Burnt Orange Nation: An SB Nation Community

Navigation: Jump to content areas:


Sports blogs for fans, by fans.
New Blog: RSL Soapbox for Real Salt Lake Fans!

Playing the Numbers Game - Brain Style: Big 12 Opener

Heading into conference play, this Texas team is averaging the highest margin of victory of all of Mack Brown's Texas teams.  Sitting diametrically opposite from the 2007 ‘Horns, these guys have come out to take care of business four times out of four, with only a tiny hiccup (Bill Little says "surge") in the UTEP game for us to complain about.  The dominating performance by Rak & Co. up front and the steady progression of everyone else not playing middle linebacker has the defense on track to becoming a brick wall very stretchy rubber sheet with lots of knives behind it by the end of the season.  On offense, Colt is playing as well as anyone has ever played the quarterback position right now, aided by the fact that Greg Davis has discovered the concept of a counter (coincidence that Applewhite arrives at the same time?  You decide), and by Jordan Shipley looking as good as everyone said he was supposed to look all this time.

All cuddling aside, though, some Texas fans are not jumping on the peace-pipe train (next stop: Colt is better than Vince) yet.  We have a tailback with a conditioning problem, one with the usual starter vitriol being poured mercilessly on him while Mack and Greg make excuses, one that hasn't stood on two legs for more than one (glorious, 6 YPC) hour, and a really smart one that has a great personality good hands.  That would all be fine, if only our line could run block.  I don't know if anyone else has noticed, but we have two receivers, and they're both returning kickoffs at the same time.  Are anyone else's crazy bells going off?  Our most experienced corners are our worst ones, and our middle linebacker continues to be told either "Please don't blow this assignment" or "Just soak up a blocker, for Christ's sake."

So which school of thought are we supposed to join, then?  Well to help us decide, let's take a little walk down numbers lane... (Who saw this coming?  Not him.)

Star-divide

First, a table.  Chart?  Table.

Big 12 Openers and Early Performance

Year

Game

Record

Avg. MoV

Result

Delta (MoV)

Final Record

Final Ranking

SD(MoV)/Avg MoV

1998

@ #5 KSU

1-1

6.00

L 7-48

-47.00

9-3

15

5.66

1999

@ BU

3-1

17.75

W 62-0

44.25

9-5

21

1.37

2000

TBPSU

2-1

29.00

W 42-7

6.00

9-3

12

0.96

2001

Tech

3-0

30.33

W 42-7

4.67

11-2

5

0.12

2002

TBPSU

4-0

34.25

W 17-15

-32.25

11-2

6

0.29

2003

#16 KSU

3-1

33.75

W 24-20

-29.75

10-3

12

0.89

2004

BU

3-0

29.67

W 44-14

0.33

11-1

5

1.09

2005

@ MU

3-0

33.67

W 51-20

-2.67

13-0

1

0.82

2006

ISU

2-1

25.67

W 37-14

-2.67

10-3

13

1.44

2007

KSU

4-0

19.00

L 41-21

-39.00

10-3

10

0.96

2008

@ CU

4-0

38.75

?

?

?

?

0.17

Sorry about the huge 'ness' of my char... table.  I'm  not exactly a WYSIWYGW (the last W is for Wizard!).  The gist is:  Delta(MoV) is Margin of Victory of Texas's first Big 12 game, and Avg. MoV.  SD(MoV)/Avg MoV is the coefficient of variation (CoV) of Texas's MoVs before Big 12 play.  Go ahead and wiki that.  If you didn't wiki it, then just know it's a relative measure of statistical dispersion (inconsistency, don't make me link to that belted WoW-rrior again...).  So now that I've given you a figure that you'll surely completely skip, let me give you a few that you may actually care to glance at...

Now charts?  Charts.  eeeeee!

Early_mov_vs

Early_cov_of_mov_vs

OK, first thing's first:  Did I have to change out my data markers for little Longhorns?  No, I did not.  We cool?

Secondly, I've slapped some trend lines on the graphs to highlight what I'm seeing in these.  You can see the trends rather easily for yourself:  Higher MoV, higher rank, lower CoV, higher rank.  Let's look a little closer, though...

Looking at the first graph, I see that the 2003 team finished ranked 12th despite an elite start in terms of MoV (whipping cruddy teams sure took the sting off of that Hog loss at home, eh?).  What this chart doesn't tell you is that we were ranked 5th before losing to Washington State in the Holiday Bowl.  Had we not pouted about our near BCS appearance and instead beaten the Cougs convincingly, you can see that the trend would be much more clear.  For you stat geeks, moving 2003 up to 5th would bump up the correlation coefficient for this graph up (down) from -0.69 to -0.81.  (I'm excluding 1998, just because it's an obvious outlier as Mack had yet to build Texas into "his" team and we played only two games (one against Top 10 UCLA) before heading into conference play.)

Looking at the CoV graph, the obvious adjustment would be to give 2005's MoV a boost for having to play OSU in the 'Shoe.  Indeed, adjusting the score of that game from 25-23 to 50-23 (a sub-average day for that team) improves the 2005 MoV to 42 ppg, and decreases their CoV(MoV) by almost 0.5 (out of 0.8!).  Geeks, that gives your correlation coefficient for the CoV graph a jump from 0.62 to 0.75, again forgetting 1998.

That was quite fun, but what does it tell us about this year?  This data shows that regardless of schedule, high victory margin and consistency are both indicators of a potentially great team.  Go ahead and glance back up at that table as you swallow that last sip of Burnt Orange Kool-Aid before conference play starts, and you'll see that this team has it in spades.  The numbers put this team in the same category as some of Mack Brown's better teams over the years, and it looks like a potential Top 5 team when it's all said and done.  If you count 2005, there are only three teams with CoV(MoV)s similar to this year's, and they all finished in the top 6.  Of Mack's 4 teams to average an MoV over 30 ppg, only 2003 finished outside of the top 6, and they went into their bowl game ranked 5th.

If the coaches can iron out a few more wrinkles in this team, then hang on for the ride, guys.  We may be sitting at ground zero when one of Mack Brown's better Longhorn teams hits the toughest stretch of schedule since the 90's with a full head of steam.  Cue this:

 

Thanks to mattadk, whoever you are, for the best Texas Football video on the freakin' internets.

2 recs  |  Comment 18 comments |

Story-email Email Printer Print

Comments

Display:

Yaa

Domination!

by BufreTX on Oct 3, 2008 8:58 PM CDT reply actions   0 recs

PS

Nice graphs!

by BufreTX on Oct 3, 2008 8:59 PM CDT reply actions   0 recs

OH

So what does this mean for next year?

by BufreTX on Oct 3, 2008 8:59 PM CDT reply actions   0 recs

wait, wait, wait

what were the p values on those regression?

by Wells on Oct 3, 2008 9:28 PM CDT reply actions   0 recs

Forgive me

I’m not familiar with p values. R squared values were pretty bad, like 0.4 or something, but given the sample size and noise (strength of schedule, injured players, schemes, weather, and on and on), I don’t think it’s realistic to expect anything approximating a formulaic relationship between these data sets. I tried to be modest and just hypothesized the simplest element of the relationship that I could: higher this equals higher that and the like.

Oh, wait, were you joking?

by Horn Brain on Oct 3, 2008 9:44 PM CDT up reply actions   0 recs

correlations are just all so much bullshit without p values

p values refer to statistical reliability – the probability that your correlations aren’t attributable to chance. oh, wait, you were joking about not being familiar with p values?

…and we’ve been running counters out of the shotgun for years.

by DrunkArmadillo on Oct 3, 2008 11:47 PM CDT reply actions   0 recs

Ok...

So I read into it a little bit, and I don’t think that p values are:

A.) within the scope of this article
B.) necessary to make my point

I didn’t even bother to explain correlation coefficient, the last thing I’m going to do is teach people about p values. Plus, when you say correlations are bullshit without p values, I’m pretty sure you’d be right if I was putting forth a linear regression as a model for predicting Texas’s finish based on the OOC schedule, but when all I’m saying is that there is some evidence that early dominance can indicate a team with the potential to be good when it counts, then there’s no need to go into all that. The claim is not specific enough to be validated or invalidated by something like that. It’s simple enough that you can look at the scatter plot and see whether the line goes in the right direction or not, and that’s all I’m really trying to get out of it. It’s like carrying eight decimals through all of your calculations but taking your measurements in cubits.

Regard the counter we run out of the shotgun. Yes, that was great, but now we have counters for the bubble screen and zone read as well. There’s no convincing evidence to prove that GD even knew he was running a counter. He could have just thought we were running a play to the other side. Now I’m joking again.

by Horn Brain on Oct 4, 2008 12:53 AM CDT up reply actions   0 recs

Oh hyperspace Swami

you there with the maidenform on your head and the perfect beat.

How do I reconcile this:

Are anyone else’s crazy bells going off?

With this:

Go ahead and glance back up at that table as you swallow that last sip of Burnt Orange Kool-Aid before conference play starts, and you’ll see that this team has it in spades.

Did you just steal the clapper and create a bigger one? And slipped a couple of cases of Red Bull to the bell ringer just for good measure?

Are you gonna be around in the fourth quarter?
 

by whills on Oct 4, 2008 2:19 AM CDT reply actions   0 recs

Jesus Whills,

It’s like you wake up every morning 65 years old, then go to bed as a 17 year-old Dead head.

by Horn Brain on Oct 4, 2008 3:00 AM CDT up reply actions   0 recs

Amazing, isn't it.

Although it’s “bed with a…” and I prefer a little older Dead head models. With a Cadillac.

Just trying to bridge the generations. heheh

btw the video is killer.

by whills on Oct 4, 2008 9:51 AM CDT up reply actions   0 recs

forgive the dumb Q, but the crazy bells

were in reference to having our two best WRs returning KOs, no? is the author saying they shouldn’t be out there together so as not to risk injury(s) to one of most vital offense tools? thanks.

JP in south Baltimore

by thejahpaul on Oct 4, 2008 10:30 AM CDT reply actions   0 recs

Great video

Mack should put a link to that video on his site, the historical sweep of it is so great. Just amazing.

The only thing that I wondered about your analysis is that it doesn’t consider strength of schedule, does it (I read it quickly so I might have missed that part)? The reason I bring that up is that this year UT has an unusually weak non conference schedule (who could know that Arkansas would be so down?) and an unusually strong conference schedule.

Also, the sample size is pretty small. But those are quibbles, it is an interesting and good analysis.

by Kafka on Oct 4, 2008 10:50 AM CDT reply actions   0 recs

No, it doesn't account for SoS

I just let that remain an outside variable, part of the noise. The good thing is that the trend still emerged, which is a positive sign for us going ahead. As for sample size, you’re right, obviously, and that’s why I’ve tried to be as general as possible when inferring things from the data. Good points, both.

by Horn Brain on Oct 4, 2008 11:05 AM CDT up reply actions   0 recs

Big shoes

Ogbonnaya’s 3.4 GPA may be good for a football player (with an honest major), but not really all that good. Dallas Griffin, on the other hand, had a 3.9 and honors with a double major in business and finance.

Before you criticize someone, walk a mile in their shoes. If they get mad, you're a mile away AND you have their shoes.

by Caradoc on Oct 4, 2008 11:05 AM CDT reply actions   0 recs

Comments For This Post Are Closed


User Tools

Welcome to Burnt Orange Nation, a blog dedicated to University of Texas athletics. Get BON updates via Twitter.
Start posting about the Longhorns »

Join SB Nation and dive into communities focused on all your favorite teams.

FanPosts

Community blog posts and discussion.

Recommended FanPosts

T1_1018_coltheisman_small
Texas vs UC Irvine Gameflow (+/- stats included)
Horns_small
Thanks to our Men & Women in the Armed Forces
Small
The Aggies are doing just enough this year to give us a good upset.

Recent FanPosts

Bevo_small
Aggie Week: Let's hear your favorite Aggie joke.
Ut_vs_fau_8-30-2008_049_small
Poor, Disillusioned aggies
Brandedbevo1024x768_small
#2 in USA Today Coaches' Poll
Small
Ohio State and Oklahoma Uni's
Small
BCS Formula Question
2401104_1__small
The Perfect Storm of College Football
T1_1018_coltheisman_small
Texas vs Western Carolina Gameflow
Lemon_small
Pat Murphy resigns
Small
Orakpo says TCU can't hang
Clock_tower_small
The Weekly BON Hit Award Baylor

+ New FanPost All FanPosts >

SPONSORS


Site Editors

Pb6_small Peter Bean

Dark_pumpkin_small awiggo

Menbooger_small GhostofBigRoy

Contributing Authors

Jersey_front_small 54b

Zombie_profilepic_small Horn Brain

Gse_multipart20834_small 40AS

Pigeons_small billyzane

Small whills

Brandedbevo1024x768_small dimecoverage

Rosebowl_small txtwstr7

Small TheElusiveShadow

Me_small burnt in ny

Official Partner of CBS Sports