I generally like the point system, and enjoy playing with it.  However, one thing does bother me, and that is how many more points you lose if a lower ranked player wins in a multi player game.  So, If I lose a 6 man game, with a range of opponent rankings, and a veteren player (lets say 2k points) wins the game, I lose 25 points (2.5k/2k*20).  And if a new player (1k points) wins the game, I lose 50 points (2.5k/2k*20).  However, the skill of the winning player (much less their ranking) does not have all that much to do with why I lost in a 6 player game.  I could have been fighting that vet from the start while the average guy was building up on the other side of the board.

Anyhow, I don't have a better answer.. just something that was bothering me.

BD

If a 1k kind of guy wins a six player game with all those heavy hitters, he deserves a [self-censored]-load of points, no matter how he did it, and I say he should get most of 'em from you.

You do have a point, but consider the alternative. What if the payout to the winner was the mean ranking of all opponents divided by the winner's ranking times 20, times the number of players? (Proceeds to be taken evenly from all losing players.) The obvious downside to this would be that the lowest ranking losing player would take a substantial beating.

Interestingly enough, the dynamics of how you describe the disparity suggests that it may be in the best interests of the better players to play differently in such situations, but I'm not sure what that might mean.

For instance, should they take it a little easier on each other and take out the competition, even in the absence of a formal agreement, knowing that at least the little guy won't be taking mega-points to the bank? ..or perhaps the opposite, expend just a little more of their resources attacking each other knowing that if they are able to neutralize the better players, they have a good chance of taking out the little guys to win?

I know I like to take a look at who I'm dealing with when I make my moves and I make decisions based on that knowledge. I just couldn’t tell you which way I tend to play.

M57 wrote:

 ..or perhaps the opposite, expend just a little more of their resources attacking each other knowing that if they are able to neutralize the better players, they have a good chance of taking out the little guys to win?

This is what I do. I try to attack Dud every turn. But I think the other way is smarter in the long term.

And how is G rating calculated? Seems to correlate with your win/loss ratio vs your expected win/loss ratio?

G rating should be the total sum of players in games you've won (including yourself) divided by the total number of games you've played. At least that is the definition I know of. So, this mean a G-rating of a one means that you win/lose pretty much as is expected. Above 1 is above average, below 1 is below average, but how much of a measure this is is hard to tell.

BlackDog wrote:

I generally like the point system, and enjoy playing with it.  However, one thing does bother me, and that is how many more points you lose if a lower ranked player wins in a multi player game.  So, If I lose a 6 man game, with a range of opponent rankings, and a veteren player (lets say 2k points) wins the game, I lose 25 points (2.5k/2k*20).  And if a new player (1k points) wins the game, I lose 50 points (2.5k/2k*20).  However, the skill of the winning player (much less their ranking) does not have all that much to do with why I lost in a 6 player game.  I could have been fighting that vet from the start while the average guy was building up on the other side of the board.

Anyhow, I don't have a better answer.. just something that was bothering me.

BD

This just came up for me last week where I was playing in a game that whittled down to three players.  One of them had a score in the 600's and one in the 1900's range.  So, with my high score, I was very worried that if the 600 won, I would lose more points than the 1900.  Thus, when it came down to trying to eliminate one opponent, I went for the low score so that even if I failed, he would not win and my points were somewhat protected.  Honestly, this was the first game where I have paid any attention to points after joining, but it made me think that they could change how the top players play.

Alpha wrote:

This just came up for me last week where I was playing in a game that whittled down to three players.  One of them had a score in the 600's and one in the 1900's range.  So, with my high score, I was very worried that if the 600 won, I would lose more points than the 1900.  Thus, when it came down to trying to eliminate one opponent, I went for the low score so that even if I failed, he would not win and my points were somewhat protected.  Honestly, this was the first game where I have paid any attention to points after joining, but it made me think that they could change how the top players play.

M57 wrote:

Alpha wrote:

This just came up for me last week where I was playing in a game that whittled down to three players.  One of them had a score in the 600's and one in the 1900's range.  So, with my high score, I was very worried that if the 600 won, I would lose more points than the 1900.  Thus, when it came down to trying to eliminate one opponent, I went for the low score so that even if I failed, he would not win and my points were somewhat protected.  Honestly, this was the first game where I have paid any attention to points after joining, but it made me think that they could change how the top players play.

The other obvious component of the decision involves taking into account the actual or implied playing strength of the players.

I was reading your post Alpha, and it occured to me that this is an area of strategy where a meaningful Aggressiveness rating would have value to players.

For instance if (and especially in a fogged game) you knew that that the 600 player had a tendency to attack and spread out his forces thinly (I would call this aggressive play), certainly it would impact on your decision, right?

We've talked about and Aggressiveness rating on other threads, but the group consensus is that it seems depending on how you calculate it, aggressiveness tends to be too subjective an attribute or too closely aligned with playing strength.

Agreed, aggressiveness would be a useful stat, but is a very hard thing to measure numerically.  I have thought about it a bit and have discuss it with a math friend, but it seems that anything we come up with would only be useful on a per game basis and is still heavily dependent upon luck.  Once you take an aggregate of games, the luck should mute how aggressive a player really is because without a strong position, it is hard for a player to be aggressive.  Of course, you could come up with who is kamikaze aggressive as they tend to repeated spread out on their first turn and over attack.

I think number of offensive dice rolls vs defensive dice rolls would be an interesting statistic.. perhaps it has something to do with aggresiveness?

BlackDog wrote: I think number of offensive dice rolls vs defensive dice rolls would be an interesting statistic.. perhaps it has something to do with aggresiveness?

I doubt it.  It's all dependent on context.  For instance, if a player rolls more 3v1's per capita than the average player, does that make them more or less aggressive?

More total 3v1 rolls, more aggressive. Higher proportion of 3v1 rolls among total rolls, less aggressive.

asm wrote: More total 3v1 rolls, more aggressive. Higher proportion of 3v1 rolls among total rolls, less aggressive.

What if player A (in a fogged game) sits around accumulating a large number of armies on a small number of territories, attacking just enough to get a card each turn.  During this time all hell breaks loose on the board as most players battle it out, and over time because they have more territories, they congregate a bunch on the borders but only one on each interior territory (I think you'd agree this is a pretty standard strategy).  Player A waits until most players have beaten each other to a pulp, and though they have more territories, Player A has amassed enough armies to take them out one at a time, with a preponderance of 3v1 kills.

I wouldn't call this an aggressive strategy, yet it might look like one using this type of statistic.

I thought you were referring to a longer-term data set. Over the course of a single game I doubt there's any way to pull any meaningful conclusions out of this data.

Right. If this player used this strategy day in and day out, wouldn't it look like they play aggressively using this method, when in fact they don't?

Alpha wrote: G rating should be the total sum of players in games you've won (including yourself) divided by the total number of games you've played. At least that is the definition I know of. So, this mean a G-rating of a one means that you win/lose pretty much as is expected. Above 1 is above average, below 1 is below average, but how much of a measure this is is hard to tell.

Where is the definition of the G-rating???

Thx!

(=

G Rating

G rating is a normalized rating based on how many games you expect to win on a game of a particular size. A G rating of > 1 means you win more games than average, < 1 is below average. For example, in a 2 player game you would be expected to 50% of the time (a G rating of 1), a 3 player game 33% (a G rating of 1), a 4 player game 25% (a G rating of 1), and so on. A player's G rating score will be used as a tie breaker when more than one player has the same Ranking score.

Added it to the Rankings help page too ( http://www.wargear.net/help/display/Rankings )

Thanks Yertle, I was wondering that myself.

Thx Yertle! (=

This just occured to me: A low G rating combined with a high ranking means that I am generally playing higher ranked players, a high G rating combined with a high ranking means that I am playing lower ranked players.

It can also have to do with the size of the games you play. If you play nothing but 1v1s and win 80% of the time, your ranking will be astronomical but your G will still be quite low.