How Nebraska Would Have Done Elsewhere

[StPaulHusker]

This was great work. Can't wait to see the other conference match ups.

[tschu]

I have a request:

 

Can you explain the basics of your statistical analyis?

 

The statistical rankings that I have seen done (by several internet posters on other forums, also stats wizards, along with a sports statistics professional) have revolved mainly around adjusted yards per play rankings. The reason for that being that yards and specifically yards per play are the largest and least variance-ridden component of the statistical description of any football team. Of course there are flaws - special teams are largely unaccounted for, coaching is unaccounted for, and so on. Very very fast offenses (oregon) may be represented statistically differently than very slow offenses (iowa). Still, I am simply wondering. Are you using a YPPbasis or something else?

[It'sNotAFakeID]

I have a request:

 

Can you explain the basics of your statistical analyis?

 

The statistical rankings that I have seen done (by several internet posters on other forums, also stats wizards, along with a sports statistics professional) have revolved mainly around adjusted yards per play rankings. The reason for that being that yards and specifically yards per play are the largest and least variance-ridden component of the statistical description of any football team. Of course there are flaws - special teams are largely unaccounted for, coaching is unaccounted for, and so on. Very very fast offenses (oregon) may be represented statistically differently than very slow offenses (iowa). Still, I am simply wondering. Are you using a YPPbasis or something else?

 

It's 1:40 in the morning so forgive me if this isn't as clear as it would be at 3 in the afternoon.

 

First of all, I'm using three different types of metrics: (1) Offensive--broken down into pass attempts, completions, passing yards, yards per completion (really the same as passing yards), and completions which wind up as touchdowns, rushing attempts, rushing yards, yards per carry (again the same as rushing yards), and rushes which wind up as touchdowns, (2) Defensive--broken down into those same variables with the word 'allowed' at the end of them, and (3) Turnovers--which are total turnovers and total turnovers forced. The next thing that happens is that all of these variables are standardized and Z-scores are produced as a result.

 

After that all of the Z-scores are thrown into what is called a cluster analysis. A cluster analysis takes what is known about the individuals in a sample and groups them together based on similarities it finds. Now there is a little bit of decision making that has to be done by the researcher in terms of how many clusters you want to have. In my case, especially when I did it with the SEC, if I would've chosen 3 clusters there would have been teams who ended up in the almost there group being placed in the worst cluster. I'm going with four unique clusters because it allows me to set up the report nicely. You have the worst group, the group that did better but still didn't live up to it, the teams that were nearly there but not quite, and then the cream of the crop.

 

The last step in the process is to compare the means of the Z-scores for each cluster against each other. This allows you to see not only which teams were similar but why they were similar and what type of team would be the most successful in a specific conference. For example in the SEC, the top teams didn't have the best offenses, the top teams had the best defenses. And for all of Georgia's and South Carolina's success last year, they didn't make elite status because there defenses didn't quite match up against the defenses of Alabama, Florida, and LSU.

 

It also allows you to catch any anomalies; like Texas A&M for example. The Aggies should have not had as much success in the SEC as they did in 2012, and I believe that Manziel really was the difference maker. Had A&M had an average QB, A&M falls down to that 3rd tier of teams.

[EZ-E]

I feel You did a great job here Big 12. But I don't think there's much difference on your list between 4-Us. Those games are toss ups as is AtM's with us.

 

Call it hubris, but for some reason I always felt confident beating AtM as I did Baylor.

[Creighton Duke]

I think we would be above ole miss, but other than that, sounds pretty good to me

 

Statistically, a game between Nebraska and Ole Miss is a toss-up. That sucks to say that, because Ole Miss has been down for...ever, but the way the games were played last year, it would be a toss up. It was also very unexpected that we stuck with Georgia for as long as we did.

I don't see why this gets thrown out there so much. That team was down, like, 6 or 7 starters for that game.

[tschu]

breakdown

Thats makes a lot of sense after googling a bit on cluster analysis. Looks like a fair bit of work too!

 

Including turnovers is an interesting choice too. I think if you're doing a pure predictive power ranking then including turnovers may be misleading because they're one of the most variance-ridden stats. But since you're not doing a predictive power ranking and you're wondering how a team would have done in a past season compared with past data from other teams, including turnovers seems like an obvious choice.

[It'sNotAFakeID]

breakdown

Thats makes a lot of sense after googling a bit on cluster analysis. Looks like a fair bit of work too!

 

Including turnovers is an interesting choice too. I think if you're doing a pure predictive power ranking then including turnovers may be misleading because they're one of the most variance-ridden stats. But since you're not doing a predictive power ranking and you're wondering how a team would have done in a past season compared with past data from other teams, including turnovers seems like an obvious choice.

 

Cluster analysis is a lot of work, but luckily I have a nice statistical software package that does all the heavy lifting. All I really need to know is when it's appropriate to use cluster analysis and what the results mean. For that I thank a certain UNL psychology professor who makes really difficult concepts very easy to understand.

[It'sNotAFakeID]

Today I tackle how Nebraska would've fared in a conference perceived by many to be weaker than the Big Ten, the ACC. Also being addressed in this these posts is how success in the ACC is stylistically different than success in the SEC.

 

The Worst

 

Teams which fall into this category are, on a normal day expected to lose more times than not to the teams which will show up in the top two tiers, but are will win just as much as they lose against teams in the next tier up, as their Z-scores for wins were identical. The reason I'm designating these teams the worst and the other teams just a slight step up is due to the fact that these teams put up their stats against an easier schedule.

 

Offensively: Philosophically, these teams liked to keep their offensive game plan balanced, running and passing the ball a similar amount of times per game. In their execution, they were average in the passing game ranking 3rd in completions, passing yards, yards per completion, and completions to touchdowns. However, failure to develop any sort of ground attack really hurt these teams as they came in dead last in yards per carry and rushes to touchdowns. Combine a one dimensional attack with the highest turnover rating in the conference and you have yourself an offense that wasn't really a threat to opposing defenses.

 

Defensively: There wasn't a whole lot of things to dislike about these defenses. If there teams had to pick one way to attack these defenses it would be through the air, but even that would prove to be a daunting task for most offenses. Overall, the defenses on these teams were slightly above average in the pass defense and way above average in run defense. A couple of things which prevent these defense from being truly spectacular to the point of taking over games: (1) an inability to prevent teams from turning completed passes into touchdowns and (2) failure to force turnovers--these defenses forced the fewest amount of turnovers in the conference.

 

Strength of Schedule: Facing the 3rd easiest schedule in the conference should've allowed at least some of these teams to exceed their statistical expectations, and that could be why some of the teams falling into this category had more conference wins than they should have had.

 

Who Are They? While these teams were above average in the defense department, that defense's inabilities to put an ineffective offense back on the field coupled with that same offense's ability to take themselves off the field with turnovers are what puts these teams into the bottom of the ACC. They are Maryland, NC State, Virginia, and Virginia Tech

[It'sNotAFakeID]

Getting Better

 

It really was arbitrary that these teams wound up under the 'Getting Better' heading instead of the worst. But I felt that because they played a tougher schedule, that they deserved to belong in this category. Any game between these teams and the 'Worst' teams would be a toss-up, and these teams would struggle against the teams in the two tiers above them, losing way more games than they won.

 

Offensively: These were your pass heavy teams, they passed the ball the most in the conference. And they were effective at completing passes, also coming first in the conference in that stat. Where they faltered offensively is their ability to put up big yards and points with the passing game, as they fell well below average (even last in some cases) in yards per completion and completions to touchdowns. These teams shied away from the running game, and for good reason--they weren't good at moving the ball on the ground. They were last in nearly every rushing statistic. But they kept their offenses on the field the most, managing to turn the ball over the fewest times in the conference.

 

Defensively: But man were their defenses awful! Teams could pass or they could run, either way, there wasn't much these defenses could do to stop them, as these teams were well above average (and last in most stats) in every statistical category. However, they did give their offenses more chances to score than the teams in the 'Worst' tier, ranking 2nd in turnovers forced.

 

Strength of Schedule: The year was a tough one for these teams, as there wasn't a game that was easy, and the SOS stat shows it. These teams faced the 2nd toughest schedules in the conference.

 

Who Are They? These teams were like opposites to the teams in the tiers below them. Whereas the teams below couldn't generate an offensive attack of any kind but made it tough for opposing offenses, these teams could at least move the ball through one medium, the passing game. But when it came to stopping the other offenses, it was more likely that a score got the defense off the field than a punt. But with a tougher schedule, I think these teams deserve to be a slight step up from the worst. They are Boston College, Duke, Miami, and Wake Forest

[It'sNotAFakeID]

Nearly There But Not Quite

 

Taking a big step up from the teams below is this next tier of teams, who would routinely beat the teams in the list below, and would lose just a few more games than they would win against the teams above them.

 

Offensively: These teams were run-heavy. The running game was the bread and butter of these offensive attacks and for good reason, they were way better than any other ACC team at doing so, coming well in 1st in nearly every rushing category (the lone exception being converting those rushes into touchdowns, but even that stat is close). These teams didn't go to the pass very much, coming in last in pass attempts, pass completions and pass yards. But when they did go to the passing game, they often reaped a big reward, managing to place first in yards per completion and completions to touchdowns. Holding on to the ball was an issue, as they were turned the ball over more than an average amount of times.

 

Defensively: It wasn't a very good idea to pass against these defenses, as the secondaries on these teams surrendered very few completions. When those completions were given up these secondaries were statistically average, surrendering an average amount of yards per completion and an average amount of completions to touchdowns. The real struggle came in stopping the run. These defenses came in either 3rd or last in every rushing statistic and were decidedly below average in stopping the run. Forcing turnovers was a mixed bag, some games a lot of turnovers would be forced, but then there would be a spell of 2-3 games where these defenses couldn't force the opposing offense into making a mistake.

 

Strength of Schedule: Without a doubt these teams faced the toughest schedules in the conference, and it wouldn't have been surprising to see these teams wind up with fewer conference wins than they did.

 

Who Are They? It's kind of funny, these teams struggled to pass the ball, but were dominant in the secondary; these teams excelled at running the ball, but had problems stopping the run game. In the end, they are better than most teams in the ACC and the elite teams in this conference would be in for a dog fight each time they stepped onto the field with these teams. They are Georgia Tech and Nebraska

[It'sNotAFakeID]

The Cream of the Crop

 

Like what I did with the SEC, I'm not going to do a full break down with these teams, because they are elite for a reason. Offensively, they were above average in all statistics, and they turned the ball over the fewest amount of times. Defensively, they were as a whole the best teams in the conference and were well above average in every statistic. They also managed to force the most turnovers in the conference.

 

But they did this going up against the weakest schedules in the conference. And while they would routinely beat Miami, Boston College, Duke, Wake Forest, NC State, Maryland, Virginia, and Virginia Tech, they would have their struggles and losses to Georgia Tech and Nebraska.

 

Without further ado, the best teams in the ACC in 2012 were Florida State, Clemson, and North Carolina.

[It'sNotAFakeID]

So, if I had to rank the teams in the ACC (including Nebraska): they would be ranked as such:

 

13. Maryland

12. Virginia

11. Virginia Tech

10. NC State

9. Boston College

8. Wake Forest

7. Duke

6. Miami

5. Georgia Tech

4. Nebraska

3. North Carolina

2. Clemson

1. Florida State

 

Remember that there is a distinct gap between teams 1-5 and teams 6-13, but not a gap at all between teams 6-13 and a small gap between teams 1-3 and teams 4 & 5

[The King]

I don't think we are worse than Ole Miss or South Carolina, nor are we worse than North Carolina.

 

Clemson and FSU would be a toss up.

 

I like the work you're putting into these things - I just think sometimes stats don't carry over. For instance, do we average more yards against those teams? Is our defense better against those teams because they pass more? Is it worse? Who knows. Love the analysis though.

[gbr93]

I think we would be above ole miss, but other than that, sounds pretty good to me

 

Statistically, a game between Nebraska and Ole Miss is a toss-up. That sucks to say that, because Ole Miss has been down for...ever, but the way the games were played last year, it would be a toss up. It was also very unexpected that we stuck with Georgia for as long as we did.

I don't see why this gets thrown out there so much. That team was down, like, 6 or 7 starters for that game.

Who were the 6 or 7? The only one I remember was the NT which the back up was hyped to be just as good. We were out atleast one starter as well so in my eyes it was completely even or we may have even had a disadvantage. The only reason I say that is because it's obvious baker was an integral part of our rush D. Again I'm not saying your wrong I just remember it that both teams had about the same amount of starters unavailable for the game. Sorry for the thread jack

[It'sNotAFakeID]

I don't think we are worse than Ole Miss or South Carolina, nor are we worse than North Carolina.

 

Clemson and FSU would be a toss up.

 

I like the work you're putting into these things - I just think sometimes stats don't carry over. For instance, do we average more yards against those teams? Is our defense better against those teams because they pass more? Is it worse? Who knows. Love the analysis though.

 

Well remember that each team in each tier has about an equal chance of beating each other. So Nebraska was statistically level with Georgia and South Carolina last year. With the ACC, there wasn't much of a difference between the elite and the next step down, but that difference was still there.