Mark Lubratt <mark.lubratt@indeq.com> writes:
> I guess I was looking more at the cost estimate and not so much at the
> rows estimate. I agree that the row estimate wasn't too bad. But the
> cost estimate seems way out of line.
The cost estimates are not even in the same units as the actual runtime.
Cost is in an arbitrary scale in which 1 unit = 1 sequential disk block
fetch. It is unknown what this might equate to on your machine ... but
it's quite unlikely that it's 1.0 millisecond. The thing to look at
when considering EXPLAIN results is whether the ratios of different cost
estimates are proportional to the actual runtimes.
regards, tom lane