In order to understand the benefit of Chrome preferring HTTPs over HTTP for omnibox navigations, we need some more specific metrics.

We a) need to understand what % of navigations are omnibox navigations that wind up HTTP and b) need to understand what % of those navigations could have been converted to HTTPS if we had a whitelist of sites to convert.

We can accomplish (b) by using RAPPOR to note the most common HTTP sites and then compare those to a whitelist offline to determine the percentage of navigations that could have been converted.

Alternatively we could package a highly compressed and pruned whitelist to get a more exact result.