Warrior General Games

http://spreadsheets.google.com/ccc?key=p9N6eCChbl0ICJyVvSpSfUg&hl=en

Game Predictions:
Again, I think it's important to remind you that I have nothing to do with these predictions aside from inputing data and posting it here

Barger Hill-6
Dubrovnik - 32

Hungarian - 0
Gen. Patton - 202

New York - 11
NFL - 119

Bisons - 8
Gliwice - 2

Sheboygan - 27
St. Petersburg - 10

Washington - 6
Sarajevo - 37

Georgia - 24
Kazan - 14

Minsk - 38
Team Phoenix - 4

Perfect!

Google spreadsheet is so awesome. I can chat with you guys now.

Man that Bisons/Gilwice game looks fucking intense! Game of the week indeed.

This is awesome man! Keep it up. However I am giving some backups starting roles this game for some much needed xp...if we lose I will shit myself.

Originally posted by psychonaut
This is awesome man! Keep it up. However I am giving some backups starting roles this game for some much needed xp...if we lose I will shit myself.

Is that what you did the last game?

Great job Houston!

I hope the BHB/Dubrovnik game doesn't go down like that...

I'd like to see most (if not all) of these games be closer than they are predicted to be. Last week, some of the scores were VERY close, some weren't...but the basic end results were all accurate.

Originally posted by Gott
Originally posted by psychonaut

This is awesome man! Keep it up. However I am giving some backups starting roles this game for some much needed xp...if we lose I will shit myself.

Is that what you did the last game?

rofl...pretty much knew I was going to lose...The Burglars have a great team.

Originally posted by HoustonMark
I'd like to see most (if not all) of these games be closer than they are predicted to be. Last week, some of the scores were VERY close, some weren't...but the basic end results were all accurate.

I think the model has problems on the extreme ends. It projects really high scores when good teams play really bad teams. Other than that, it seems to be pretty solid.

Yeah, this model seems to be OK for predicting who should beat who, but it's not accurate on the margins they're likely to win/lose by at all. For instance, Hungary's played 3 teams with better records than the Demon Brigade. In those 3 games, they lost by 79, 95, and 72. Yet the model has them projected to lose to the Demon Brigade by 202. That's nowhere near accurate.