I think this is a significant sample because:
- almost all games have screenshots or short debriefs (except for Ace who usually links to Dedcon recordings, and some of the old hosting links which have expired)
- it has a large number of games (111 so far, although when you break that down into 15 pairings, the numbers are individually still low)
- all the games are random territories so there's no selection bias
- all the games are played at quite a high standard, so there's less noise from poor tactics (although good play still clearly makes a difference)
Anyway, compare and contrast the following with my earlier back-of-the-envelopethread.
Summary:
Asia: Strong (26-14)
Africa: Strong (26-14)
SA: Neutral/weak (14-17)
EU: Neutral/weak (17-21)
NA: Neutral/weak (16-21)
Russia: Weak (12-24)
Code: Select all
vs NA SA EU Africa Rus Asia
NA . 3-4 6-7 2-4 2-3 3-3
SA 4-3 . 2-5 1-3 3-1 4-5
EU 7-6 5-2 . 1-5 2-3 2-5
Africa 4-2 3-1 5-1 . 9-5 5-5
Rus 3-2 1-3 3-2 5-9 . 0-8
Asia 3-3 5-4 5-2 5-5 8-0 .
So this just confirms what we already knew - the only whitewash pairing is Asia-Russia, although Africa-EU is pretty one-sided, but not impossible (I remember seeing other EU victories that didn't make it in to this sample). Africa-Russia also gives a statistical advantage, but one that can be overcome by good play. And SA-NA is pretty even - grr those last few million kills White Rabbit!
I'll keep on updating the spreadsheet occasionally and we'll see how the numbers change - although this is about 3 months worth of ladder matches so it won't change that fast.
Comments, suggestions anyone?
Cheers, Tripper







