Various questions

[Finalgenesis]

Originally Posted by thejeff

There was a long debate about how sites were generated a while back. The conclusion was that the 4 checks were separate. The other mechanism just didn't match the results.

Finalgenesis's numbers look right for the theoretical chance based on rolling until a check fails.
Here's what I've found in one EA game: (Site searching mostly complete)
31 total provinces
0 sites: 4 - 13%
1 site: 8 - 26%
2 sites: 12 - 39%
3 sites: 5 - 16%
4 sites: 2 - 6%

I haven't calculated the chances for separate rolls or figured in terrain, but that just doesn't match the roll until you fail mechanic.

Thanks for the date (Still working on mine), in EA the spawn is 45% I think, but you got 0 sites only about 13% of the time, whereas the "roll 1 by 1 until failure" mechanics would mean that 55% provinces should have 0 sites? Your data would fit a "roll 4 dices" mechanics better where the chances are:

assume 45% spawn:
0 sites - 9% (55%^4)