USA Today FTW: The Main Character article, Rivalry Week edition

Submitted by crg on November 29th, 2021 at 12:00 PM

Jason Kirk's article now the he's over at USA Today (basically his take on the former TWIS).  This week's edition has some good stuff... especially near the end.

Enjoy!

https://ftw.usatoday.com/2021/11/college-football-results-rivalry-week

Bonus:  link to team confrontation in tunnel at half time https://twitter.com/BleacherReport/status/1464670028012220416?ref_src=twsrc%5Etfw%7Ctwcamp%5Etweetembed%7Ctwterm%5E1464672777424846866%7Ctwgr%5E%7Ctwcon%5Es3_&ref_url=https%3A%2F%2Fftw.usatoday.com%2F2021%2F11%2Fcollege-football-results-rivalry-week

JHumich

November 29th, 2021 at 1:57 PM ^

WOW IF YOU HAVE GOTTEN THIS FAR WITHOUT CLICKING THE LINK, CLICK IT IMMEDIATELY, AND SCROLL DOWN TO THE OSU-MICHIGAN SECTION AND READ IT

I know, I know. Just broke like three internet rules.

But this may be THE GREATEST THING I HAVE EVER READ ON THE INTERNET.

Rules don't apply.

Just go read it.

Seriously.

Then, come back, and undo the neg you gave me for pleading with you in such a manner, because you will then know that it was worth it.

Brhino

November 29th, 2021 at 12:18 PM ^

Is it just me, or do teams seem to lose these "99% chance of victory at this time... oh... oh no.  oh noooo." games way more often than a 99% would tend to indicate? 

wile_e8

November 29th, 2021 at 12:43 PM ^

There has to be a better term for it that I can't think of right now, but I think this is a form of survivorship bias. You don't hear about the vast, vast, majority of "99% chance of victory" games because the teams with 99% chance of victory end up winning comfortably. But thousands and thousands of sporting events every year means things with only a 1% chance of occurring happen regularly. And those are the ones you hear about. 

Brhino

November 29th, 2021 at 12:57 PM ^

I get what you're saying, but it still seems like a lot.  I would further separate 99% Chance to Win games into two categories:

1. Team A is up 35 points on Team B at halftime.  Yeah...

2. Team A is up 15 points on Team B with 8 minutes left.

And yeah, for category 1 games, Team A wins 100% of the time, or 9.99%.  But those other games, I don't know.  I'd love to see the numbers on those sort of still-vaguely competitive games, especially when ranked teams are involved.

For our game, ESPN says we hit a 99% chance to win when we were up 15 and OSU turned a 3rd and 14 into a 4th and 7 at our 24 yard line. OSU picked up 4th and 7, found themselves in 3rd and 20 on the next set of downs, turned that into 4th and 5 and then a touchdown, and it was back to a one-score game. I don't know about you, but it didn't feel like a 99% chance to win to me until HH hurdled one last fool.  Maybe we've just been bitten so often that we're gunshy, I don't know.

tee wrecks

November 29th, 2021 at 1:22 PM ^

On your point about the first type of game, when OSU was up on MSU 49-0 at halftime last week, ESPN's win probability metric showed that OSU had a 99.7% chance of winning.  Where, exactly, does that other 0.3% exist in the data and what would you have to do to get to 99.9% probability by halftime?  The metric suggests that MSU had about a 1-in-333 shot at coming back, which seems to drastically overstate the actual chances.  No team in college or pro football history has ever come back from more than a 35 point deficit at any point in any game.

jmblue

November 29th, 2021 at 1:29 PM ^

And then another issue is, even if some team had come back from 49 down, was it comparable to MSU and its opponent comparable to OSU?

To go back to the OSU example, being up 15 midway through the 4th quarter might well be a 99% chance against Northwestern, but against OSU?

Yeoman

November 29th, 2021 at 1:25 PM ^

99% is how I felt about it. How was OSU ever going to get a stop when they could barely get to second down?

To me it felt over after the Cameron Brown fracas. If I could have seen what Herbstreit saw in the tunnel I might have gotten there a little sooner.

wile_e8

November 29th, 2021 at 1:41 PM ^

You're just projecting your biases as a Michigan fan onto one anecdote. Sure, after years of heartbreak in this game, it was extremely hard to be confident in a win until the final whistle actually blew. But the entirety of the data shows teams in that same situation winning 99% of the time. It's just that you were fearful we were going to be the unlucky 1%. 

jmblue

November 29th, 2021 at 1:16 PM ^

That could be, but some of the math doesn’t seem to check out.  Auburn had a 99.7% chance to win a seven-point game with 1:31 remaining?  When it only takes one play to tie and there are multiple plays left in the game (and Auburn needed to gain another first down to ice the game at that point), it doesn’t seem like a 3 in 1000 chance.

Brhino

November 29th, 2021 at 1:29 PM ^

It wasn't just Auburn. There were 3 games involving top-25 teams last weekend in which one team had a 99%+ to win but didn't:

Texas A&M lead LSU by 4 with 1:30 left, LSU facing 3rd and 11 at their own 14. TAMU had a 99.7% chance to win.  Instead, LSU drove the field and scored the game winning touchdown.

Auburn, with a 3rd and 1 in Bama territory, a 7 point lead, and less than 2 minutes left, was at 99.8% chance to win.  We saw how that worked out.

UNC had a 99.9% chance to win when, up by 9 with 2 minutes left, they sacked NC State.  NC State went 64 yards in the next play to score, onside kick, scored again, and then a defensive stop for the win.

So last weekend, in the 23 games involving top 25 teams, we had 3 examples of teams pulling off 1-in-300 odds to win.  Either it was one of the craziest weekends ever, or these numbers are bunk.

wile_e8

November 29th, 2021 at 1:49 PM ^

Anecdote vs. data. With 1:31 left in the Iron Bowl, a team that had scored 3 points all game needed to drive 97 yards for a touchdown with no timeouts. While you're focusing on the one time a team actually pulled it off, you're ignoring all the extremely more numerous similar examples where the underdog failed with a whimper. 

jmblue

November 29th, 2021 at 2:04 PM ^

But Alabama isn’t just any team. If these odds are based on a generic team being in that situation, that’s misleading.  Alabama is, at a bare minimum, one of the three most talented teams in the country and its odds of winning any given game are going to be greater than average.

As for them scoring only 3 points to then, that’s true, but in that situation you’re also going to use all four downs, and you’re going up against a defense that has played nearly four full quarters and is probably a bit fatigued.  And you’re still Bama.

The Mad Hatter

November 29th, 2021 at 12:44 PM ^

This guy nailed it...

"Michigan beat Ohio State exactly how Michigan daydreams of beating Ohio State."

It wasn't just that we won, which was sweet enough, but that we pushed them around all day like it was nothing, running an offense that looks and feels like a Michigan offense.

OSU has become a pussy ass air-raid team, and if they keep it up we're beating them again next year.

 

Tauro

November 29th, 2021 at 12:49 PM ^

My favourite line of the article:

 

In the fourth quarter, as Ohio State kept moving within a single score, there was nevertheless chatter about storming the field. After all, that’s what Michigan spent the whole game doing.

MGoGrendel

November 29th, 2021 at 1:15 PM ^

There was something just a little bit different from the get-go. Michigan fans are known for their wind-battered panic mode, their instant turtling at the onset of sports traumas after enduring such a disproportionate load of them in the past. Life has taught Wolverines to flinch at the first sign of football trouble, lament what has not even gone wrong yet, and bury their emotions in a basketball program they somewhat care about, because those losses hurt less.

Bingo!