re: College Football Historical Relevancy Rankings

quote:

---

This is just hard to understand.

---

I’m sure this happens to you a lot, CatfishJohn

quote:

---

I think you need to rework your formula because it's pretty damn hogwash if it has Auburn ahead of LSU.

---

This is not a ranking of head-to-head meetings. This is not a ranking of which program is better or more accomplished. This is a ranking of how prone programs have been at experiencing stretches of irrelevancy as defined as having a football season in which they finished outside the AP rankings.

As noted in the OP, LSU had experienced four separate LOOONNNGGG stretches of 7+ seasons of finishing unranked. Auburn has one such stretch, and only two stretches that lasted longer than four seasons.

This data is basically saying that Auburn has historically been more likely to avoid long droughts of irrelevancy, lasting 5 or more years, as compared to LSU. And the numbers undoubtedly prove that.

quote:

---

This should probably be adjusted. FSU didn't field a football team until 11 years after the inception of the AP poll. It's hard to be relevant before you even exist.

---

I would counter that by pointing out there is no truer definition of irrelevancy as being non-existent.

quote:

---

This is not a ranking of head-to-head meetings. This is not a ranking of which program is better or more accomplished. This is a ranking of how prone programs have been at experiencing stretches of irrelevancy as defined as having a football season in which they finished outside the AP rankings.

---

I would go with weeks inside the AP instead of season ending in the AP. There are teams that are very relevant for a season but have a few tough losses and end up outside the AP 25. And vice versa, there are teams on no ones radar that squeak into the AP at the end of the season. Weeks are a more concise measurement of time. And even better might be doing this while removing first couple of weeks to wash away some preseason bias, although one could argue that media bias and publicity are indeed measurements of relevancy (I'm not arguing that necessarily)

quote:

---

I’m sure this happens to you a lot, CatfishJohn

---

Thanks

quote:

---

This data is basically saying that Auburn has historically been more likely to avoid long droughts of irrelevancy, lasting 5 or more years, as compared to LSU. And the numbers undoubtedly prove that.

---

So that is what this is telling you. Not overall historical relevancy, it is about who had longer droughts of being ranked at the end of the year. If a team spends more weeks in the AP poll and finishes more seasons ranked in the AP poll, wouldn't that mean they've been relevant historically more often?

quote:

---

Does a 7-5 Ohio State or Alabama finish ranked no matter what? Does a 7-5 Ole Miss get the same evaluation?

---

We all know the answer to that one.

It's like Notre Dame. They can go 0-12 and be ranked in the preseason top 10 the following season.

quote:

---

I would go with weeks inside the AP instead of season ending in the AP.

---

This falls into the same argument of "ranked opponents at game time" vs "ranked opponents at end of year". LSU fans typically favor the latter argument.

quote:

---

So that is what this is telling you. Not overall historical relevancy, it is about who had longer droughts of being ranked at the end of the year.

---

Yes, pretty sure that was clear from his original post.

Regarding "overall historical relevancy", that's more opinion than something that can be determined analytically.

Which team is more historically relevant overall:
Nebraska or Michigan

Depending on your criteria, you may end up with two different answers.

quote:

---

LSU fans typically favor the latter argument.

---

News to me...

quote:

---

Yes, pretty sure that was clear from his original post.

---

His OP is contradictory.

He says this is a

quote:

---

College Football Historical Relevancy Rankings

---

And

quote:

---

This is not a ranking of the “Best” or “Most Accomplished” programs. This is a ranking of how common it is for a program to be relevant at any given point since the inception of the AP Poll (1936).

A team is considered relevant for each season if they finished ranked in the last AP Poll.

---

So he defines relevancy equals # of seasons ending ranked in AP poll.

But then says relevancy in his model is not that. It is actually

quote:

---

The below methodology is based on the standard deviation of season streaks of finishing outside the Final Poll.

---

So he is not using his original definition.

The definition of relevancy to him is streaks of being ranked, not how many times you've ranked or how many weeks youv'e been ranked or the timespan of ranked seasons.

The other major issue is that relevancy should have several other factors. There are weights to relevancy e.g. how highly you're ranked, hardware, notable talent, etc.

I don't give a shite if someone did an analysis and LSU is outside the top 50 if I agreed with the methodology and the result passed even an initial smell test. Who cares.

quote:

---

It’s happened 20 times since the 1990s.

---

That's most years though, right?

quote:

---

I would put LSU ahead of TN. Other than that, it looks pretty accurate

---

Well, that would mean you don't know how math works.

quote:

---

8. Penn State- 1.53

---

quote:

---

WAY too high,

---

Did you even read how it was calculated?

I’m going to take this opportunity to say that since SEC football was integrated, Ole Miss has had just 5 10 win seasons and 1 season with fewer than 3 losses. 2023.