are you sure Draug? I just looked through some of my early games for a 19+ win, and its a bit ambiguous - this is a 21 center win in a 119 point pot - 18/34 should be 63 , 21/34 should be 73,5 , but the winner got 70 - equivalent to exactly 20/34 of the pot...

and then the link...
gameID=38074

This was changed and I think that games end date predates the change. But I do remember Kestas saying he was doing that to prevent just that kind of point hoarding.

okay - havent played much ppsc lately.

the rules specifically say you can't get more that half.

I think there can be some oddities with this, especially when someone is CD. I'm not sure how exactly it works, but the points for non-active centers somehow get spread amongst the active players.

I can think of, but not link, one good example of someone getting the entire pot when playing PPSC. He managed to push his neighbors out of their homes so only he had home centers and then he rounded up each enemy unit and crushed it, leaving only his units on the board. All this at 17 or less centers. Then with 17 units he moved to where in one year (two moves) he could step into the other 17 centers and claim a complete victory, in which case he did get all the points. Was quite beautiful. I imagine someone has a link to that somewhere.

^I would like to see that link if anyone can find it.

lets find out http://webdiplomacy.net/board.php?gameID=71304

I am not sure exactly how it works but you can also get more points if there are CD's, I took this game over for 105 , won 152. Couldn't pass up the opportunity when I saw it in open games. The only thing I'm not sure about is if the extra points are solely from CD's or also because I took 24 centers.
http://www.webdiplomacy.net/board.php?gameID=69627#gamePanel

Well, for that game if you take your 18 valid centers plus the other 7 active centers, that's 25. Divide the pot, 210 , by 25 to get 8.4 per center. Your 18 centers would get you 151.2 , and you actually got 152. Rounds up I guess? Other player had 7 centers at 8.4 per center is 58.8 rounded up to 59. Those numbers seem to work out. Other times I've tried this same thing and not gotten numbers working out or maybe I just did it differently and figured it out this time. Not sure.

The point to take from all of this is simply:
Don't play PPSC. WTA is more pure to the game and ends up with less people giving the game away so they can get second place points.

The tricky one is when someone has say 20 centers. They get paid for 18, but the others get paid for their centers, so either the system has to remove the extra centers from the equation of points per, or there is lost points in those 2 centers.

I would think the pot would be divided by the number of valid centers, where a valid center is the first 18 of the victor and any non-CD'd centers. So if you take 20 centers, then the post is divided by 32 instead of 34. But that does increase the points you get for winning more than 18 centers since you get 18/32ths instead of 18/34ths.

So possibly better would be to remove CD centers first and the winner gets 18/X. In your scenario the winner would get 18/34ths of the pot. The remainder of the pot is then divided by the remaining active centers, so the 16/34ths portion is divided by 14 as the value per center for the survivors in your scenario. The winner doesn't get more and the survivors are splitting the same amount, just at slightly different ratios depending on who lost the extra centers to the winner.

But I haven't looked at the code so don't know how it is really currently being done.