Warrior General Games

Originally posted by matta
Projected Standings (remember, this thing seems to favor teams with weaker schedules, the bias decreases over time):

[U
14 . . The State of Texas Highwaymen . . ..2 . . 14 . . 0 . . ..13% . . .....16 . . 5,329
15 . . South Texas Skull Crushers . . ..0 . . 16 . . 0 . . ....0% . . .......0 . . 2,294
16 . . Michigan Rage . . . . . . . . . . . . . . . ..0 . . 16 . . 0 . . ....0% . . .......0 . . 2,935


One thing I've always liked about these spread sheets, is the fact that they almost always pick 2 or more teams in the same conference to go winless, even though they play each other, at some point.

yall really think usmc will score that many on southside?

Originally posted by shimmer
yall really think usmc will score that many on southside?

Being that Rapid City and Chicago only scored in the 50s to 60s, nah. It will probably be in the same range.

Originally posted by Len67

One thing I've always liked about these spread sheets, is the fact that they almost always pick 2 or more teams in the same conference to go winless, even though they play each other, at some point.

Yeah, it's a fluke, but that happens when a game is projected to go 0-0. It counts that as a loss for both teams. When it projects a game to tie at anything higher than 0-0 (e.g. 7-7), it could that as a win for both teams. So sometimes you see multiple 16-0 teams.

As it gets more data, the probability of a tie gets less and less.

Originally posted by matta
Originally posted by Len67


One thing I've always liked about these spread sheets, is the fact that they almost always pick 2 or more teams in the same conference to go winless, even though they play each other, at some point.

Yeah, it's a fluke, but that happens when a game is projected to go 0-0. It counts that as a loss for both teams. When it projects a game to tie at anything higher than 0-0 (e.g. 7-7), it could that as a win for both teams. So sometimes you see multiple 16-0 teams.

As it gets more data, the probability of a tie gets less and less.



Thanks, I've always wondered about that. So, I get it now, if the teams have yet to score a point, and there is more than 1 of them, they will get a loss, by this system, when they play each other. Works the exact opposite when the are unbeaten.
And no I am not being sarcastic, it's that I never this. Thanks again.

Originally posted by shimmer
yall really think usmc will score that many on southside?

Not gonna happen. Our second game is skewing the results. It's going to jack up the league awards also. I don't think anyone can catch our return man. His average might go down, but he's got such a huge lead in TD's he should be a lock.

Us getting State of TX as our East matchup isn't going to help.

See you in Lubbock Nov 1.

Originally posted by Len67
Originally posted by matta

Originally posted by Len67



One thing I've always liked about these spread sheets, is the fact that they almost always pick 2 or more teams in the same conference to go winless, even though they play each other, at some point.

Yeah, it's a fluke, but that happens when a game is projected to go 0-0. It counts that as a loss for both teams. When it projects a game to tie at anything higher than 0-0 (e.g. 7-7), it could that as a win for both teams. So sometimes you see multiple 16-0 teams.

As it gets more data, the probability of a tie gets less and less.



Thanks, I've always wondered about that. So, I get it now, if the teams have yet to score a point, and there is more than 1 of them, they will get a loss, by this system, when they play each other. Works the exact opposite when the are unbeaten.
And no I am not being sarcastic, it's that I never this. Thanks again.


Right. So SoTex and Michigan have yet to score a point (well, at least when those scores were entered), so the spreadsheet assumes neither team is capable of scoring, and therefore their game will go 0-0. You can't win with 0 pts, so both teams lose.

The other case, where there are multiple 16-0 teams usually occurs only in week 1 where there are two 1-0 teams that won by the exact same margin. Since there's only one datapoint, the spreadsheet treats a 21-18 win over NBSK the same as a 21-18 win over Michigan. The next week, when NBSK wins by 100 and Michigan loses by 100, the spreadsheet figures out that NBSK is a better team than Michigan and adjusts the predictions.

Originally posted by matta
Scores:

Vegas (62) over Memphis (21)

Lots of gutted teams in the East...

I think the spreadsheet fliped the names :-) j/k Good Game LV

Nice game Atlanta. Good luck the rest of the way.

The spreadsheet nearly pegged the score.

Good game Southside. You had a couple long drives we just couldn't stop. I'll be keeping my eye on you next season.

Nice game Pittsburg. You guys got a good offense. Keep up the hard work.