Jump to content

Texas opens as a 5 point favorite v Oklahoma


TexasMan

Recommended Posts

Just now, Rickylovesweed said:

The issue with the advanced metrics is they don't value Wyoming as much as they should based on their style of play. 

For instance, Texas Tech is still ranked above Wyoming in almost every advanced metric by a good bit even though Wyoming beat them. When Wyoming beats a ranked Fresno State team this weekend people will start to realize that the advanced metrics aren't an accurate reflection of them as a team. 

So you have figured out a "style of play" loophole in advanced metrics that Sagarin, Connelly, Vegas, etc., are all blind to and unable to correct? Please describe exactly how Wyoming's style of play creates this conundrum. Does it bake in that their starting QB missed the Texas game? Also, if Tech and Wyoming played again, Tech would be favored again. Tech ran fairly bad to lose that game. I agree Wyoming isn't garbage, but if you play in the Cotton Bowl the way you did the first three quarters of that game, Texas will get their asses handed to them. 

  • Haha 1
Link to comment
Share on other sites

11 minutes ago, Frank Drebin said:

Spread was Texas -7.5 last year.

We were discussing this last year over 7am Fair parking lot booze and gummies and decided it had to mean Gabriel was playing. If you could have assured me he wasn't taking snaps, I would have bet literally every penny of disposable income I had on Texas. Beville is legit the worst QB I've ever seen start for OU and I went there and attended every home game for the Schnelly and Blake years. 

Link to comment
Share on other sites

1 minute ago, TXSooner518 said:

So you have figured out a "style of play" loophole in advanced metrics that Sagarin, Connelly, Vegas, etc., are all blind to and unable to correct? Please describe exactly how Wyoming's style of play creates this conundrum. Does it bake in that their starting QB missed the Texas game? Also, if Tech and Wyoming played again, Tech would be favored again. Tech ran fairly bad to lose that game. I agree Wyoming isn't garbage, but if you play in the Cotton Bowl the way you did the first three quarters of that game, Texas will get their asses handed to them. 

All of this is true.  I was just about to post the same thing that Tech would be favored against Wyoming if they played tomorrow.  Ricky needs to stop posting about things he doesn't understand

  • Hook 'Em 1
Link to comment
Share on other sites

12 minutes ago, Rickylovesweed said:

Now go look at yards per play in the USC and Colorado game and compare that to OU and SMU. 

I'm just not a believer at all in OU based on the SMU game. It was the only semi-decent opponent they've played and the yards and yards per play were very similar. This was an even game minus the turnovers and OU's special teams pinning SMU inside the 20 4 or 5 times. 

I'm not factoring any context of Texas into this. I'm just saying OU has played 1 decent opponent and it was a major struggle bus for them. 

I can understand this perspective. Our 2018 team struggled with Tulsa at home, but went on to beat a Kyler Murray-led OU team and play in the Big 12 Title Game. Sometimes a team does not always play up to their potential. It is also worth noting that SMU is not the pushover win that they were 10-20 years ago, either. They've poured some serious money into that program lately, and have upgraded their talent significantly. 

Link to comment
Share on other sites

4 minutes ago, Fumbles said:

You have to cherry pick the score at a certain time in one game to find an argument.  Weak.  See you Saturday.

Like you cherry picking Bama's ranking a month ago to argue out of your ass that Texas has the #1 SOS in the country? See you Saturday. 

Link to comment
Share on other sites

High stress games with amped up players and crowds increase to opportunity for emotion to help drive the game farther from the expected outcome in either direction. Clearly, either team could win. Injuries, fluke turnovers, and the like can move the result several TDs from where it "should" be. With that said, Texas played Alabama in a similarly amped up environment. Both teams took their best shots. Texas benefited neither from a preponderance of flukes nor other particular events that would move the result in our favor from where talent, skill, coaching, etc "should" have had it. We beat Alabama on their field by double digits, and they are a decent benchmark. Both Texas and OU have had good and bad quarters of football. Both are undefeated. Only Texas has been tested against an opponent remotely of the quality it will face this weekend, and Texas has passed the test. I can't predict the future, but my reason tells me covering by Texas is  likely.

  • Hook 'Em 5
  • Like 2
Link to comment
Share on other sites

4 minutes ago, Skipper said:

All of this is true.  I was just about to post the same thing that Tech would be favored against Wyoming if they played tomorrow.  Ricky needs to stop posting about things he doesn't understand

Of course Tech would be favored again. When did I deny that? 

My point is that Wyoming is not the 80th best team in the country. That will be obvious this weekend when they beat a really good Fresno State team. 

Link to comment
Share on other sites

Just now, Rickylovesweed said:

Of course Tech would be favored again. When did I deny that? 

My point is that Wyoming is not the 80th best team in the country. That will be obvious this weekend when they beat a really good Fresno State team. 

When you said "For instance, Texas Tech is still ranked above Wyoming in almost every advanced metric by a good bit even though Wyoming beat them."  If you agree Tech is better, then of course they should be higher in advanced metrics. 

Link to comment
Share on other sites

3 minutes ago, TXSooner518 said:

When you said "For instance, Texas Tech is still ranked above Wyoming in almost every advanced metric by a good bit even though Wyoming beat them."  If you agree Tech is better, then of course they should be higher in advanced metrics. 

No, I just don't take advanced metrics as the end all be all. At the end of the day, it's about wins and losses. 

When Tech is 20 or 30 spots higher than Wyoming in the advanced metrics (like ESPN efficiency) it's not about "who would be the favorite" it's just that Wyoming is probably undervalued. 

When Wyoming is 5-1 after this weekend with wins against a P5 team, a decent G5 team in App State, a really good G5 team in Fresno State, and a close loss to Texas I could give two shits if they are ranked outside the top 70 in advanced metrics. They're definitely better than that based on their resume. 

 

 

Link to comment
Share on other sites

Advanced metrics are models. The best model in the universe can't reliably extrapolate from the ability to beat bad teams to know whether you'll play well against a good team early in the season. There is literally a complete absence of the data of interest resulting in total dependence on guessing how to interpret proxy data which is known not to map closely. OU looks good to models that are analyzing turds for data. If OU had even 30 snaps on each side of the ball this year against a solid opponent, the computers would have at least something on which to chew. As it stands, those metrics are just guessing, as shown by the fact that expert humans all seem to agree that OU should be the underdog, but ESPN's machines disagree. The machines are not stupid; they lack useful data.

  • Hook 'Em 3
Link to comment
Share on other sites

2 minutes ago, Rickylovesweed said:

No, I just don't take advanced metrics as the end all be all. At the end of the day, it's about wins and losses. 

When Tech is 20 or 30 spots higher than Wyoming in the advanced metrics (like ESPN efficiency) it's not about "who would be the favorite" it's just that Wyoming is probably undervalued. 

When Wyoming is 5-1 after this weekend with wins against a P5 team, a decent G5 team in App State, a really good G5 team in Fresno State, and a close loss to Texas I could give two shits if they are ranked outside the top 70 in advanced metrics. They're definitely better than that based on their resume. 

 

 

Wait I had it on good authority Wyoming got the piss beaten out of them. @Fumbles get in here!  

And predictive metrics are exactly about who is better and who would win going forward, not about evaluating resumes. I'm waiting for that explanation about how Wyoming's style of play confounds all the systems though. 

Link to comment
Share on other sites

1 minute ago, Magus Ossis said:

Advanced metrics are models. The best model in the universe can't reliably extrapolate from the ability to beat bad teams to know whether you'll play well against a good team early in the season. There is literally a complete absence of the data of interest resulting in total dependence on guessing how to interpret proxy data which is known not to map closely. OU looks good to models that are analyzing turds for data. If OU had even 30 snaps on each side of the ball this year against a solid opponent, the computers would have at least something on which to chew. As it stands, those metrics are just guessing, as shown by the fact that expert humans all seem to agree that OU should be the underdog, but ESPN's machines disagree. The machines are not stupid; they lack useful data.

Most models have OU the slight underdog. And those models are used heavily in a multibillion dollar industry, it's pretty laughable that you think they are wholly unreliable. The biggest thing is of course that every team has a range of possible results, and so you get outlier outcomes when one team is at the far end of its range and the other is at the opposite end of theirs. Or individual fluky plays. But if OU and Texas could run this game a million times, the end result would be pretty close to 500k-500k. Tech-Wyoming pretty close to 800k-200k Tech. 

Link to comment
Share on other sites

3 minutes ago, TXSooner518 said:

I'm waiting for that explanation about how Wyoming's style of play confounds all the systems though. 

You don't think the style of play has an impact on certain predictive metrics/systems? 

Bill Connelly has personally admitted that SP+ (S&P back in the day) for whatever reason did not properly predict/evaluate Herman's teams. Even when he was at Houston. Ian Boyd and Bill Connelly had a long discussion about this a few years ago. 

  • Hook 'Em 1
Link to comment
Share on other sites

Just now, Rickylovesweed said:

You don't think the style of play has an impact on certain predictive metrics/systems? 

Bill Connelly has personally admitted that SP+ (S&P back in the day) for whatever reason did not properly predict/evaluate Herman's teams. Even when he was at Houston. Ian Boyd and Bill Connelly had a long discussion about this a few years ago. 

It's hard to code in turtling.

  • Haha 1
  • Rage+1 1
Link to comment
Share on other sites

1 minute ago, Rickylovesweed said:

You don't think the style of play has an impact on certain predictive metrics/systems? 

Bill Connelly has personally admitted that SP+ (S&P back in the day) for whatever reason did not properly predict/evaluate Herman's teams. Even when he was at Houston. Ian Boyd and Bill Connelly had a long discussion about this a few years ago. 

What, specifically, does Wyoming do and how, specifically, does it affect the metrics?

Link to comment
Share on other sites

The issue about all these models is that they are generally on a per play basis and the limited amount of data that football games produce. It’s not like baseball or basketball where there’s quite a few data points. That said, OU is the default fraud team from those ratings as they have had a joke of a schedule but has performed pretty well. They’ll drop like a rock when they face actual completion (ie look at their defensive rating this week versus last week). 
 

they are no different than the Mizzou or Ole Miss of years past that blitz a shit OOC schedule and then drop once they play teams with a pulse. 

Link to comment
Share on other sites

6 minutes ago, TXSooner518 said:

Most models have OU the slight underdog. And those models are used heavily in a multibillion dollar industry, it's pretty laughable that you think they are wholly unreliable. The biggest thing is of course that every team has a range of possible results, and so you get outlier outcomes when one team is at the far end of its range and the other is at the opposite end of theirs. Or individual fluky plays. But if OU and Texas could run this game a million times, the end result would be pretty close to 500k-500k. Tech-Wyoming pretty close to 800k-200k Tech. 

Straw man alert. I didn’t say the models were useless. I said they were NOT stupid but rather were relying on poor proxy data to model a phenomenon in the instance of early season predictions against good teams before playing any good teams. I gave the specific example of negative correlation between the multibillion dollar industry you reference and one of the models.

Link to comment
Share on other sites

9 minutes ago, gmr548 said:

It's hard to code in turtling.

Yes, which is why the advanced metrics also probably don't like Wyoming. 

They fall on the ball offensively and let their defense win games. It's the reason Connelly said Herman's teams were tough to evaluate/predict. Herman was in a lot of close games with bad teams, beat down several good teams, etc. 

The turtling against bad teams (even though they were in control of a lot of those games) made Herman difficult to evaluate under S&P. That's straight from Connelly. 

Link to comment
Share on other sites

17 minutes ago, closetojumping said:

For all of the advanced stats outputs, are any of them incorporating things like injuries? Suspensions? Prior performance against the same opponent per player? I don't know how deep some of it gets.

 

Not for SP+ 

Connelly is pretty open about "loopholes" in his system which is why the @TXSooner518 discussion above is funny to me. Connelly has discussed multiple times that "style of play" loopholes definitely impact SP+. 

 

Link to comment
Share on other sites

17 minutes ago, Rickylovesweed said:

Not for SP+ 

Connelly is pretty open about "loopholes" in his system which is why the @TXSooner518 discussion above is funny to me. Connelly has discussed multiple times that "style of play" loopholes definitely impact SP+. 

 

What is it about Wyoming’s style of play that has them underrated in SP+? 

Link to comment
Share on other sites

Just now, Constant said:

What is it about Wyoming’s style of play that has them underrated in SP+? 

I think it was in the thread that talked about Herman’s teams overperforming… grind clock, run the QB, do well on 3rd, 4th, and red zones (offense and defense). 

It read like being a grinding, hard out wasn’t reflected well in his algorithm. 

Link to comment
Share on other sites

7 minutes ago, Constant said:

What is it about Wyoming’s style of play that has them underrated in SP+? 

IMO it's very similar to Herman's teams. 

They win ugly against bad teams. See Portland State and New Mexico. They take very few risks on offense (I.e. Herman's turtling) and defensively they keep the ball in front. That style of play wins games but it also keeps bad teams in the game and doesn't look good from an advanced metrics perspective. 

 

 

Edited by Rickylovesweed
Link to comment
Share on other sites

3 minutes ago, Constant said:

What is it about Wyoming’s style of play that has them underrated in SP+? 

If this is a serious question, their approach to Texas was to take the air out of the ball and limit possessions. It worked. Texas ran like 57 plays on only 7 possessions. 

Obviously, with Texas, the purpose was to preserve some pride. In Laramie, the approach works better against comparable talent. Get the opponent into a mud wrestling contest and wait for them to run out of breath in the back half of the game. 

It's ugly as sin, but they're a hell of a good cook. That Wyoming team probably finishes 8-4 or better this year and Bohl has them going to bowl games regularly. 

  • Hook 'Em 2
Link to comment
Share on other sites

1 minute ago, closetojumping said:

If this is a serious question, their approach to Texas was to take the air out of the ball and limit possessions. It worked. Texas ran like 57 plays on only 7 possessions. 

Obviously, with Texas, the purpose was to preserve some pride. In Laramie, the approach works better against comparable talent. Get the opponent into a mud wrestling contest and wait for them to run out of breath in the back half of the game. 

It's ugly as sin, but they're a hell of a good cook. That Wyoming team probably finishes 8-4 or better this year and Bohl has them going to bowl games regularly. 

I didn’t see any of the game with it being on LHN. 

Link to comment
Share on other sites

2 minutes ago, Constant said:

I didn’t see any of the game with it being on LHN. 

I genuinely recommend watching their game against Fresno State this Saturday. It will be a night game in Laramie, airing on Fox nationally. If you're a bit of a sicko it is great television. 

  • Like 2
Link to comment
Share on other sites

1 minute ago, Constant said:

I didn’t see any of the game with it being on LHN. 

It was a boring, if predictable, game. Texas coming in off of the Bama game, Wyoming being tough and plucky with excellent coaching, it ended up about how I figured. I assumed Texas would cover, but a fumble in the 4th while marching down the field ruined that. 7 possessions in an entire game is pretty wild. Wyoming ran like a 20 play drive to soak up almost all of the 3rd quarter by picking up 3 yards each down.

Link to comment
Share on other sites

3 hours ago, TornACL said:

My biggest takeaway from the OU Cincy game is that Gabriel will make insanely bad decisions when pressured.

Not saying we should bring the house, but I am saying that when we get pressure, Gabriel is gonna throw at least a couple of idiotic passes that need to be picked off.

And taken back for 6

And when he gets up to make the tackle Collins will find him

Link to comment
Share on other sites

1 hour ago, closetojumping said:

For all of the advanced stats outputs, are any of them incorporating things like injuries? Suspensions? Prior performance against the same opponent per player? I don't know how deep some of it gets.

 

Nope. Reason number a million why advanced metrics in football are largely pointless. 

Link to comment
Share on other sites

It doesn't feel like TX OU weekend with all this nerd talk. Where are the poorly constructed and poorly conceived ms paints and animations? 

I am here to see past legends of the rivalry being sodomized by various farm animals, not to discuss advanced analytics and models.

  • Hook 'Em 2
Link to comment
Share on other sites

2 minutes ago, Rickylovesweed said:

They aren't worthless but they also aren't an exact science. 

 

I said “largely” pointless. There’s way too many variables and way too little of a sample size for a lot of useful date to be learned from it. It’s not baseball or basketball or hockey. 

  • Hook 'Em 1
Link to comment
Share on other sites

  • Pancho locked this topic
Guest
This topic is now closed to further replies.


×
×
  • Create New...