VERY interesting stat on W-L Correlation

[Go Gophers Rah]

I've spent the morning trying to find one single stat of opposing teams that correlates to our Win-Loss record. I kept coming close, but would find that AFA or SDSU wouldn't fit any pattern - that might be because their stats are skewed by their non-BCS opponents.

So, including only BCS-league teams, here is the only correlation that I could find:

National Rank - Rushing Offense (game result)
16) Wisconsin (loss)
22) Ohio State (loss)
25) Illinois (loss)
35) Cal (loss)
41) Penn State (loss)
67) Michigan State (win)
73) Purdue (win)
89) Northwestern (win)
95) Syracuse (win)
98) Iowa (???????)

Now I'm not saying that this stat is a predictor that we'll win or anything, I just find it to be interesting.

Go Gophers!

 

[gophmeister]

Now check how well their defenses are against the run.

 

[KoolAid]

Are you using 2009 W-L as your Y-variable? That's going to be too small a sample size to draw specific conclusions.

Over the years, the best predictor of W-L% is Pass yards per attempt. With the exception of the MSU game, I'm going to guess Minnesota's Y/A is pretty bad, hence the mediocre W/L%.

Run another regression with this stat and see what picture it paints.

 

[dpodoll68]

Not bad...at least it's better than your ground-breaking research indicating that teams play worse on the road.

 

[Go Gophers Rah]

dpodoll... you never cease to be an unpleasant human being.

 

[MBAGuy]

Not bad...at least it's better than your ground-breaking research indicating that teams play worse on the road.

dpodoll68: Believing he's better than you since 1978.

 

[Schnoodler]

That is interesting GGR.

 

[howeda7]

I've spent the morning trying to find one single stat of opposing teams that correlates to our Win-Loss record. I kept coming close, but would find that AFA or SDSU wouldn't fit any pattern - that might be because their stats are skewed by their non-BCS opponents.

So, including only BCS-league teams, here is the only correlation that I could find:

National Rank - Rushing Offense (game result)
16) Wisconsin (loss)
22) Ohio State (loss)
25) Illinois (loss)
35) Cal (loss)
41) Penn State (loss)
67) Michigan State (win)
73) Purdue (win)
89) Northwestern (win)
95) Syracuse (win)
98) Iowa (???????)

Now I'm not saying that this stat is a predictor that we'll win or anything, I just find it to be interesting.

Go Gophers!

AFU would never fit this pattern because they run SO much that they're going to rank highly even if they're not efficient. I think you're on to the fact that the teams that could ram it down our throats gave us the hardest time, but that doesn't mean I suddenly think we can beat Iowa.

 

[dpodoll68]

dpodoll68: Believing he's better than you since 1978.

Actually, I've only been alive since 1980 and didn't figure out that I was better than everyone until around 1985 or so.

And I was paying him a compliment. This actually had some meaningful information, unlike earlier contributions mentioned above.

Even at that, I do question the validity of a study that includes one year of data and eliminates 18.2% of that data.