Idea for new scoring system

from Stardock Forums
I have been thinking a lot about the scoring system in GalCiv, and it occurs to me that there is a system which is both simpler and better measures skill. I would have scoring work according to:

SCORE = [ VICTORY * (SIZE / TURNS)] * BASE,

where SCORE is the score for the game, SIZE is dependant upon the size of the galaxy for that game, VICTORY is dependant upon victory type, TURNS is dependant upon the number of turns it took to win, and BASE is a constant like 1000 or something so that scores look all high and cool. That's because I would have SIZE, VICTORY, and TURNS each average to 1 so that they basically become multipliers to the 1000.

The best way to make the variables each average to 1 would obviously be to use metaverse data. Either you don't get a metaverse score until you submit, or each update to the game includes the latest figures. As long as the metaverse keeps track of those 3 variables for each game, which shouldn't be hard, it can recalculate everything each time the metaverse data changes too.

The idea is to make it so that (a) there is no advantage to playing a smaller or bigger galaxy (since in theory the size only affects game length, not difficulty), (b) there is no advantage to getting a particular victory type (unlike now, where you always want a military victory for max points) and (c) score is based purely on how fast you win, not things like population. Using population and other game stats just encourages running games longer than they could take to win, which does not require skill. Skill is defined by one thing: how fast you can win.

Also, I would like to see a score board where only each player's HIGHEST score is listed, rather than the total of his scores. Maybe each player could have one score listed for each victory type, or something. Not that I expect to be first (I cannot beat masochistic to save my life), but I think a person should be able to shoot from last to first on the score board in one game if he is that good. Otherwise the score board basically measures how dedicated you are to playing (which is the only reason I'm in the top 10, lol).

Any thoughts?

-Drake



~SDC~
127 views 3 replies
Reply #1 from Stardock Forums Top
Umm, you forgot game difficulty. The present scoring system already discounts difficulty enough, no need to discount it further. To me, the lamest win possible on Maso should score higher than the best win possible on Normal, or even tough. Right now, that definitely is not the case.

But I agree somewhat with the idea of decreasing the importance of LOTS of games running up your score. Then maybe non-college-students-on-summer-break will have a shot as well =)
Reply #2 from Stardock Forums Top
Oops. Obviously difficulty should play in, yes. Missed that.

I would lean toward averaging it toward 1 also. The idea is that WHATEVER your settings, your score is computed relative to average scores at those settings, with 1000 being average performance and the score going up or down from there depending on how you did compared to others. In that case you could outscore a masochistic game on a lower difficulty level, at least in theory. I can see where you would want to weight it toward the higher difficulty levels to encourage people to move up, too, though.

-Drake

~SDC~
Galactic Civilizations III
Reply #3 from Stardock Forums Top
I should add that in order to perform that hypothetical outscoring of the masochistic player under my scoring system while playing at a lower difficulty level oneself, one would have to do REALLY well, since obviously the averages for that lower difficulty level will be higher. So the player who can barely beat the game on masochistic would get about the same score playing at crippling, because he would totally thrash the computer at that level.

-Drake

~SDC~