Stat junkies rejoice, here is my run down:

1) Anonymity would be a must.

2) A dual ranking system. First column, instead of rank by score, score is simply an expected finish number. To use an 11 person FFA as an example your ranking at best would be 1.0 (first every time) and at worst 11.0 (last every time). This would make the rankings really tight and always make every elimination order have value.

2b) Ranking by expected finish would make it hurt your score to surrender or simply suicide on an opponent.

3) Second column could be a kill to death ratio. In other words how many opponents have you eliminated in contrast to how many times you have been eliminated.

Overall rank could be a hybrid score between finishing value and players eliminated. This would balance out strategies where players align with everyone just to try and finish in the top 5 to the ones where someone aggressively goes for the win but comes up short because the remaining players align against him. It would also help dissuade games from becoming never-ending stalemates due to players not wanting to risk becoming eliminated.

Other stats that could be tracked on a stat page could include:
- Players favorite starting spot (assuming its always played on the same map)
- Players avg. income on turn 10, 25...
- Average turn for there first elimination kill.
- Average length they survived for.
- Biggest stack ever held.
- Most armies on board at one time.
- Fastest turn for victory
- Number of wins (a no brainer)
- many more

Your thought?

Of course, a FFA ladder would never happen until anonymous games are an option. Vote for it at:
http://warlight.uservoice.com/forums/77051-warlight-features

Your expected finish number is something that can unfortunately be skewed rather easily. When someone realizes that they have no hope of losing, they simply ball together all their armies and head for a non-desirable location and stack each turn. The other players that still have a chance to win have zero motivation to waste armies on killing the turtling player off, thus it becomes a legitimate strategy to boost up that particular rating/ranking.

Also, eliminations/kills would only be able to track the player who actually delivered the deathblow, and could thus allow one player to do 95%+ of the work on a player only to have another player come in from another side and record the actual elimination/kill (I have a variant I'll create one day that utilizes this mechanism).

I think that I've seen ELO systems for 3+ team games before, but I can't recall where right now.

Very good points Troll. I think kill/death ratio would still be worth looking at but perhaps just for a fun stat. I would be curious to know how the ELO existing rankings work.

Maybe the elimination could go to whoever did the highest percentage of damage?

A for ranking, I guess the top spots could be heavily favored. 100 pts for a victory... 85 for second.... 75 for third and then a drop off... 4th = 35.... 5 pts for 10th and 0 for 11th.

Quick knockouts would likely work out as a wash for everyone in the end. In addition, the turtle strategy could get someone as high as 3rd but I doubt that would happen very often. Still, it would be nice to reward bold playing and cautious playing somehow.

Some of the things that I think will be important for an FFA ladder: multiple game sizes (variable number of opponents), multiple maps, anonymity. I agree with Troll that expected finish is too easy to manipulate.

Make the game point based where you get points for certain actions instead of a ranking against other players:

1) What place you get - 100, 85, 75...etc
2) Elim/Kills - 25 total points

To make sure the Elimination/Kills work out, we have the graphs of how many armies each player has killed of each other player. Just put two weight factors; Who's killed the most armies of that player and then the Deathblow. So say there are 25 points to give out for both factors. The person that killed the most armies would get 20 and 5 points to the Deathblow (I think doing most of the work should be rewarded more).

This also would promote killing people to the end because you are shooting for army kill.

3) First person to 20, 30, 50, 75 armies a turn
4) Points for lasting for 10 turns, 15 turns, 20 turns, etc..

There can be a couple more but I can't think of them.

To balance it out we could have negative points for actions as well like getting eliminated and other things.

Once the game is finished, tally up all the points for everyone and then rank everyone according to the game points. From there you could give a standard Rating change for each place. This could spawn possibilities of the 2nd place person in the game actually winning if they did all the work against the other players while someone turtles and then goes for the last kill.

I think these things would make for a really exciting game where everyone is trying their hardest and not giving up.

Another thought is to have Mission Cards dealt out for extra points. Some examples from the actual Risk game are:

capture Europe, Australia and one other continent
capture Europe, South America and one other continent
capture North America and Africa
capture Asia and South America
capture North America and Australia

I'm sure with all the different maps there could be different set ups for the missions.

If games were not Anonymous then you could add a hunt variable into the mix and get extra points for knocking a certain player out. If they were anonymous, the card could say, try and knock out the player with the color of _____.

I had to respond quickly as I liked your post Ruthless. I overlooked and under-exploited how the detailed stats page could easily be tied into determining the percentage of each player elimination.

If this is a case maybe each elimination is worth 25 total points. Thus every player gets points based on total percentage of damage they do to eliminated player "X"

Breakdown for player "X" being eliminated:

Player one dealt 22% damage = 5.5. pts
player two dealt 40% damage = 10 pts
player 3 dealt 13% damage = 3.25 pts
player 4 dealt 25% damage = 6.25 pts

In addition, with an exception of extra points for bonuses hit, points should be finite per game. Therefor, I don't like the idea of extra points per number of turns survived since this would reward slow play. In other words, the same number of total points up for grabs in an 11 person FFA that lasted 30 turns should equal the same number of point given out to one lasted 300 turns.

I agree with you there BP. I think the point I was trying to make with Lasting 10, 20, 30,...was that at a certain point that would probably stop. I'm guessing an arbitrary number would be chosen like 50 or 75. But for people to try and survive that long so that they don't just give up.

read the post not the comments: if youre not first your last in my opinion. maybe second should get credit but past that no, otherwise when people realize theyre out theyll just try to hang around as long as possible and not attack anyone. also kills is worthless, player A can destroy player B but if player C kills off the last territory why should he get the credit.

after reading the comments i agree with troll precisely

if you read mine and BP's reply, you would have seen that we came up with a solution to that problem

everyone would come out about the same