Make me Laugh. Try to explain SOS - Strength of Schedule

[4down20]

Well, it's simple for anyone who passed 6th grade math. Seeing that you're a Bamer, I'm understand that you'd think it's super complex.

I bet you haven't even seen an actual SOS figure. Which makes me lol.

#UCFacts

I've made my own ranking and SoS formula and probably know more about the topic than just about anyone around here. There is no "simple math" to good ones, there is a process that has to be done with many steps and lots of complex math.

But go ahead and name a SoS formula and I'll tell you the flaws in it. You want a simple math one like win% based? Those are terrible because they treat all teams with the same record as being equal, without regard to who they've beat. A slightly better version of that is a cumulative win% which takes opponent records into account, but it still suffers from the same things.

A good SoS method will look at things like rushing/passing yards for offense/defense etc. And rushing for 100 yards in a game vs Alabama and rushing for 100 yards vs UTEP are quite different. So you normalize that data a bit and compare the stats against each other to get what I call "quality stats", which recognize the power differences of the 2 defenses. Based on this data you can start to get a real understanding of team strengths.

And then from there you can treat that data in a few different ways. I personally just took the average scores of those teams and that was the SoS. However as I already pointed out earlier in this thread, that method is flawed. Since then I've found another SoS that instead looks at predicted losses if an "elite team" were to play that schedule. This method is much better and doesn't suffer the problems of averages and if I were to start doing my own formula again I would switch to that. Most formulas just use averages.

You would basically never do these things by hand or with simple math because comparing every single game and every single team and all the stats would probably take even NASA weeks to complete. As such, if you really want to question me - don't ask me to "show the math" you can start by asking me what language I used.

But it's not like you're really looking for real answers here. Lucky for you, I like talking about the topic.

 

[TheDayMan]

It's UCFunny that tshirt bandwagoners like you talk about, and
cite something that you can't explain nor understand. Lemmings.

But I’ll give you another chance to redeem yourselves. Define what SOS and show the math.

#UCFacts

I’ll sum it up real nice for you.

See UCF’s schedule.

#UCFuckyourself

 

[TheDayMan]

Hashtags are for 16yo girls or soccer moms trying to sell their essential oils.

And #UCFags

 

[Voltaire26]

I'll hang up and listen....

#UCFacts

Strength of Schedule is that thing you need to win one of these (not make your own).

 

[UCFhonors]

So you normalize that data a bit and compare the stats against each other to get what I call "quality stats", which recognize the power differences of the 2 defenses. Based on this data you can start to get a real understanding of team strengths.

Normalizing data is just a UCFancy way to say, interject bias into data.

Are you using Python or R? Or what language?

#UCFacts

 

[Bullitt08]

What Opie looks like when he sees a dick?

 

[Wamu]

Strength of Schedule is that thing you need to win one of these (not make your own).

Did you know Coach Khaki Pant passes out Spring Game Championship rings to the winner of the maze 'n blew game each year? They're sweet!

 

[Olyduck]

I’ll give you a C+ for trying. And B for your thinking being on the right track.

There is actually a really simple and basic answer to the OPie.

#UCFacts

is it really that simple tho? I have watched college football going on 29 years or so and its still a metric not everyone agrees on.

 

[Wamu]

@UCFhonors


I can make you a national title trophy out of macaroni shells & paint it UCF colors. Only $19.95 + shipping & handling.

 

[socaljim242]

Easy peasy
UCF this season plays

Tulsa
Florida A&M
Connecticut
Florida Atlantic
East Carolina
Temple
Tulane
South Florida


This means your Schedule Overall Sucks and even if you win all your games . It still sucks.

 

[UCFhonors]

is it really that simple tho? I have watched college football going on 29 years or so and its still a metric not everyone agrees on.

No stat can or ever will be able to predict games. That why I think it's UCFoolish for people to over complicate stats - its the Law of Diminishing Returns.

#UCFacts

 

[UCFhonors]

Easy peasy
UCF this season plays

Tulsa
Florida A&M
Connecticut
Florida Atlantic
East Carolina
Temple
Tulane
South Florida


This means your Schedule Overall Sucks and even if you win all your games . It still sucks.

Temple and usf have been better than USC for a while now.

But it's clear you think SOS is about name recognition or what you did in the 1980s, like most Uncle Ricos

#UCFacts

 

[socaljim242]

Temple and usf have been better than USC for a while now.

But it's clear you think SOS is about name recognition or what you did in the 1980s, like most Uncle Ricos

#UCFacts

Shows how much of an idiot you are. Last time Temples finished ranked was 17th in 1979. USF finished ranked between 19 and 21 in 2016-17-18.

USC finshed ranked - 24 times since Temples only ranking and more than twice USF at 7 times in the past decade of Kiff Sark Ed O and Clay. So in a horrible stretch USC is still way better than those two. No wonder you can't comprehend SOS . You don't have a clue that your team plays a majority of stiffs.

 

[4down20]

Normalizing data is just a UCFancy way to say, interject bias into data.

Are you using Python or R? Or what language?

#UCFacts

All teams are treated equally and the data is normalized to remove bias.

The first version I did was written in C, the last version I did was a web version and was written in Perl.

 

[Voltaire26]

Did you know Coach Khaki Pant passes out Spring Game Championship rings to the winner of the maze 'n blew game each year? They're sweet!

View attachment 201989

And like the UCF Ring, they don't mean a damn thing ... win or shut the fuck up, so I will keep quiet!!!

 

[B_dub]

@UCFhonors


I can make you a national title trophy out of macaroni shells & paint it Oregon colors. Only $19.95 + shipping & handling.

^^^^^
@OregonDucks
Get in on this shit quick!

 

[UCFhonors]

Shows how much of an idiot you are. Last time Temples finished ranked was 17th in 1979. USF finished ranked between 19 and 21 in 2016-17-18.

USC finshed ranked - 24 times since Temples only ranking and more than twice USF at 7 times in the past decade of Kiff Sark Ed O and Clay. So in a horrible stretch USC is still way better than those two. No wonder you can't comprehend SOS . You don't have a clue that your team plays a majority of stiffs.

Rankings are BS click bait. How many times in the last few years did Temple and USF win more games than USC?

#UCFacts

 

[CJH9972]

SOS can be defined any number of ways. The problem is that college football hasn't picked one method to play by allowing teams, especially G5 teams, to schedule accordingly. Ultimately, it's greatest influence seems to be about eliminating G5 teams outright while giving them no blueprint of what kind of schedule is good enough.

 

[Wamu]

^^^^^
@OregonDucks
Get in on this shit quick!

 

[Wamu]

Rankings are BS click bait. How many times in the last few years did Temple and USF win more games than USC?

#UCFacts

Wait, what, huh? You should have looked some info up before you said this. So how many times in the last few years did Temple & USF win more games in a season than USC? Going back 3 seasons since you said in the last few years.

Temple's done that once. Last year. 8 W's for the Owls & 5 W's for the Trojans.
USF's done that once. In 2016. USF won 11 games & USC won 10.

#check-yo-facts