Started By
Message
re: Updated ESPN FPI - National Rank and SEC Ranks
Posted on 7/8/19 at 1:14 pm to Farmer1906
Posted on 7/8/19 at 1:14 pm to Farmer1906
quote:
How are there 10 tougher schedules?
Because like many dumb statistical analysis these days they just take a total average, and that total average includes #109 Texas State, #120 UTSA and 1AA Lamar. A&M's "average" ranking per their 12 opponents is 42.8. I'm guessing some other teams have their bottom 1/3 of the schedule filled with #78, #85, #87 or something.
And dumb schedule rankings ALWAYS just do mass averages, which are ridiculous because they treat the difference in #67 and #87 the same as #3 and #23.
Posted on 7/8/19 at 1:14 pm to KaiserSoze99
We literally play a third of our schedule vs better competition than their toughest game.
Posted on 7/8/19 at 1:15 pm to Farmer1906
quote:
A&M's Schedule:
1. Clemson (12.3-0.7) (88.2%) (29.2)
2. Alabama (11.0-1.6) (46.0%) (27.9)
3. Georgia (9.9-2.7) (27.1%) (22.3)
4. LSU (9.5-2.8) (11.4%) (21.7)
9. Auburn (7.7-4.4) (2.1%) (16.8)
16. Mississippi St (7.7-4.3) (1.2%) (13.7)
18. South Carolina (6.1-5.9) (0.9%) (12.3)
41. Ole Miss (5.8-6.2) (0.0%) (6.1)
60. Arkansas (4.8-7.2) (0.0%) (0.5)
How are there 10 tougher schedules?
Perhaps because the other 10 schedules have 12 games on them.
Posted on 7/8/19 at 1:16 pm to SummerOfGeorge
quote:
Alabama 74.8% vs LSU
Alabama 73.6% @ Auburn
Alabama 74.3% @ Texas A&M
Very interesting.
Posted on 7/8/19 at 1:18 pm to KaiserSoze99
quote:
How can this be?
See my previous post. Mass averaging.
USC Average Opponent : #34
Texas A&M Average Opponent : #43 (assigning Lamar #130)
It's stupid, but that's how almost every dumb arse "ranks" SOS. In reality, anything below #50 should be counted as an almost auto win. If I'm doing schedules, I want to see how hard your hardest games are and how many of them you have.
Average of 8 hardest games
Texas A&M : #12
USC : #28
Not even close. Now, with USC's schedule, you could definitely argue that they really don't have any total layup games. Their 2 easiest non conference games are a Fresno team that won 10 last year and BYU. But yea, it doesn't make sense to average that stuff together.
This post was edited on 7/8/19 at 1:22 pm
Posted on 7/8/19 at 1:20 pm to SummerOfGeorge
The moral of the story:
Play average teams only.
Play average teams only.
Posted on 7/8/19 at 1:22 pm to MykTide
quote:
Alabama 74.8% vs LSU
Alabama 73.6% @ Auburn
Alabama 74.3% @ Texas A&M
Very interesting.
I too find it interesting that no one is interested in:
Alabama 88.4% vs Tennessee
Posted on 7/8/19 at 1:23 pm to KaiserSoze99
quote:
The moral of the story:
Play average teams only.
A lot of basketball programs did exactly that to game the RPI for a decade (specifically the Missouri Valley).
Posted on 7/8/19 at 1:24 pm to WildTchoupitoulas
quote:
I too find it interesting that no one is interested in:
Alabama 88.4% vs Tennessee
I mean, I'm interested in kicking their arse, but I don't think anyone would be interested in listing out expected win %s at almost 90%.
Posted on 7/8/19 at 1:26 pm to SummerOfGeorge
quote:
Because like many dumb statistical analysis these days they just take a total average, and that total average includes #109 Texas State, #120 UTSA and 1AA Lamar. A&M's "average" ranking per their 12 opponents is 42.8. I'm guessing some other teams have their bottom 1/3 of the schedule filled with #78, #85, #87 or something.
And dumb schedule rankings ALWAYS just do mass averages, which are ridiculous because they treat the difference in #67 and #87 the same as #3 and #23.
Hey, someone gets it!
I use these SoS metrics because they don't use averages at all. Instead it's based on the likelyhood of what teams of a certain level are expected to do against the opponents.
LINK /
The only thing is that after the regular season they kind of move a good bit due to bowl games. Alabama ended up #1 only after it played Georgia, Oklahoma and Clemson.
Clemson was #100 in this last year until the playoffs, then jumped up to #21 as a result of playing Notre Dame and Alabama.
Which is good IMO because those do actually end up begin the toughest schedules. If you play the SEC West and then Georgia, Oklahoma and Clemson you definitely have played the toughest schedule. But I wish it had regular season metrics separate, so you could get an accurate picture of what they did in the season. You can live, but not after the fact.
UCF was almost dead last until their conference title game and the LSU game, which brought them all the way up to....#100.
All 14 of the SEC teams are in the top17.
This post was edited on 7/8/19 at 1:30 pm
Posted on 7/8/19 at 1:26 pm to SummerOfGeorge
quote:
I mean, I'm interested in kicking their arse, but I don't think anyone would be interested in listing out expected win %s at almost 90%.
C'mon, George, isn't the TSIO the biggest game of the year for Alabama? Seems like people should be interested no matter what the odds. After all, this is a Big Rivalry game, anything can happen.
Posted on 7/8/19 at 1:29 pm to WildTchoupitoulas
quote:
After all, this is a Big Rivalry game, anything can happen.
We're Alabama, the entire point of Saban's existence is to eliminate the notion that anything can happen
Tennessee probably won't be ranked for our game and that means that they would have to be the unranked first team in over a decade to beat us. And do it in Tuscaloosa. Odds don't seem great.
Still can't wait to whoop their arse again, though.
Posted on 7/8/19 at 1:31 pm to WildTchoupitoulas
quote:
After all, this is a Big Rivalry game, anything can happen.
It can.
LSU fans should know better than anyone.
Posted on 7/8/19 at 1:33 pm to SummerOfGeorge
I like Tennessee big in that one.
Posted on 7/8/19 at 1:35 pm to MykTide
quote:Are you referring to the game commemorated by this fine china from the Franklin Mint...?
It can.
Posted on 7/8/19 at 1:35 pm to SummerOfGeorge
Interesting that Tenn and SC get bumps for %chance at conference title, and Mizzou gets 0% for the category. I suppose that's from the post season ban??? It would have to be given our schedule is so much easier than both of theirs.
Posted on 7/8/19 at 1:36 pm to ZOUtiger
quote:
I suppose that's from the post season ban???
I was wondering the same but yes, that has to be it.
Posted on 7/8/19 at 1:36 pm to SummerOfGeorge
I think comparing strength of schedule is dumb especially within the division. We play 5 common oppenents, play each other, 2 cross division games, 1 FCS game, 2 group of 5 games and a Power 5 OOC game. Everybody plays that schedule. The OOC schedules are made years in advance. So, nobody really know how good these teams are going to be in 5 years.
Posted on 7/8/19 at 2:02 pm to SummerOfGeorge
It's such horse shite.
A&M's average for top 2 games = 1.5
USC = 8.5
Top 3 game average:
A&M = 2
USC = 11.33
Top 4 game average:
A&M = 2.5
USC = 13.5
Top 5 game average:
A&M = 5.2
USC = 15.2
Top 6 game average:
A&M = 7.3
USC = 18.16
Using that metric, the shitty to mediocre FudgePac12 has the following SOS rankings:
2. USC (top 3 average = 11.33)
4. Stanford (top 3 average = 11.33)
6. Colorado (top 3 average = 15.66)
8. Oregon State (top 3 average = 15.66)
9. UCLA (top 3 average = 17.66)
A&M's average for top 2 games = 1.5
USC = 8.5
Top 3 game average:
A&M = 2
USC = 11.33
Top 4 game average:
A&M = 2.5
USC = 13.5
Top 5 game average:
A&M = 5.2
USC = 15.2
Top 6 game average:
A&M = 7.3
USC = 18.16
Using that metric, the shitty to mediocre FudgePac12 has the following SOS rankings:
2. USC (top 3 average = 11.33)
4. Stanford (top 3 average = 11.33)
6. Colorado (top 3 average = 15.66)
8. Oregon State (top 3 average = 15.66)
9. UCLA (top 3 average = 17.66)
Posted on 7/8/19 at 2:24 pm to KaiserSoze99
quote:
9. UCLA (top 3 average = 17.66)
UCLA you say?
Popular
Back to top
Follow SECRant for SEC Football News