I think this is not implausible. Let's make a few coarse-grained assumptions and run some numbers:
Given the demographics of your district, let's say the top 30% are equivalent to the top 5% of the general population. (That could be because of intelligence leading to affluence; enriched early experiences leading to better brain development; or plain old hothousing.) Of those, 1/5 will actually be in the top 1%. What about the others?
Remember that there will always be measurement error. That puts those kids within range of scoring in the top 1% by accident. Most of them won't (they are just as likely to under-perform as to over-perform). But that's just on a single measurement. They get at least two shots (the two subtests of the OLSAT) and possibly more -- possibly several more -- if the parents go for private testing.
I could easily see something like half of those who are "really" in the top 5% managing to get at least one score in the top 1%.
