You can do some crude approximations by looking at reviews and ratings. (Reviewing or rating is only possible if you downloaded the app from the store.) Those are obviously very problematic metrics so take this with a grain of salt.

Unretina was downloaded 3,283 times in the US App Store, it was rated 16 times and reviewed 8 times. 0.5% and 0.2% of all downloaders rated and reviewed it, respectively.

Here are some top paid apps which received a similar amount of ratings and reviews (excluding Apple’s apps):

1. Compartments ($10, 17 ratings, 7 reviews, 4 stars, #16 in Top Paid)

2. Contacts Cleaner ($5, 15 ratings, 8 reviews, 4 stars, #18 in Top Paid)

3. Courier ($5, 19 ratings, 7 reviews, 5 stars, #20 in Top Paid)

(…)

All those apps are bunched together pretty closely (all are between #16 and #70, none have any lower spot on the Top Paid list; the number of ratings and reviews generally tapers off on the list). That certainly supports my hypothesis that there is a relationship between ratings, reviews and downloads. I would dare to speculate that all those apps sold a few thousand copies on the first day.

You have to take into account, though, that those are the successful apps. You probably could expect to sell a few thousand copies of your $10 app only if you made it into the top 70 or so. (Unretina is similarly one of the more successful free apps. It has the #39 spot on the Top Free list.)

TUAW puts the number of available apps in the UK App Store at about 1,000 (http://www.tuaw.com/2011/01/06/mac-app-store-by-the-numbers-...), if we assume that number to apply in the US App Store, too, you would have to have been in the top 7% of all apps to sell a few thousand copies on launch day.