I wanted to mention how the ratings are calculated on the Leaderboard. Ratings are calculated using a modified form of the ELO rating system. In a nutshell, it works as follows: After a game, your rating is adjusted based on whether you won or lost (up if you won, down if you lost), by an amount that is inversely proportional to the likelihood of your winning. Basically if you weren't expected to win and you do win, your rating will go up more than if you were expected to win and you did win. The expectation of win is based on the ratings of the two players before the game begins. A 1900 rated player for example is expected to beat a 1600 rated player.
Let's take an example. Take two players, (A) rated 1600 and (B) rated 1900. If A beats B, the rating boost for A is much higher, than if A won over another 1600 player. Similarly if A loses against B, the rating drop is lower than if A lost over another 1600 player. This is because the 1900 player is better and is expected to win. Similarly if B wins over A, the rating boost is lower since B was heavily expected to win, than if B played another 1900 player, where it is more of an even match.
The bottom line is this: the fastest way to boost your rating is to win against players with higher ratings than you have. The fastest way to drop your rating is to lose against players with a lower rating than you have. This approach is also applied to teams of players (allies) however in this case, the win is split between players on a team based on the percentage of units killed relative to all kills for the team. Much more detail than you may want is in the Wikipedia ELO page.
One last note: the two problems mentioned in the last post are fixed and have been for awhile now. The Leaderboard is humming along, and new multiplayer missions created in the Mission Editor are visible to the game servers within 20 minutes of being posted to the Mission Sharing forum.