Warrior General Games

I played with the League Projector Tool from this thread: http://goallineblitz.com/game/forum_thread.pl?thread_id=470138 a little bit last season and have it setup and ready to go for projections for this season. Since there is no data to have projections for week 1 I can start posting them starting week 2. Although according to the creator of the tool it doesn't really start getting accurate until it has 5 weeks worth of data in it.

So would people like me to start posting its projections starting week 2? 5? 6? Lemme know what people think for it.

yea sounds good the roughnecks would love to see it, and when ever you want to start

Well, I'm allready doing predictions for the Eastern Conferece, so if you take the Western, it'd be cool

Originally posted by Bolts-Fan
Well, I'm allready doing predictions for the Eastern Conferece, so if you take the Western, it'd be cool

Are you doing them with the tool or just your opinion? Cuz these projections will be based on the tool that I linked and are statistical calculations and will be actual score projections.

Atm I use my own opinion, as the download link didn't work when I tried it... but I'm gonna try again

Works for me and I love it. It is pretty neat.

Anyone still want the Power Rankings (also statistical calculations that increase in accuracy as the season progresses)? I put in a couple of minor tweaks after last season, so it is ready to go if people want them.

Originally posted by iguazu_cn
Anyone still want the Power Rankings (also statistical calculations that increase in accuracy as the season progresses)? I put in a couple of minor tweaks after last season, so it is ready to go if people want them.

power rankings are always fun. they give us a general feel of where teams are at for the most part.

Hell yeah, you did great with that last season.

Throw up the projector! looks cool

Ok, so I know I said that I might throw up the projections starting for week 2, but I just finished putting week 1's game into it and all the projections just look disgusting since there is no cross data yet. It is projecting scores like 799-953 for example, although that is the extreme end. I'll post some results when I get some more data for it. So look for it in a couple weeks most likely.

So the projections finally look somewhat normal. Obviously they will get more and more accurate as time goes on. So without any more delay, here are the week 4 projections:

Western Conference
Alberta Express vs South Jersey Rotobowl Rice Riders
19 - 55

AU Tigers vs Ft. Richardson Rangers
20 - 39

Ceará Yorkshires vs YellowKnife Bandits
21 - 13 (inaccurate based on recent changes not reflected in these stats)

Chilliwack Thrashers vs Thunder Bay Bloodhounds
9 - 65

Hill City Heat vs New Jersey Hood
8 - 49

Moose Jaw Blazers vs West Coast Trojans
32 - 23

Oh Canada Ehs vs Saskatoon Spitfires
16 - 7

Prince George Bulls vs The Anti-Mod Squad
13 - 17


Eastern Conference
Aspermount Roughnecks vs Hudson Bay Trappers
11 - 39

Canadian Bacon vs Halifax Fire
31 - 5

District Challengers vs Paradise Cheeseburgers
14 - 49

Edmonton Moose vs Old Milwuakee Boob Punch
27 - 8

Fort Myers Mustangs vs Willow Creek Phantoms
5 - 70

Lumberjack City Yaks vs Medicine Hat Witch Doctors
23 - 14

Moose Jaw Mice vs Vancouver Valients
16 - 290 (Only game that seems extremely off for the score. I will look over the sheet again to check for any typos)

Prince Edward Island Sharks vs Sherbrook Colonels
57 - 19


So there you have it, Week 4 projections are done.

edit: Played with some numbers to see where they were coming from. Since Vancouver beat Hudson soundly, and Hudson has since beat other teams soundly, that is creating the drastic score. Again, things will get more accurate with more data.

So you are having the same problems as me with the Valients team. ok. i thought I entered something wrong. The formula must be typed in wrong somewhere there.

Originally posted by fixxxer101
So you are having the same problems as me with the Valients team. ok. i thought I entered something wrong. The formula must be typed in wrong somewhere there.

It basically comes down to the fact that there isn't enough cross data yet. Give it a few more games and things should balance out. Although you may very well be correct, there might be a formula wrong somewhere. Time will tell.

Im not seeing the mistake with the Valiants score, looks pretty good to me.