Often tests are not performed on a random sample of the population but only on people that are symptomatic.

But not all symptomatic people receive tests.

Knowing those two number (number of symptomatic people who would have qualified for a test, were it available; number of test performed) could paint q better picture.

Until recently Ontario had very stringent testing pre-requisites: 2 or more symptoms, history or travel in the last 14 days. Even then, on average, only 2% of tested people tested positively for SARS-Cov-2. In my opinion that makes the notion of much more larger population having it and not noticing due to the lack of testing, at least at this stage, highly unlikely.

They should be doing some tests via random sampling anyways just to make sure.

Ideally, yes. But given the limited testing capability that is a politically hard thing to do. Even if 50-75% of cases are asymptomatic (and that is a very big if), that would mean that on average you have a number of asymptomatic people that is a factor of 1 to 3 of confirmed cases. While it would be great to isolate those people, at this point that is still not a huge number that means anything for the response.

Why should we assume different states have the same starting conditions?