How hard is it to match CVC-roots?

Straightforward application of the ‘same birthdays’ problem shows that it is surprisingly easy to find a match for any CVC-root when wordlists of many languages are compared simultaneously. A simple algorithm can be used to estimate the expected incidence of multiple CVC-matches. A test of a sample from Greenberg's ‘Amerind Etymological Dictionary’ using that algorithm finds no ‘cognate sets’ whose resemblances are clearly not the result of random factors. The same test can and should be applied to all comparative etymological dictionaries.