K-State Football...

[Skycat]

While I don't have any data, I always feel pretty queasy when we get a 1st and goal outside the 5.  First and goal from the 8 feels like an automatic FG try for our offense.

I doubt that is unique to KSU.

Most teams %'s will go down in that situation, I would imagine.

I'm sure you are right.  But my guess is that better running team's %'s dip less than others.

[sys]

Kougs, the r squared for that data is .4583.

That is not in the neighborhood of being statistically significant.

the r square doesn't speak to significance.  rather to the strength of the relationship.


the data look fairly convincingly anti-rusty, though the sample size really is too small to do much with.

[michigancat]

Kougs, the r squared for that data is .4583.

That is not in the neighborhood of being statistically significant.

the r square doesn't speak to significance.  rather to the strength of the relationship.

It's a real weak relationship.

[kougar24]

Kougs, the r squared for that data is .4583.

That is not in the neighborhood of being statistically significant.

the r square doesn't speak to significance.  rather to the strength of the relationship.

It's a real weak relationship.

Yeah, it is, but there is definitely an upward trend to the data. You can't deny that.

So, you gonna buy that software? 

[mjrod]

Kougs, the r squared for that data is .4583.

That is not in the neighborhood of being statistically significant.

the r square doesn't speak to significance.  rather to the strength of the relationship.

It's a real weak relationship.

That's because the sample is too small.

[GoldbrickGangBoss]

Correlation between rushing y/c and turnovers.

[michigancat]

Yeah, it is, but there is definitely an upward trend to the data. You can't deny that.

Yeah, but yards per play has a very similar trend.  Shockingly, teams with better offenses overall do better in the red zone than crappy offenses.  The correlation (which is very slight) between rushing and red zone efficiency does not explain causitation.

Kougs, the r squared for that data is .4583.

That is not in the neighborhood of being statistically significant.

the r square doesn't speak to significance.  rather to the strength of the relationship.

It's a real weak relationship.

That's because the sample is too small.

No, like I said above, you have 275 red zone attempts and 2400 carries in that data.  It's enough.

[michigancat]

Here is the correlation data:

Code: [Select]

TEAM TD% YPC YPP YPA
1) Oklahoma 0.83 5.40 5.50 7.60
2) Texas 0.78 5.20 6.60 8.30
3) Missouri 0.70 4.00 6.20 8.30
4) Texas Tech 0.66 2.80 6.90 8.30
5) Kansas 0.65 4.80 5.60 6.70
6) Nebraska 0.62 3.90 5.60 7.10
7) Oklahoma State 0.57 5.40 6.70 9.10
8) Texas A&M 0.57 4.70 5.20 6.00
9) Colorado 0.52 3.80 4.90 6.20
10) Kansas State 0.52 4.20 5.70 6.90
11) Iowa State 0.45 2.90 4.20 5.60
12) Baylor 0.35 2.10 4.40 5.60
correlation 0.673 0.636 0.665
Notice that yards per attempt and yards per carry are virtually identical.

[michigancat]

If you add the SEC, the correlation gets weaker:

YPC:  0.55
YPP:  0.52
YPA:  0.51

Code: [Select]

TEAM TD% YPC YPP YPA
Oklahoma 0.83 5.40 5.50 7.60
Texas 0.78 5.20 6.60 8.30
Missouri 0.70 4.00 6.20 8.30
Texas Tech 0.66 2.80 6.90 8.30
Kansas 0.65 4.80 5.60 6.70
Nebraska 0.62 3.90 5.60 7.10
Oklahoma State 0.57 5.40 6.70 9.10
Texas A&M 0.57 4.70 5.20 6.00
Colorado 0.52 3.80 4.90 6.20
Kansas State 0.52 4.20 5.70 6.90
Iowa State 0.45 2.90 4.20 5.60
Baylor 0.35 2.10 4.40 5.60
Georgia 0.71 4.60 5.60 6.90
LSU 0.64 4.80 5.90 7.40
Tennessee 0.68 4.80 5.80 6.80
Mississippi St. 0.71 3.70 4.40 5.40
Kentucky 0.70 4.70 6.00 7.10
Alabama 0.61 4.00 5.10 6.30
Arkansas 0.63 6.40 6.60 6.80
Florida 0.75 5.00 7.00 9.90
South Carolina 0.59 3.60 5.30 6.80
Auburn 0.61 3.80 5.00 7.00
Vanderbilt 0.47 3.90 4.80 6.10
Ole Miss 0.36 3.90 5.40 6.80
correlation 0.55 0.52 0.51


[michigancat]

Adding the ACC, here are the correlations:

YPC: 0.43
YPP: 0.45
YPA: 0.44

Keeps getting worse for bookie.

(up to 36 teams and almost 1000 red zone opportunities for those scoring at home.)

[catzacker]

So if there's no correlation what does this tell us, exactly?

[michigancat]

So if there's no correlation what does this tell us, exactly?

There is correlation between TD% in the red zone and YPC.

The problem is, the correlations between TD% and both YPA and YPP are virtually identical to the YPC correlation.

What it tells you, is that better offenses are better at scoring touchdowns in the red zone than bad offenses, regardless of their strengths, and people that say "we need to establish the run to score more TD's" are full of sh*t.

[sys]

you can't compare ypplay and ypcarry because ypc is included in the ypp data.  they aren't independent.

personally, i would prefer regressions to correlations.  you have a hypothesized causitive relationship.  also your sample size is the number of teams.  that is the unit of data you are analyizing  (you are not, f.ex. analyizing if a run of 8 yards resulted in a score more often than a run of 3 yards, your data are the averages).  with the sec and big 12 you are close to a good sample.  add another conference and it should be fine.

to keep it simple, i would just run simple linear regressions on ypc and yppass if you wanted to see whether one is more important than the other on red zone scoring.  if you want to look at all factors that impact red zone scoring, maybe also do turnover rate and penalty number.  prob. in a multiple regression framework.

[michigancat]

you can't compare ypplay and ypcarry because ypc is included in the ypp data.  they aren't independent.

personally, i would prefer regressions to correlations.  you have a hypothesized causitive relationship.  also your sample size is the number of teams.  that is the unit of data you are analyizing  (you are not, f.ex. analyizing if a run of 8 yards resulted in a score more often than a run of 3 yards, your data are the averages).  with the sec and big 12 you are close to a good sample.  add another conference and it should be fine.

to keep it simple, i would just run simple linear regressions on ypc and yppass if you wanted to see whether one is more important than the other on red zone scoring.  if you want to look at all factors that impact red zone scoring, maybe also do turnover rate and penalty number.  prob. in a multiple regression framework.

As in, r-squared?  Refresh my stats, plz.

[cireksu]

I have no stats to back this up but...

i think our scoring problems in the redzone are a result of our pass first offense.  We run the ball pretty well between the 20's as indicated by ypc.  Our passing game sets up the run well when teams have to play both safety's 15-20 yds off the ball.

This changes in the redzone however and teams don't have to respect the deep ball.  essentially the offense is playing in a smaller box.  teams man up on our below avg recievers and te's, double team jordy, and can leave the lb's and safety's close. 

We just don't have the oline, fb for a "power running game" inside the 20.  I think if RP can recruit more physical linemen that can really push the dline around this would change.  All goes back to recruiting.

[sys]

regression gives you an r2 as well as a p value.  (also an equation, the slope of which might be interesting to this question).  regression assumes a causitive relationship.

correlation gives you an r and a p value.  it evaluates any relationship, not assuming one variable influences values of the other.  

[michigancat]

Here it is w/ the Pac 10, Big XII, ACC, and SEC.

[sys]

good stuff.  both are pretty weak.  looks like luck has a big impact.

[kougar24]

This has turned into a remedial course in statistics.

Fascinating. (Srsly)

[catzacker]

So in summation, good offenses find a way (whether by pass or rush) to score.  A team's offensive "balance" doesn't necessarily matter, it's the manner at which they utilize their strengths. 

[mjrod]

OK, I see now.

I haven't done any statistics in a while, and now I understand what's going on.

Correlation is extremely weak.   Fortunately, most football coaches get glossy eyed when it comes to statistics.  Next time a coach says we need a running game to help in the redzone, I'll just snicker.

[Bookie Pimp]

Adding the ACC, here are the correlations:

YPC: 0.43
YPP: 0.45
YPA: 0.44

Keeps getting worse for bookie.

(up to 36 teams and almost 1000 red zone opportunities for those scoring at home.)

Actually, what "keeps getting worse" is your FBIQ....

Tell us, oh great sage, what happens to YPC stats when a team punches it in from the 3 yd line or closer.  Is that better, or worse, for their YPC stats in the redzone?  Additionally, what you fail to factor in to your little study is the fact that any QB sacks that occur within the 20 yard line will actually impact a teams YPC stats.

No where did I say that the ability to score touchdowns was directly related to YPC.  What I said was, the ability to score touchdowns is related to a teams ability to run the football.  With few exceptions to the rule, this will remain a constant across all levels of the game for each of our foreseeable lifetimes.

Please help us all understand why having a strong running game is a bad thing.  That was the original question, wasn't it?

Let me spell this out for you in a way that should be easy for all to understand.... Statistics equations class be damned.  In order to run the football with any consistency whatsoever, a team would need to have an offensive line that can consistently control the line of scrimmage.  Assuming a team has an offensive line that can control the line of scrimmage, then passing also becomes much easier simply due to the fact that the QB will have more time to survey the coverage and not be running for his life.

I understand that you likely have not had any football gear on your person since junior high or so, nor have you ever taught the game.  So, I'll give you a pass for not understanding the nuances of lining eleven players up on each side of the ball with differing game circumstances (read: field position, score, gameclock, down & distance, etc.) when considering "statistics" inside the redzone.  I would like to congratulate you, however, on your ability to plot basic statistical equations on a graph though.

 

[Houstoncat93]

Somebody run yards per carry versus winning percentage.  That is really what people who say we need a running game are saying.  A good running game = more wins.  

[kougar24]

Please help us all understand why having a strong running game is a bad thing.  That was the original question, wasn't it?

No. That leap of logic isn't founded based on Rusty's argument.

A more accurate take on Rusty's viewpoint would be to say that the ability to score is not dependent on the ability to run the ball.

[michigancat]

Tell us, oh great sage, what happens to YPC stats when a team punches it in from the 3 yd line or closer.  Is that better, or worse, for their YPC stats in the redzone?  Additionally, what you fail to factor in to your little study is the fact that any QB sacks that occur within the 20 yard line will actually impact a teams YPC stats.

This is all YPC, so a three yard TD run (or ten three yard TD runs, for that matter) would have little impact over the course of a season and sacks would affect all teams.

No where did I say that the ability to score touchdowns was directly related to YPC.  What I said was, the ability to score touchdowns is related to a teams ability to run the football.  With few exceptions to the rule, this will remain a constant across all levels of the game for each of our foreseeable lifetimes.

Is YPC not directly related to "ability to run the football"?  Do you have a better way of quantifying "ability to run the football"? 

Please help us all understand why having a strong running game is a bad thing.  That was the original question, wasn't it?

No one said having a strong running game is a bad thing.  There are many ways to have success on offense.  The original question was whether or not KSU's running game was keeping KSU from scoring points.  Evidence suggests that it is not.

I understand that you likely have not had any football gear on your person since junior high or so, nor have you ever taught the game.  So, I'll give you a pass for not understanding the nuances of lining eleven players up on each side of the ball with differing game circumstances (read: field position, score, gameclock, down & distance, etc.) when considering "statistics" inside the redzone.  I would like to congratulate you, however, on your ability to plot basic statistical equations on a graph though.

LOL, you have nothing to suggest that KSU's running game is keeping KSU from scoring points, other than "I coach jr. high football and understand nuances".

[cireksu]

How about, what's the overall record for teams that average 180+ yds/game vs teams that avg, less than let's say 110 yds per game.

[cireksu]

if you look at the stats basically all the bcs contenders run the ball for over 180 yds/game or are in the top 25 team rushing category.


if you look at the bottom 25 team rushing category LOL.



only about 3 teams in the bottom 25 average over 30 points/game, Tech, Hawaii, and Bowling green.


there are about 5 teams in the top 25 team rushing that average under 30 points /game.




does that help anyone's argument?

[catzacker]

Usually, a good team has a good offensive line and that good offensive line can run and pass block.  There probably isn't a correlation between scoring and just someone's ability to run or just someone's ability to pass.  Look at KSU's offense in 2001, we could run on anyone, but we couldn't score (averaged 29ppg).  

[Bookie Pimp]

LOL, you have nothing to suggest that KSU's running game is keeping KSU from scoring points, other than "I coach jr. high football and understand nuances".

Getting stuffed at the one yard line kept KSU from scoring seven points against a PITIFUL Iowa State defense just after averaging less than 2 YPC on three attempts from inside the ISU six yard line.

 

[catzacker]

LOL, you have nothing to suggest that KSU's running game is keeping KSU from scoring points, other than "I coach jr. high football and understand nuances".

Getting stuffed at the one yard line kept KSU from scoring seven points against a PITIFUL Iowa State defense just after averaging less than 2 YPC on three attempts from inside the ISU six yard line.

 

Sincerely,
OSU getting stuffed twice inside the 5 by a terrible KSU rush defense.