I find all the math kinda interesting. I see how one gets 45.1% for 12-0 by multiplying all the probabilities together, but how do they get the other percentages?
Sent from my SM-T230NU using Tapatalk
Printable View
I find all the math kinda interesting. I see how one gets 45.1% for 12-0 by multiplying all the probabilities together, but how do they get the other percentages?
Sent from my SM-T230NU using Tapatalk
Since the probability of winning any game is the percentage given, the probability of losing is 100 minus that percentage. Therefore, to get, for example, the probability of going 11-1, you need to add up all the combinations of multiplied percentages that include exactly one loss from remaining games.Quote:
Originally Posted by td577
Audit can correct me if I'm wrong but I worked it out for the 11-1, 8-4, and 7-5 scenarios and nailed 40.1%, 0.1%, and 0.002%. Would take a little longer for 10-2 & 9-3 since there are quite a few more combinations and I'm just using my phone calculator.
you got it dude.Quote:
Originally Posted by Strategery
Audit, how would you figure this one out?Quote:
Originally Posted by westnodak93bison
A methodology that is relatively straight forward is discussed in Mathletics, but getting the data and the computing power to do it right is an issue.Quote:
Originally Posted by westnodak93bison
If you had all the scores for all the games you could compute an offensive and defensive rating for each team and understand how much the model's predictions vary from reality which would then give you an idea of how likely a particular defense would be to pitch a shutout against a particular offense.
In college football you've got like 250 teams x 2 variables per team (off & def) + 1 for home field + whatever else you want to throw in, so 501 variables or more. Excel's solver can't handle that many variables. It can do a single conference or even the whole NFL (65 variables) but not 500+.
If we just use the MVFC conference games played to date (which is only 3 or 4 games per team so, sample size issues!, plus you're willfully ignoring a bunch of information you know is relevant!) such a model suggests that NDSU's defense is 13 points better than the conference average while USDs offense is 9 points poorer than the conference average. Those are top and bottom of the table respectively. The predicted game score: NDSU 39 USD 4 (which is way more than the 25 point margin suggested by Sagarin even though the conference only model's home field advantage is almost x2 Sagarin's). Comparing the model's predictions to actual results to date, the standard deviation of predicted points allowed compared to actual is 5.58 and the errors are normally distributed around the mean so we can use the normal distribution curve to estimate roughly a 20% chance of a shut out on Saturday.
Again, that's just based on performance in the 3 or 4 conference games played to date and it's completely ignorant of everything except the final scores and who's field the game was played on. But's that's a way you could get at it.
Excel solver can't handle it! OMG. What will we do.
Use Cognalysis!Quote:
Originally Posted by NDB2
Slide RuleQuote:
Originally Posted by NDB2
I'm surprised.Quote:
Originally Posted by NDB2
What if I had two copies of Excel running at the same time? Could I grip it by the husk?Quote:
Originally Posted by NDB2
I think I have a copy of VisiCalc laying around here somewhere. Will that help?Quote:
Originally Posted by NDB2
Maybe this can help.
http://sploid.gizmodo.com/genius-app...one-1649161239
Caution! If you point your phone at the saragin ratings with this app, it might explode.
56 is trying to open a file with it right now.Quote:
Originally Posted by gumby013
Better make sure he has the key to the filing cabinet.Quote:
Originally Posted by bisonaudit
You mean I need a key?????? I am doomed!!!Quote:
Originally Posted by Tatanka
I just pulled some interesting stats from Massey's archives for a thread on AGS and figured I'd share them here too.
I got this bit from CAS here a couple years ago: Interesting tidbit that could likely be verified if you spoke to either individual, but Erk Russell was reported to have had a conversation with Rocky Hager (then Bison coach) sometime in the late 80s or early 90s in which he said that he felt the DII Bison would have given those championship Georgia Southern teams of that era everything they could handle.
Readily admitting the limitations of computer models -- especially across divisions -- this is still interesting: Massey concurs with Erk Russell, indicating that NDSU held a slight edge in both '86 and '90, which were championship years for both programs in their respective divisions.
Per Massey archives, here's a decade's worth of NDSU's odds of beating Georgia Southern on a neutral field at the end of each respective year:
1983 - 93% (NDSU vs DIAA Champ SIU - 25%)
1984 - 73% (NDSU vs DIAA Champ MT St - 34%)
1985 - 43%
1986 - 57%
1987 - 32% (NDSU vs DIAA Champ NE LA - 14%)
1988 - 35% (NDSU vs DIAA Champ Furman - 31%)
1989 - 6%
1990 - 53%
1991 - 15% (NDSU vs DIAA Champ Youngstown - 21%)
1992 - 65% (NDSU vs DIAA Champ Marshall - 33%)
Edited to add: Hard to say with any degree of certainty, but I think those teams from the 80s would have been competitive at the next level up. Especially with another 28 scholarships or whatever the difference was.
When is NDSU going to lose again? 2016? Undefeated is looking pretty good right especially since we are playing @ 80% on offense and still blowing teams away. IF NDSU fixes their problems on offense, no team will come within 17 points of them in their next 8 games.
Who says they have to lose again?Quote:
Originally Posted by Strongman
Sent from my SPH-L720 using Tapatalk
Two records are a long, long way away. Longest DI winning streak is 47(Oklahoma 1953-57) and longest DI unbeaten streak is 63(Washington 1907-17*).Quote:
Originally Posted by td577
*Can you imagine going a decade without losing a game? (59-0-4) Impressive no matter the era.
Those are impressive streaks. Both of them.Quote:
Originally Posted by Hammersmith
Sent from my SPH-L720 using Tapatalk
47 really isnt that far away at this point. Especially since no FBS next year (Not that those have been a problem)Quote:
Originally Posted by Hammersmith
It is and it isn't. We're 68% of the way there. But it's still a full season away with the playoffs in the middle. How many teams in the playoff era have been able to go undefeated through a full season? And how many very good teams couldn't quite make it and ended up with one loss?Quote:
Originally Posted by SamsRams
If there is any program that can do it, it is this one
Well with Wentz a senior next year, it might not be until 2016. Seriously.Quote:
Originally Posted by td577
Updated rankings are out, up to 34.
One spot ahead of Marshall, who is also 8-0. The wacky world we live in.
Remember when that UNI game looked really scary? Not so much anymore. (on paper) Even more so with SDSU. In the top-25, Harvard jumped and McNeese took a tumble. Montana out, Idaho State in(who would have thought that a few years ago?). In the conference standings, the SoCon jumped into 2nd and pushed everyone down a slot. Big South stays in, Big Sky stays out. And the Valley is still nipping at the heels of the AAC & MWC.
Sagarin's Week 10 Predictions
Sagarin FCS Top-25Code:Sagarin Predict Result Diff
34 North Dakota State 78.70
61 Iowa State 70.85 +4.57 +20.00 +15.43
211 Weber State 41.32 +34.10 +17.00 -17.10
235 Incarnate Word 28.85 +53.13 +58.00 +4.87
141 Montana 54.98 +27.00 +12.00 -15.00
148 Western Illinois 53.53 +21.89 +7.00 -14.89
112 Southern Illinois 60.45 +21.53 +28.00 +6.47
96 Indiana State 63.14 +18.84 +17.00 -1.84
173 South Dakota 48.01 +27.41 +40.00 +12.59
83 South Dakota State 65.67 +16.31
82 Northern Iowa 65.84 +9.58
103 Missouri State 61.31 +14.11
98 Youngstown State 62.55 +19.43
53 Illinois State 72.29
Total Diff: -9.47
Home Field 3.28
Sagarin Top-5 FCS ConferencesCode:1 34 North Dakota State
2 53 Illinois State
3 75 Jacksonville State
4 76 New Hampshire
5 79 Villanova
6 82 Northern Iowa
7 83 South Dakota State
8 90 Coastal Carolina
9 93 Harvard
10 96 Indiana State
11 98 Youngstown State
12 100 Eastern Washington
13 101 Richmond
14 102 SE Louisiana
15 103 Missouri State
16 105 Chattanooga
17 108 Sam Houston State
18 110 McNeese State
19 112 Southern Illinois
20 115 Liberty
21 117 Fordham
22 121 Samford
23 124 Idaho State
24 125 Eastern Illinois
25 127 Montana State
Code:1. MVFC 63.12
2. Southern 50.13
3. Colonial 49.93
4. Big South 49.83
5. Southland 49.79
FBS Comparisons
Non-P5 Top 10 ConferencesCode:#1 in the AAC, C-USA, MAC, MWC, & Sun Belt
#5 in the Big Ten
#6 in the ACC & Big 12
#8 in the PAC-12
#12 in the SEC
Code:1. AAC 63.90
2. MWC 63.78
3. MVFC 63.12
4. C-USA 60.19
5. MAC 57.36
6. SBC 55.11
7. SoCon 50.13
8. CAA 49.93
9. Big South 49.83
10. SLC 49.79
Has our conference rating basically topped out?? I know it'll change a little bit from week to week based on how are teams do compared to others. But considering we are all in conference play it feels like there is nothing our teams can do to improve. Is the only way we can improve as a conference is for our BCS wins (Iowa St and Ball St, etc) to actually win games??
It just seems strange that UNI is still so highly rated.
There losses have been by 1 3 3 and 8 and the 2 worst ones were on the road to FBS teams.Quote:
Originally Posted by No_Skill
Yeah, I know the reasoning behind it. Maybe "strange" wasn't the right word.Quote:
Originally Posted by bisonaudit
Can you imagine being a uni fan right now? That has to be tough.
I feel for the fans. Farley is another matter.
Because of the way Sagarin chooses to compute conference rankings, there is a way for the Valley to improve. Sagarin uses central mean to compute conference ratings instead of simple average. With simple average, every Valley team would be worth the same, 1/10 of the total score each. But central mean favors the middle of the conference pack at the expense of the outliers. In the Valley, the 1 & 10 teams are worth 1/30 each. 2 & 9 are worth 2/30, 3 & 8 are worth 3/30, 4 & 7 are worth 4/30, and 5 & 6 are worth 5/30. So the middle two teams combined are worth 1/3 of the conference total, and the middle four teams combined are worth 3/5.Quote:
Originally Posted by stevdock
What's that mean? Well, if the middle four teams(currently SDSU/YSU/ISUb/MSU) were to play NDSU & ISUr very close and absolutely pound WIU & USD, then the ratings of the middle of our conference would go up while the top and bottom of the conference would go down. But since central mean favors the middle, the up effect would be greater than the down, and the conference rating would rise.
This is one reason Sagarin hates the Big Sky. While NDSU's impact on the Valley's rating is only 1/30, EWU's impact on the Big Sky's is an even tinier 1/49. The mid-pack Big Sky teams aren't that good, so it drags the conference rating down.
Now is a big shift for the Valley likely to happen? Nope. But it could because of central mean.
Based on Sagarin's RATING numbers (3.28 home field) and a 2.61 pt kicker for coming off a bye week, here's the outlook for the remainder of the regular season:
South Dakota State 88.6% (+4.8% from last week)
Northern Iowa 76.1% (+1.5%)
Missouri State 85.2% (+4.6%)
Youngstown State 92.4% (-0.4%)
12-0 53.0% (+7.9% from last week)
11-1 37.1% (-3.0%)
10-2 8.9% (-4.0%)
9-3 0.9% (-0.9%)
8-4 0.03% (-0.1%)
Assumes a normal distribution of outcomes w/ Sagarin's projected spread as the mean and a standard deviation of 13.86. In round numbers this means that if the spread is 14 the model predicts that roughly 2/3 of the outcomes will be between a tie and a 28 point victory. Half of the remaining outcomes or about 1/6 of all the outcomes would be a victory by more than 28 and there would be roughly a 1 in 6 chance that the team favored by 14 would lose.
It would be interesting to know, based upon previous history, how accurate is Sagarin at predicting the winners/losers of the last 4 games? I would think that their models become pretty accurate about this time of the season. The +7.9% change from last week, assuming the Bison go 12-0, seems like a big jump. I'll take it!Quote:
Originally Posted by bisonaudit
UNI had the 7th and 8th highest Sagarin rating in 2013 and 2012 and didn't make the playoffs either year. Too many FBS losses and close losses in conference play.Quote:
Originally Posted by No_Skill
For the first three weeks of November last season the standard deviation between Sagarin's predicted margins and the actual outcomes were 15.12, 16.30 and 13.20. My model uses a standard deviation assumption of 13.86.Quote:
Originally Posted by KilldeerBison
It would suck to be a UNI fan for sure. Have a team that consistently is good for many years, but never good enough to win it all.
Invite some up and commer who says they will dominate you within five years and that team then goes on to threepeat.
That'd fucking suck.
Sent from a TI-83+ Calculator
Dude.
They haven't been good enough to make the playoffs.
What? Why not 14.01?Quote:
Originally Posted by bisonaudit
How much ratings value does NDSU add? I.E., what would be MVC rating without NDSU? Conversely, what would each non Big 5 conference rating be if NDSU was in each conference? I would be very interested to see how much NDSU would inflate the MAC and Sun Belt's rankings.