SB Nation - Login for mobile commenting

Burnt Orange Nation

Pundit Roundup Talks Narrative

Punditroundup

Highlights this week include: some talk about the national narrative in college football and how that affects Texas' chances at reaching the national championship game; the Undulating Curve of Media Hype and more.  Click on through to keep reading!

Star-divide

I'm tired of people talking about "style points," both on this website and elsewhere.  (Though, to be honest, very few people talk about style points outside of FanPosts on BON.)  Texas fans have been in a frenzy over style points ever since OU won the tiebreaker last season, with some calling for Mack to run it up this season and others instead imploring him to keep it "classy."  It's an inverted echo chamber of nonsense, and it needs to stop.  The reason isn't so much that either side is wrong, but rather that it just doesn't matter.

First of all, Texas was done in last season by the computers, not human voters.  And as you no doubt recall, computers are not allowed to take into account margin of victory, and they still had OU ahead of Texas.  The human voters actually had Texas ahead of OU overall (albeit by a very slight margin) after the regular season despite all of OU's "style points."  Now, the argument could be made that those style points are what made the human vote close to begin with and Texas would have easily overcome their deficit in the computers if the humans hadn't been so impressed with style points.  And that may be true.  But my point is that a completely objective measure (whether it's right or wrong, it's not swayed by "style") had OU ahead and a mostly subjective measure had Texas ahead.  This despite the fact that OU's offense was throwing up 60 points per game every week towards the end of the year both in an attempt to woo voters and an attempt to not lose games due to their atrocious defense.  So if an objective measure that can't take into account "style points" has OU ahead and a subjective measure that can take into account "style points" has Texas ahead, how important can style points actually be?

Billy-zane_medium Hi, my name is billyzane and I'm calling for a moratorium on talking about "style points."  Furthermore, did you know that when I had hair, apparently I looked sort of like Marlon Brando? Who knew?

The answer to that is this: only as important as their place in the overall narrative of the season.  "Style points" have very little value on their own.  Any idiot voter can see that scoring 60 points on Idaho State isn't impressive.  But if the narrative of the season is, "Geez, this might be the greatest offense of all time," then scoring 60 points against a mediore to bad team is a story: "Well, OU did it again!  60 points for the 4th straight game!"  The narrative is what counts and "style points" (whatever the hell they even are) only matter insofar as they fit into that narrative.  Of course, scoring a ton of points had a lot to do with starting that narrative going in the first place (though it wasn't the only reason--the fact that the Big 12 was perhaps the greatest offensive conference ever and OU had the best offense in that conference had something to do with it), so it's a bit circular.  But the important thing to note is that Texas' narrative in 2008 "45-35!  We're getting screwed here!" won out over OU's narrative ("OMG.  Greatest offense ever!") when Texas won the human vote.

The preseason 2009 narrative started off as a national championship cerberus which neatly coincided with the top three Heisman candidates.  Florida was the "national champions returning everyone on defense and Jesus on offense."  Texas was the "team that may have gotten screwed last year and it out for revenge."  OU was the "team that lost the national championship but had players forego the draft for one last shot." Obviously, a few people made noise about USC and Alabama, but for the most part it was neat and tidy.  All Texas had to do was go undefeated, beating OU in the process, and they would be in the national championship game.  But then OU lost and Sam Bradford got hurt.  Now the narrative strength that would come from beating OU has diminished slightly if not entirely.  Everything is is flux and speculation about USC passing Texas even if both go undefeated runs rampant.  You know that old Hollywood adage (coined by William Goldman), "No one knows anything."?  That's true of college football too.  No one knows what twists, aberrations or tangles of football idiosyncrasy may come in the weeks ahead to distort the narrative further.  But we can still do our best to parse out where the narrative is headed now, where it might diverge in the future, and how this affects Texas' national championship hopes.

Twitter_tracker_medium

Bruce Feldman: Lost in Lubbock: 'Ur lookin for the Marsha Sharp Parkway... we re not naming roads after the bailiff on Night Court.'

[I don't really know why this is funny, but it totally is.]

Although the AP no longer has a say in the BCS rankings, and thus the media by and large does not directly affect the rankings, nothing drives the narrative like the media and nothing affects the prism through which voters perceive a team more than that media-driven narrative.  So let's see where we currently stand with the media's narrative.  You'll notice that I primarily talk about USC here.  That is because this narrative is predicated on Texas going undefeated.  If Texas does not go undefeated, then the narrative completely changes and we can talk about that situation if it arises (Personally?  I think Texas is probably screwed.).  So, if Texas goes undefeated, only one of Florida, Alabama and Ole Miss can go undefeated, and I think only undefeated Florida would be ranked ahead of Texas.  The only other potential threats to pass an undefeated Texas are from Penn State (won't happen because they play in the Big 10 and no one thinks fo them as a threat), Cal (a potential threat if they beat USC and go undefeated, but they're a threat to Texas for the same reasons as USC but to a lesser degree), and a mid-major like BYU.  I think BYU has a great shot to go to the national championship game if a bunch of teams lose.  BYU is not jumping an undefeated Texas.  Sorry, boys.  Thanks for beating OU though.  So that's where we are: in terms of teams that might jump an undefeated Texas, it's pretty much only worth talking about USC at this point.  Let's get to it.

The Good:

  1. Just about everyone in the media who's not senile (ahem, Beano and Lou) predicted a national championship game matchup of Texas and Florida.
  2. OU is, for the time being, no longer a part of the national championship narrative, leaving only a few teams to talk about.
  3. Everyone assumed a USC blowout of Ohio State and, though USC did win, they did not look particularly dominant in doing so.  Beyond that, at this point a victory over Ohio State does not have the same cachet that Texas' virtually identical victory over Ohio State in 2005 had.
  4. Texas still has the "we got screwed last year" narrative going for them.
  5. Thanks to the hype of USC/tOSU game and the Michigan/ND shootout, talk of Oklahoma State's loss to Houston has been far more muted than was talk of Oklahoma State's win over Georgia in Week 1.
  6. Colt McCoy vs. Tim Tebow is an extremely compelling storyline that is ready-made for the maudlin storylines that mainstream writers adore.
  7. Although at some point in the first half of this weekend's game, ESPN ran a headline saying, in effect, "Texas struggling with Wyoming," at the end of the game, the headline said "Texas Rolls" and, frankly, no one is talking about it.  Relatedly, because of Versus' contract disputes with DirecTV, a LOT of people did not see Texas' first half offensive struggles.  Um, thanks Comcast?
  8. Despite USC's big win (and Bama's victory over Virginia Tech), Texas is still ranked #2 in both the AP and Coaches polls.
  9. Texas has a primetime game on national television this Saturday with which to make a statement (a la the Missouri game in 2008).
  10. USC has games against a few Pac-10 teams that are a lot better than their national perception would indicate (i.e. Oregon State and Washington).  Texas' version of these games is against a team (Baylor) that at least everyone is talking about being a sleeper.  That is to say that USC's conference schedule is probably harder than it is perceived to be, while Texas' conference schedule might be easier than it's perceived to be.
  11. The Bad:

  12. Florida leads by so much in the human polls that, barring a Gator loss, they may be the only team that is assumed into the national championship game, and the #2 team may be thought to be up for grabs.
  13. The Big 12 seems to have taken a step back this year, at least in terms of national perception, with the losses by OU and OSU, plus the complete implosion of both Colorado and Kansas State.
  14. Formerly marquee games versus OU and OSU have lost some luster.
  15. USC plays Cal before Texas plays OU.
  16. Matt Barkley is, for some reason, considered by the media to be a golden god in the making.

Speaking of Matt Barkley, here is a chart approximating the level of his media hype during the Ohio State game, with an overlay showing the quality of both his and Joe McKnight's performances over the course of the game:

Barkley_hype_chart_medium

Well, that seems logical. 

Getting back to the narrative, there are, in effect, two types of media-driven narratives to consider.  One is the season-long narrative and one is the week-to-week narrative.  Texas should feel quite comfortable with the season-long narrative as the balance of evidence seems to still favor Texas.  First, the preseason narrative had Florida and Texas as the presumptive favorites and thus, if both of them go undefeated, I think the default position of the media would be that they should both be in the national championship game, even if USC is undefeated.  This notion is supported by the idea that McCoy and Tebow are both seniors and presumptive Heisman favorites while Barkley is a freshman and will have another chance (plus the McCoy/Tebow personal storyline).  Additionally, Texas is still viewed as aggrieved from the 2008 debacle and might be viewed as deserving of the benefit of the doubt.

All of this is to say that if Texas wins out, and does so while playing well, I don't think media members will start to advocate an undefeated USC over Texas.  Certainly some will, just as some probably would have advocated Penn State over Texas in 2005 had PSU not lost to Michigan on the last play of the game.  But just like that year, Texas entered the season #2 and a presumptive national championship game participant, and despite struggling some against Oklahoma State and A&M, were never really threatened by anyone (even before everyone else lost a game).  And contrary to vocal minority belief around these parts, the media is not in the tank for USC.  (Anyone remember the insane hype around Colt McCoy a few games into his freshman year? Or the fact that USC has finished the regular season with the same number of losses as a team that made the national championship game each of the last two years but wasn't the team chosen?)  When the narrative favors USC, then the media props that narrative up (remember, it's the narrative not the individual school). But this year, the narrative favors Texas and Florida, not USC.

Hey, You Know Who Doesn't Suck?

Those Versus announcers. I had to look up their names because I had no idea who they were, but Joe Beninati (play-by-play) and Glenn Parker (color) did a very competent, sometimes excellent, job of calling this game, giving frank assessments and often pointing out usually-overlooked things like a great block that freed a receiver.  I'm shocked when I get good announcers at all, and when I get them from a football outpost like Versus, well I'm flabbergasted but appreciative.  You paying attention Fox Sports?

However, and this is a large caveat, oftentimes the week-to-week narrative can overcome the season-long narrative if one team isn't holding up their end of that season narrative.  On the one hand, in 2004, USC and Oklahoma (the preseason 1-2) could not be overtaken by undefeated #3 Auburn because the Trojans and Sooners won all of their games that year by convincing margins: i.e., "We thought these teams deserved to be here to begin with, and they have done nothing to prove that they don't belong here so the team that gets left out is the one that we didn't think belonged here in the first place."  That kept the season-long narrative alive.  However, if OU or USC had faltered somewhat towards the end of the year (while still pulling out victories), then the media might have agitated in favor of Auburn and voters may have taken a look at how the teams were playing week-to-week and thrown out the season-long narrative.

To that end, the week-to-week narrative in 2009 will of course depend on how Texas and USC play in any given week, but there is a broad outline we can draw about how that might go.  First of all, Texas has a great chance this weekend in prime time on national television to show how dominant this team can be.  Considering Texas is still ranked #2 even after USC beat tOSU without the Horns playing anyone of substance, a win over a good conference team with the nation watching would go a long way towards pushing Texas higher in the rankings.  Beyond that, it would be great to have a dominant (semi-)signature victory before USC plays top-10 Cal on October 3.  The fact that this game comes before Texas plays OU may have a profound effect on the week-to-week narrative if Texas doesn't dispose of Tech handily. 

For that reason, to keep the season-long narratives alive, Texas needs to play very well against Tech.  They don't have to run up any score or get any "style points", but they need to win and they need to play well doing so.  USC plays a vastly improved Washington team this weekend that still has the stench of last year's 0-12 record.  I'm not saying Washington has much of a chance of winning that game, but Jake Locker is exactly the kind of mobile quarterback that Pete Carroll has had trouble defending (I've been saying this since 2007, but when you watch him play, he looks more like Vince Young than anyone I've ever seen; he's incredibly athletic, has the same slippery elusiveness that Young had, and frankly I don't think teams take him seriously as a runner because he's white).  A closer-than-expected Washington/USC game combined with a strong Texas victory rights the ship and I think Texas will be fine after that.

Now, to placate the "ZOMG STYLE POINTS!!1!"-style people that I respectfully disagree with, I will say that if this season devolves into a week-to-week narrative battle between USC and Texas, then "style points" (still not exactly sure what that means...) will probably play a part in how Texas and USC are viewed because after October, the only teams of any consequence that USC and Texas play are UCLA and Kansas, respectively.  It's a battle of who can beat up the bad teams better.

But the overarching point of this was to say that if Texas maintains a high level of play throughout the season, and as such maintains the season-long narrative of "we got screwed last year, we deserve this, and everyone knows we deserve this," then we have nothing to be worried about.  As I said earlier, even though a vocal minority around here likes to complain about the media being in the tank for USC, the past two seasons the Trojans have not made the national championship game despite finishing the regular season with the same record as a team that did.  Last year they did not pass Texas, Florida, Alabama or OU despite having the same number of losses, and the year before they did not pass LSU, Virginia Tech, OU, Georgia, or Missouri(!!!) despite having the same number of losses.  USC jumping teams with similar records is not only not a foregone conclusion, it's historically counterfactual!  Texas only has to worry about USC if they don't play up to their own high standard the rest of the year.

And finally, without further ado, but with apologies to New York Magazine and Adam Sternbergh, here is the Undulating Curve of Media Hype.  If you can't see it, click it and it'll open in full size.

Undulating_week_2

5 recs  |  54 comments

Comments

Excellent write-up

Any problems if I simply link back to this article the next time I see one of the nervous nellies insist that USC is going to pass us? I get tired of banging my head against the wall.

The one quibble I would make is terming as “bad” the fact that the USC-Cal game is before the Texas-OU game. I’d rather have the shot to make the big impression in a big game later in the season rather than sooner. OU certainly benefited last year by having its high profile games in November after we had concluded our run of tough games.

This is my favorite post every week.

I will quibble with you over USC not being the media’s love-interest, however. They did pass us up last year by beating PSU in the Rose Bowl while we beat tOSU, despite the “Texas got screwed” narrative thing. We didn’t even finish third. I remember how everyone was talking about how we needed a plus one system so that USC could play Florida. WTF?

I think the verbal tongue bath laid on USC is driving the perception that they are the media’s pet team more than any actual vote garnering (at least when it matters).

USC passed us on merit.

As did Utah, at least in the AP, based on the relative impressiveness of the teams’ bowl performances. If Texas had defeated OSU as impressively as USC handled PSU, I have no doubt we would have remained ahead of the Trojans.

People look at the polls after the season?
BZ, you're really going above and beyond

Wonderful read.

Wow

“I’m tired of people talking about “style points,” both on this website and elsewhere."

Well said supreme leader. Dissent will not be tolerated.

Not a dictate. People can talk about whatever they want (within reasonable bounds).

I just think that the way people are talking about it is unproductive and hysterical (to pay homage to PB’s insane yet apropos metaphor). This was just an attempt to frame the discussion.

If you say you don't like my opinion your a communist fascist
The beatings will continue until morale improves
Great write up-thanks

Too bad the pundits at ESPn can’t analyze or write this well.

I would've agreed with you about USC not being a media-darling

until this last week, when I was reminded once again that they are. I don’t know how many times I saw those stupid puff pieces about USC and what classic songs they are currently rallying around. And the love Barkley is getting is completely disproportionate to how good he actually is, which you do a great job of pointing out, yet don’t attribute to media bias, which I think is exactly the reason. Tate Forcier was much more instrumental in his teams win, and also in his second game ever, and it’s MICHIGAN, but he hasn’t gotten nearly the screen time.

Much of the reason is that the USC game was on ESPN, so they are obviously going to promote their own product, but this doesn’t reduce from the problem at all, because ESPN is the overlord of sports narrative. Also, the school is in LA. You cite the fact that USC hasn’t gone to the NC in the last few years despite having the same number of losses as the teams that played, but I think that had mostly to do with the timing and quality of the losses. Last year Oregon State looked bad at the time of the upset, and the top 5 was so bottlenecked with the Big XII South and Florida that there was just no room to move back in. In ‘07 they lost 2 games in three weeks, including Stanford, which is much more detrimental to a team than 2 losses far apart (see LSU), and in ’06 the loss was the last game of the season. The BCS has done a good job of fighting the bias to put the correct teams in the NC game (in that they haven’t explicitly ignored the credentials of more deserving teams), but that doesn’t mean the bias isn’t there.

Perhaps the fact that USC had the biggest game of the week had something to do with the attention they received last week?

I imagine USC fans will be similarly complaining about pro-Texas bias the week of the RRS.

Isn't that the....

….. same Saturday USC plays at South Bend? You don’t really believe our game is going to receive more media attention than their game with Notre Dame do you?

Good piece Zane.

I do.

We’ll be on ABC (or, technically, ESPN on ABC) and USC-ND will be on NBC. If one subscribes to the theory that ESPN’s true bias is not a pro-USC bias but rather one that is pro-whatever the hell its family of networks is broadcasting any particular week, then I would expect, assuming no additional losses, Gameday would be in Dallas and that the RRS will receive more attention from the primary source of college football news.

Plus, given last year’s soap opera, the build-up to Texas-OU will provides better theater than USC-ND, so I imagine that the game will also be the primary focus of outlets other than ESPN, like SI, etc.

If ND were undefeated going into the game and had a chance to catapult from an upset of the Trojans into the national championship picture, I might think otherwise, but obviously that’s an impossibility now.

Agree

with the part about RRS getting more hype because the game is on ABC, although if neither USC or ND loses between now and then I’m sure they’ll spend a while talking it up during the week, but still not more than OU-UT. Gotta chase them ratings. I’m going to appreciate how important they are going to make the game seem after last year’s drama, despite the fact that all the controversy was due to the fact that the result of last year’s game didn’t in fact mean anything.

If Bradford makes it back that week (or before) you know that ESPN/ABC will be all over the game. USC, ND, etc. will all take a backseat. They (talking heads) just love to hype a usually nonexistent story.

Unless OU gets beat by Miami.

They did not look too bad last night beating up GT. Could be tough for the dirt burglars.

This

despite the fact that all the controversy was due to the fact that the result of last year’s game didn’t in fact mean

As BZ noted above, IT did matter to the human voters. An even some pundits on ESPN, including Herbie, championed this point as a reason for Texas to be ranked above OU. Unfortunately, at the end of the day it wasn’t enough in the CPUs thanks to Mizzou and Neb.

A 10-2 Notre Dame team still likely to make a BCS Bowl
ND is automatically in

if it’s in the top eight of the final BCS Poll, but is a near sure thing as an at-large at 10-2.

I’ll predict 8-4 (other losses to USC, 1 from road games at Pitt/Stanford, one to either Michigan State or Purdue). At 9-3, Irish have a shot at BCS, probably dependent on whether a non-BCS team qualifies.

As I said before,.....

…. a 10-2 Notre Dame team still likely to make a BCS Bowl.

narrative

Another thing to remember is that if we win ‘em all, even the narrative PROBABLY won’t matter. It only matters if 3+ teams go undefeated, which rarely happens. Or if we end up tied with other one loss teams, but most horns seem to be resigned to not playing in the MNC if we have a loss. WIth a freshman quarterback, its even more unlikely that USC will actually go undefeated. So since the chances are that if we win out, we’ll be the only team or one of two to do that, we can probably win ‘em all ugly and still be ok. Even if a team like BYU wins out too, we’d still probably go ahead of them even with ugly wins. Not saying we should do that, just sayin if we did, history says we’d likely be ok.

Don't forget the Big 12 championship.

 Voters tend to remember the last thing they see and hear. The commentators blathered on and on about how Oklahoma was the greatest offense with possibly the next Heisman winner, yada, yada, yada…

We were not edged out of the BCS championship until after that game. If it’s a race to the tape with a PAC 10 teamI hope we have a great showing at our championship with our own Heisman candidate that deserves a shot at the crystal football.

we were edged out before

thats why we didn’t play in the Big12 championship. But if we are one of several undefeated teams, the Big12 championship will help our SOS and provide a last statement game for us before the final BCS standings.

My bad, you are correct...

As far as SOS, should we win out, will take some hits…at leat two losses for top ranked Big 12 south opponents, OU (sux), Okie lite and maybe an abysmal big 12 north team in the championship.

Nebrask might be sporting a decent record

If not, there’s KU and Missouri. Missouri doesn’t play OU, and teams like Iowa State, Coloraod, and KSU, “should” provide wins for all these teams. Of course you never know.

The same Missouri team that struggled with Bowling Green?
And the same Missouri team that plays in the North

I didn’t say they’d be great. I said they might have a decent record.

Yes, and elephants might fly in Houston tomorrow.
Washington could beat USC this week

and end this debate. They won last week and looked pretty good against LSU in their opener. Jake Locker is probably mad that Barkley is getting all the hype. This is a trap game for USC. Plus I think I read that Barkley may not play, so USC is pretty questionable at the QB position.

I actually think this isn't too far fetched also

They’ll be on an emotional low, on the road, against a coaching staff who knows their system and a team who is better than they get credit for, with a QB who is either hurt or has never started a game. And USC drops a game like this every year.

except that...

I’ve been reading this diagnosis quite a bit in the mainstram media as well, and in my experience it seems that whenever an upset pick becomes trendy, not only does the upset not occur, but the favored team usually blows the underdog out of the water.

But then, nobody knows anything, right?

BZ: Thanks for the words

A lot of them. Well put together.

Glad somebody else noticed the Versus announcers. I thought the play-by-play guy was B+ (his little phrases accompanying each player in the starting lineup were inane). The analyst was superior. Inciteful. Brief. On point. Solid A grade.

Both knew the players, didn’t stumble and fumble around reaching for names. They’re probably paid half what ESPN’s egos (there are a few worthy ESPN throats, but not many) are. You’d have expected some Wyoming bias; if there was, I didn’t hear it. They played up the big Cowboy second quarter — justifiably. Best of all, they let the game tell the story and talked about it. No shilling, no cutesy comments, no stale storylines.

Too bad we’ll never hear ’em again.

directv

thanks to my satellite tv, I could not watch. Good point in this article about how a lot of people probably did not see Texas struggle in the first half.

But I have no doubt the announcers did a better job than the usual suspects we have to deal with. I don’t mind Herbstreit or Andre Ware. The worst is that dude Palmer. It boggles my mind how he got that job.

the bachelor prob. got him the job..nt
It seems to me

that the reason USC didn’t go to the championship the past two years despite having the same record as a team that did is because their losses were to worse teams than the losses of the schools that did go or that they didn’t pass. In the scenario you’re talking about they would have no losses, so those precedents wouldn’t apply. People would also focus on our easier OOC schedule while not focusing on their easier conference schedule.

Speaking of narratives, if UT and USC both go undefeated, the narrative fight will be between “Texas got screwed last year” and “No team with a freshman quarterback starting the entire season has ever won the championship”. I think which of these would win out in the end is a tossup, but many people would take fresh new hope and something unprecedented over it’s their turn or sour grapes. These people will give McCoy the Heisman as a consolation and USC the championship game.

I could see that happenning

I know there are some statistics that say USC and the media do not have a deal with the devil, but every year I see examples that this deal exists. This is not just Longhorn bias. I started to despise USC at the end of 2003 when they declared they were going to play Michigan in the Rose bowl for the national title! Funnily enough, USC won the AP vote. But LSU won the crystal ball. Last I checked all the conferences and coaches agreed that the national champion is the owner of the crystal football. I guess USC likes to have it both ways, since they claim they won a national title in 2003. In Rod Smart lingo, “I hate they”.

The announcers who did not suck

I read somewhere that they are usually hockey announcers? I did also enjoy their frankness.

2005

Another factor helping us is the parallel between 2005 and 2009. The pundits know how this can play out — Florida proving it’s the best team ever with the best QB ever going untouched through the season, Texas hanging in there with the most efficient QB ever taking us through an undefeated season, all leading up to the game of the century in the Rosebowl (with Heisman as sideshow). I just don’t see how they can resist that.

Style points still matter and the objective computers are flawed by the subjective BCS system...

While the computers themselves can claim their rankings are objective, the way the BCS includes them in the formula isn’t. The computers often end up with a much wider disparity than the polls due to their being an average of just 4 rankings rather than a percentage of hundreds of votes cast and so the difference between one position and the next is often much higher than it will be between the polls. This is certainly a problem since a third of the equation makes a greater point distinction than it should between teams ranked next to each other. For Texas and OU, the .0400% difference in their November 30th computer numbers (OU was #1, we were #2) equates to a 100+ point difference in the Harris poll (where we had 6 more points than OU or .0021%) and a 60 point difference in the Coaches poll (where OU had 1 more vote than us or .0007%). So the computers we suggesting OU was a much better team than were were while the voters on average saw it as a draw. The BCS should be using the computers’ point differences between teams rather than just awarding whole points based on overall rankings. This would be a closer approximation to how the BCS deals with poll voters, which is where I’ll go next.

I think you are misrepresenting the role the voters play in the BCS formula. If the BCS just used the overall aggregate like they do with the computers (i.e., you get 25 points for being #1 in the Harris poll, 24 for #2 , etc), you could make the argument that a tiny majority of voters seemed to recognize 45-35>60 pt. barrage, but the BCS is ultimately giving voice to each voter. So even though we ranked higher than OU in the Harris poll we were ultimately done in by the fact that a lot of poll voters either put OU a spot ahead of us or more egregiously put OU many spots ahead of us. Art Briles ranking us at #7 rather than #2 cost us 5 points in the coaches poll (almost 10% of our deficit to overcome OU’s computer numbers). Each of those votes directly effected the point total of the BCS. The actual total ranking didn’t matter. While I in theory think this is a fair way of doing it, since a consensus #1 should be well ahead of the pack (and Alabama was just about consensus #1 in the polls on November 30th even their #3 computer ranking wasn’t enough to drop them.), it also raises the importance of having legitimate and fair voters which is where the theory of fairness and reality don’t mesh.

If the voters really did understand how the teams would stack up they would have had OU at 5th rather than USC. If OU had USC’s vote total from Nov. 30th even the overcompensating computer ranking wouldn’t have been enough to push them past us.

So while the computers can’t factor in the style points, except in the sense that a 10-win Cincinnati team, rightly or wrongly, was worth more points than a 10-win Rice team, the BCS’s use of the computers overemphasizesthe rankings rather than point difference between each position which is how the BCS deals with the polls. Add to that the voters were still overranking OU by quite a bit based in large part on their offensive style points and you have a double whammy. It wouldn’t have even taken a lot of votes just people ranking us fairly. How many other voters in both polls were making such wildly uninformed votes as Art Briles either because of their grudges or because of style points?

Anyway, sorry for going on so long, this is less about us being screwed last year and more about the continuing flaws in the stupid BCS. I think OU’s style points were ultimately part of our undoing. The media made a big deal about it and if that caused dozens of voters to rank OU several spots ahead of us then that ultimately cost us the division championship.

If I understand you correctly, then you're incorrect.
The BCS should be using the computers’ point differences between teams rather than just awarding whole points based on overall rankings. This would be a closer approximation to how the BCS deals with poll voters, which is where I’ll go next.

You seem to be saying (correct me if I’m wrong) that the BCS treats computers differently than humans, but that’s absolutely not true. The BCS treats the human polls and the computer polls exactly the same. Each human voter ranks 1-25 and they award 25 points for a 1st place vote on down to 1 point for a 25th place vote. Likewise, each computer voter ranks teams 1-25 with the exact same points system. The BCS then just equalizes each of these on a scale such that a team with 100% first place votes in a poll will have a score of 1.0.

The problem you’re outlining (which does exist) comes from the fact that there are far fewer computer pollsters than human ones. Because of that, each individual computer’s ranking carries more weight than each individual human’s ranking. But to take into account the actual computer scores of each computer rather than the 1-25 rankings would be to actually treat the computers differently than the humans. Now, you might be saying that you know this and you’re arguing for it because treating the computers differently than the humans means that you’re treating the entire computer poll more like the entire human poll. Which is a valid argument, but one I still don’t think I agree with.

I see what you are saying...

But I don’t see the human voters and the computer rankings as the same. They would need to increase the number of computer rankings by at least a factor of 10 to balance how they are scoring the results if they want to argue the validity of the methods.

I see it the way you explained it in your article. The computers are the objective measure and the polls are the subjective measure. The problem is the BCS takes each computer’s objective result and then turns it into something largely subjective by ignoring how each computer scored each team and rather they take the aggregate ranking which is an entirely arbitary way of looking at the data. If the computers are objective then it is important that Texas was only 0.25% better than team behind it rather than 4.0% that the aggregate supplies.

It then takes the subject measure and takes smallest, most subjective detail (the individual voter) and amplifies it. Why is it so important to score that Art Briles ranked us #7, but so unimportant that that Anderson/Hester had us .002 points ahead of OU rather than 1 full point? They are applying the wrong methods to the two types of testing they are using.

You're still thinking about this backwards.
Why is it so important to score that Art Briles ranked us #7, but so unimportant that that Anderson/Hester had us .002 points ahead of OU rather than 1 full point? They are applying the wrong methods to the two types of testing they are using.

That doesn’t make any sense. It’s an incorrect analogy. The correct analogy would be between (i) how close one cmputer thinks two teams are and (ii) how close ONE voter thinks two teams are. So the analogous situation to your Anderson/Hester example above would actually be Art Briles thinking #6 and #7 were actually extremely close together, but having to rank one team #6 and another #7.

I understand what you’re saying though. You want BCS rankings to take into account how exactly a computer rates teams, not how they rank them. For example, if a computer thinks Florida and Texas are virtually even, but Florida is .0001 ahead of Texas, then Florida gets a full point more than Texas does in the rankings. But the human polls do the exact same thing. A human voter may think Florida and Texas are virtually even, but he just thinks Florida is slightly ahead for whatever reason and puts Florida #1 and Texas #2, even though they are virtually identical in his head. It’s the same thing.

Each computer is treated exactly the same as each human. The ONLY difference between the polls is that there are more humans than computers (and, for some stupid luddite reason, they throw out the highest and lowest computer rankings). That’s it. If you changed the computer system to how you want it, you would also likely have to change the human system by giving each human a certain number of ranking points per week and allowing them to divvy them up however they pleased.

As for your objective/subjective thing. You don’t make anything subjective by looking at how they’re ranked as opposed to how they rate according to your system of rating. Objectivity just means looking at the facts and only the facts. Which is what computers are forced to do. Certainly, how their programmers program the formula is subjective, but once that’s set, the rankings themselves are objective. What you do with the data that’s spit out doesn’t change that the data is objective.

I think the analogy of comparing a computer process to a human's thinking is the bad analogy.

Humans aren’t given any criteria for their votes. They don’t even have to watch the games. It is ultimately why they take hundreds of voters then take an average. They know each individual voter is fallible. The polls are quite careful to avoid showing their work. They know they have credibility only in their multitudes. The computers aren’t chosen because they need a lot of them to balance out the mistakes. There are no mistakes or ‘hidden’ biases (At least I think all the computer polls show their methodology, though figuring out the biases in each methodology would take a math expert!). Why they drop the highest and lowest is even more silliness, but also puts to lie the idea that the BCS treats the computers like the human polls.

The only reason I see for the BCS doing it they way they do it is because its easy and the mathematics are simple enough to not make everyone think there is something shady going on. I actually think if they added back in the finer details of the computer models they would actually help their perception since it would reveal how teams are clustered in each method rather than blunt force rankings.

I think the objective way of looking at ranking teams is closer to the computer models which give results that are deeper than just 1-119. Team #1 has a better composite score than Team #25, but the computer model is going to show you that perhaps Team #1 is more like Team #2 than perhaps Team #3 is to Team #2. This is important to the model. While beating #1 might have big payoffs, beating #2 might not be much different, while beating #3 might actually not help a team any more than beating team #20 if there really isn’t that big of a difference between them. Humans do this too, but everyone does it differently and some don’t do it at all or blatantly ignore that reasoning to push a different goal. The computer is tasked with something very specific and the modelers aren’t constraining it to use a simple set of 1-119, but the end result that the BCS uses reverts back to this less precise number which really begs the question of why they even bother with the computers in the first place.

Thanks Billy

for another round of kitchen-table-styles pundit’s debate. Nice graphic of the Undulating Curve…Luv it!

I really don’t worry about USC at all. Our spot for MNC game is set and destined if we do our job not losing a single game this season. Actually, what I’m worrying about right now is if I have good enough money to cover the ticket, hotels and other expenses in Pasadena!

Glad

You’re back BZ.
To me the narrative is set and i agree with all above, especially 6.
Keep the posts flowing holmes. Where is lady BZ?

Irrelevant unless...

we do a an ‘08 Missouri on Tech. If it is close ,I’ll bet you my left-uhhhh horn, that usc beats cal and jumps us with the coaches.Then all we can hope for is to run the table and hope usc pulls a brain fart against an unranked team again.

So you read all of that, disagreed, and your rebuttal, with no argument attached, is "Irrelevant"?

Give me a little bit more to go on here. “Irrelevant” isn’t an argument unless you say why it’s irrelevant.

style points misconception

I’m with BZ on style points, and this comment is intended to help end the talk about them, like BZ’s post, because I don’t think many people have thought about what style points really are.

Style points are getting a big lead in the first half (MAYBE starting the second half strong too). Every team and every coach, including Mack, already tries to do that, regardless of the opponent. It just doesn’t always work – see Wyo game.

Style points are NOT adding a couple touchdowns in the fourth quarter to an already big lead – thats running it up. All you (or voters) have to do is look at a box score to see the difference. And I think they do.

Like it or not, the vast majority of points scored in OU’s string of 60 pt games came in the first half, and with the exception of the Big 12 championship where Stoops ran it up, very few of OU’s points came in the fourth quarter. Scoring an a$$load of points in the first half is dominance and IMO it is legit to consider that in looking at a team’s body of work. Piling it on in Q4 is bad sportsmanship, and again, people who follow college football CAN tell the difference.

(I have no problem with letting the second stringers score in the fourth quarter, because there is a legit goal apart from running it up – to get your backups real game experience. Just not first string when we already have the game in the bag.)

So, lets stop the cries to pile it on. Instead, lets do what we always do anyway and pull for our team to dominate when the game matters. We don’t need a change in philosophy towards putting nice big numbers up there, even if those points come late. We just need to play better from opening kickoff. Mack, the team, and the fans already want that and are trying to do that every week. Talking about style points implies running up the score, and there just isn’t any reason for it to enter the discussion.

60 points are the mag rims on a beat down...all style, no substance.

49-14 over NU at half-impressive; 49-14 at end of 3rd-impressive; 62-28-knob slob Bob
38-14 over A&M-sort of impressive; 66-21-impressive; 66-28-knob slob Bob
42-7 over Tech-impressive; 58-14-impressive; 65-21-knob slob Bob
21-13 over OSU-not impressive; 37-26-hmm OU might lose; 61-41-knob slob Bob
38-7 over Missouri-impressive; 41-14-impressive; 62-21-knob slob Bob

I am actually mostly in agreement with you, since its hard to say how much effect 60 points scored has over 49 or 59 points in a blowout. I do think the Tech score had an effect on voters. If the game finished 42-21, I could see us gaining a lot of votes.

I guess we agree then

I hestitate to respond because OU sucks, Stoops sucks, I’m not an apologist. Its just that I like facts better than fiction, and we run the risk of being accused of being whiny and of doing the same thing if we make too big a deal out of something (McCoy still in and UT scores TD with 8 minutes left in the game against Wyo). So I’ll just mention:

MU – I pointed out was an exception. Stoops ran it up. MNC berth on the line, but still.

NU – your numbers are wrong. No pts in Q4.

A&M – no points in Q4, running it up in Q3 is grey zone, I don’t know if backups scored those points or not.

Tech – only 7 pts in Q4.

OSU – game was still contested into Q4.

Scoring even one touchdown in the fourth quarter with a big lead (which OU did in some other games last year) may be running it up, but only a little, and the overall point was that the big lead happened early, not late. Winning 60-20 is impressive if 40 or 50 of those points come in the first half. Winning 60-20 with a lot of points late not so much. We try to do the former every game anyway, so talking about style points is stupid and makes us look bad.

I will say in the Tech game

The reason they only scored once was because they only had the ball twice that quarter. They started off with the ball and Bradford remained in the game and they scored. After that, it was their backup, but they still tried to score, including going for it on fourth and goal but getting stopped at the 1. It was clear Stoops was gunning for 70. Also, against Oklahoma State, the last touchdown was totally unnecessary and meaningless, although you can arguably just blame Oklahoma State for not tackling.

I think talk of style points is getting out of hand, but there’s no question what Stoops was trying to do. I wouldn’t have a problem with it if I was confident that the voters watched the games and knew HOW the score got there, rather than just look at the final tally.

FYI -

I played that drinking game where you duct tape a 40 oz in each hand and drink every time BZ says the word “narrative.” I woke up naked in a gutter in Connecticut with 3 Domino’s Buffalo chicken Kickers stuck to my chest and a bad case of the Mondays. I may have lost a step since college.

Oh, what? Like you’ve never done it?

Dude, it was only 41 times.

You can’t hold your liquor anymore.

You must Login with your SB Nation account and be a member of Burnt Orange Nation to post a comment.