SportsHoopla Top 25 week 6 voting

[24seven]

Yea, I can see what you mean. But I can't see what other statistical category I could use besides defensive and offensive production.

If you're a team that is winning without producing on offense or defense, then you are probably a very lucky team. And this program is going to figure that out. If you're a team that is winning because of offense, then you're gonna be ranked well because you have a top-tier offense, and this program will figure that out. Vice versa with defense.

Idk, it's the best I can possibly do with determining which statistics should be used. As you can see, my team is ranked #25, so I wasn't biased with which categories I should base it on.

They should be #26, homer.

 

[Jack_John_Mark]

They should be #26, homer.

Your team is ranked high enough dammit.....ye should keep your mouth shut

 

[Jack_John_Mark]

Did special teams factor in? Those aren't really luck. If your team starts from the 40 every drive, and someone else starts from the 15..... you don't need as much offense. Likewise, if your opponents are starting at the 15 and someone else's opponents are starting at the 40, you don't need to hold teams to as few yards.

Turnovers? Those are sometimes about luck, but not usually. And they make a bigger difference than almost anything else.

No those things are not factored in.........but good grief no team is going to consistently start at the damn 40 yard line.

I know you're pissed because your team is undefeated and my little computerized ranking system has them behind some 1-loss teams........but if they keep winning I guarantee you they will move up. There is no possible way at the end of the season that my program will have an undefeated GT team outside of the top 3.

 

[CalvinJohnson4God]

No those things are not factored in.........but good grief no team is going to consistently start at the damn 40 yard line.

I know you're pissed because your team is undefeated and my little computerized ranking system has them behind some 1-loss teams........but if they keep winning I guarantee you they will move up. There is no possible way at the end of the season that my program will have an undefeated GT team outside of the top 3.

LOL

It has nothing to do with where GT is. I actually like computer ranking systems. I just disagree that they are without bias. It's just a different kind of bias.

And yes, the example given would obviously not be typical. But the point is that you left out a large portion of a team's stats. (Georgia Tech's special teams happen to suck, by the way). The fact that you don't consider special teams a part of the game obviously shows some bias.

Just as you don't think that one of the most crucial parts of the game, protecting the ball, is important.

 

[Jack_John_Mark]

Dammit.........I spent like 3 hours compiling all the offensive and defensive stats for all 120 teams. I have to type all that shit out by hand

If I had someone help me to compile those other stats I would definitely put those in as well. I mean, Nebraska is obviously in the top 5 in the country in special teams........so again, I didn't mean to be biased.

The question is, how much do I allow special teams and turnovers to weigh in?? Straight up offensive and defensive production should outweigh special teams and turnovers by quite a bit............though I do agree a turnover or kickoff return, punt block, etc can completely flip a game.

 

[CalvinJohnson4God]

Dammit.........I spent like 3 hours compiling all the offensive and defensive stats for all 120 teams. I have to type all that shit out by hand

If I had someone help me to compile those other stats I would definitely put those in as well. I mean, Nebraska is obviously in the top 5 in the country in special teams........so again, I didn't mean to be biased.

The question is, how much do I allow special teams and turnovers to weigh in?? Straight up offensive and defensive production should outweigh special teams and turnovers by quite a bit............though I do agree a turnover or kickoff return, punt block, etc can completely flip a game.

I don't know..... and how do you decide how to weigh each stat? That's why there is an inherent bias. I wasn't ripping your rankings.... they look pretty decent. I just disagree about the bias.

Are you good with Excel at all? I use that when I mess around with stats. You can go to the NCAA's stat site... copy an entire chart and paste it into excel. If you're trying to use several different stats.... after pasting, just sort them all by team name (alphabetically), and then you can do whatever you want with it.

 

[KansasSooner]

Dammit.........I spent like 3 hours compiling all the offensive and defensive stats for all 120 teams. I have to type all that shit out by hand

If I had someone help me to compile those other stats I would definitely put those in as well. I mean, Nebraska is obviously in the top 5 in the country in special teams........so again, I didn't mean to be biased.

The question is, how much do I allow special teams and turnovers to weigh in?? Straight up offensive and defensive production should outweigh special teams and turnovers by quite a bit............though I do agree a turnover or kickoff return, punt block, etc can completely flip a game.

Why? Go to the NCAA stat site and they are in spreadsheet form. I just copy and paste into a spreadsheet from there.

I weight this way: offense*% of offensive plays + defense*% of defensive plays + special teams*% of special teams plays. It works out to about 47% for offense and defense and about 6% for special teams.

 

[KansasSooner]

Every computer ranking carries bias. Not toward certain teams based on name and preconceived opinions, maybe... but you decided which stats you felt were most important and how they should be weighted.

I do like it better than rankings where team bias is so prevalent... but bias is impossible to avoid.

I'll guarantee you my multiple regression analysis is the most unbiased of any model out there. The whole purpose of using the stepwise procedure to select the variables which most highly correlate to winning % is to select a model free of any bias as far as possible. There are 15 offensive categories, 14 defensive categories and 8 special teams/other categories that the NCAA recognizes as being legit data. Any bias in the model is easily seen in the residuals if one knows how to look for it. This allows further investigation into the data to see if the bias can be removed. However to date, I have not seen any bias in the model's results.

 

[BoiseStateFan27]

JJM, sir

if you're able to without it causing extra time, I'd suggest you add in turnovers and Red zone efficiency on both offense and defense to your rankings. Both would make everything much more accurate

because honestly unless you're grasping for straws on something no one cares if you put up a ton of offensive yards if you can't avoid turning it over or can't convert anything in the redzone.

there's my two cents, sir

it's already pretty good though.

 

[Jack_John_Mark]

I don't know..... and how do you decide how to weigh each stat? That's why there is an inherent bias. I wasn't ripping your rankings.... they look pretty decent. I just disagree about the bias.

Are you good with Excel at all? I use that when I mess around with stats. You can go to the NCAA's stat site... copy an entire chart and paste it into excel. If you're trying to use several different stats.... after pasting, just sort them all by team name (alphabetically), and then you can do whatever you want with it.

Yea I thought about using Excel, the problem is I reformatted my PC a few months ago and now I don't have Microsoft Office anymore. It's funny how my computer came with Microsoft Office on it, but they didn't ship me a CD to reinstall it if I need to.

So then I tried to use GoogleDocs spreadsheet..........it just doesn't work. It pastes all of the stats into one column which won't work.

 

[CalvinJohnson4God]

Yea I thought about using Excel, the problem is I reformatted my PC a few months ago and now I don't have Microsoft Office anymore. It's funny how my computer came with Microsoft Office on it, but they didn't ship me a CD to reinstall it if I need to.

So then I tried to use GoogleDocs spreadsheet..........it just doesn't work. It pastes all of the stats into one column which won't work.

I will keep you in my prayers

 

[Jack_John_Mark]

I will keep you in my prayers

Much appreciated

 

[johnshadows64]

I guess I'm eligible to vote:

1. Alabama
2. Oklahoma
3. Wisconsin
4. LSU
5. Oklahoma State
6. Stanford
7. Clemson
8. Boise State
9. Oregon
10. Arkansas
11. Georgia Tech
12. Michigan
13. West Virginia
14. Illinois
15. South Carolina
16. Kansas State
17. Arizona State
18. Nebraska
19. Baylor
20. Virginia Tech
21. Texas
22. Texas A&M
23. Georgia
24. Michigan St.
25. Houston

 

[ACC_HOKIES]

To go along with JJM's computer ranking, I've got one brewing as well. Based on wins, margin, sos, and a small quality wins bonus. I've also ignored all FBS>FCS results (FBS<FCS count against).

1 Louisiana State
2 Alabama
3 Oklahoma
4 Wisconsin
5 Stanford
6 Boise State
7 Oklahoma State
8 Michigan
9 Clemson
10 Illinois
11 Georgia Tech
12 Houston
13 Kansas State
14 Oregon
15 South Carolina
16 Arkansas
17 Arizona State
18 West Virginia
19 Texas
20 Baylor
21 Nebraska
22 North Carolina
23 Penn State
24 Rutgers
25 Virginia Tech

 

[KansasSooner]

Yea I thought about using Excel, the problem is I reformatted my PC a few months ago and now I don't have Microsoft Office anymore. It's funny how my computer came with Microsoft Office on it, but they didn't ship me a CD to reinstall it if I need to.

So then I tried to use GoogleDocs spreadsheet..........it just doesn't work. It pastes all of the stats into one column which won't work.

Just go here: OpenOffice.org - The Free and Open Productivity Suite

Fully compatible with Microsoft Office, Lotus, and WordPerfect Office. And IBM just gave them the entire Lotus libraries...

 

[KansasSooner]

JJM, sir

if you're able to without it causing extra time, I'd suggest you add in turnovers and Red zone efficiency on both offense and defense to your rankings. Both would make everything much more accurate

because honestly unless you're grasping for straws on something no one cares if you put up a ton of offensive yards if you can't avoid turning it over or can't convert anything in the redzone.

there's my two cents, sir

it's already pretty good though.

The NCAA site lists red zone offense and defense categories for that. Turnovers are broken down into total, fumbles and interceptions, too.

 

[Jack_John_Mark]

Alright dudes..........I am going to work on figuring in all those categories for next week. The easiest thing for me to do might honestly be to create a program in c++ which reads input from txt files. That way I can just copy and paste the tables right off the page into a txt file and save......then compile and run the program.

 

[24seven]

Alright dudes..........I am going to work on figuring in all those categories for next week. The easiest thing for me to do might honestly be to create a program in c++ which reads input from txt files. That way I can just copy and paste the tables right off the page into a txt file and save......then compile and run the program.

I'll then take yours and switch all the odd numbered rankings with the even numbered ones next to them and BAM, I'm done!

 

[rolltide14_0]

Based on common sense.

1. LSU
2. Alabama
3. Oklahoma
4. Wisconsin
5. Oklahoma state
6. Boise st.
7. Stanford
8. Clemson
9. Oregon
10. Arkansas
11. Georgia Tech
12. Michigan
13. West Virginia
14. South Carolina
15. Illinois
16. Virginia Tech
17. Nebraska
18. Kansas state
19. Baylor
20. Arizona state
21. Texas A&M
22. Michigan st.
23. Georgia
24. Houston
25. Notre Dame