a re-vote if the number of ballots held in flawed storage is enough to change winners. An alternative to re-votes is to create backups of the paper ballots soon after they are voted, so there is an alternative to flawed storage of the original ballots. If audit samples find no error, there is still a known limited risk the results are wrong in the audited contests, as with any sample. The cost of hand-checking has meant in practice checking one or two non-randomly chosen contests in each jurisdiction, and not checking any of the close contests, because of the large sample sizes needed in close contests. In such cases the audit does not attest the reliability of unaudited contests. As with other election audits, the goal is to identify not only intentional alterations of ballots and tallies, but also bugs in election machines, such as software errors, scanners with blocked sensors or scanners skipping some ballots. The approach does not assume that all ballots, contests or machines were handled the same way, in which case spot checks could suffice. The sample sizes are designed to have a high chance of catching even a brief period when a scratch or fleck of paper blocks one sensor of one scanner, or a bug or hack switches votes in one precinct or one contest, if these problems affect enough ballots to change the result.