Add aggregate statistics covering all players at a certain level

Re: my comments (Eric) here about small sample sizes, a reasonable thing to do when you have little or no data on a player is to look at the aggregate data that you've collected on all players to get a sense how the overall player pool plays. As to how to do this, there are a lot of ways, some of which I brought up in that blog post, but primarily, I'd be interested to be able to bring up histograms of the distributions of various stats in PCP (not necessarily in the HUD) both weighted by hands played and by head count, along with the important summary stats about the distribution.

For instance, if we were looking at VPIP we'd calculate the VPIP for each player in the database at a given level, and then build up two histograms: first, a head count weighted graph, where we simply increment the count in each VPIP bucket by one for each player that has a VPIP in that range, and second, a hand weighted one, where we increment the count in each bucket by the number of hands that each player w/VPIP in range played. The head count version would be useful to see because once a hand database is large, unknowns will mostly be newcomers at the level and the head count stats arguably represent their play distribution better (if you have no data on someone, chances are they aren't a regular), whereas when your database is small, you still don't have many hands on each of the regs, so a hand weighted distribution is a better guide.

Particularly for stats where it's very difficult to get enough data from an individual player to have any certainty (fold to 3bet, for instance), aggregate data would be hugely useful, since the overall probabilities are very different depending on the stake you're playing, and these stats can make the difference between a +EV and -EV play.