The USC effect

[walshtrips]

Seton Hall's NET was 66 prior to the USC's victory yesterday. We had two Quad 3 losses yesterday morning and have only one this morning, as USC's win moved them from a NET of 105 to a NET of 97. Despite going from two Quad 3 losses down to just one, our NET didn't budge.

These metrics make no sense to me.

 

[LeftCoastPirates]

Seton Hall's NET was 66 prior to the USC's victory yesterday. We had two Quad 3 losses yesterday morning and have only one this morning, as USC's win moved them from a NET of 105 to a NET of 97. Despite going from two Quad 3 losses down to just one, our NET didn't budge.

These metrics make no sense to me.

Because the computer doesn’t evaluate your Quad 1/2/3/4 records. Those buckets are fun categorization tools to carry a general conversation as to where the general quality of your wins have come from.

But make no mistake about it, at the end of the day when the committee of human individuals sits around a table to discuss your team sheet. A win vs the number 3 ranked NET team (UConn) when compared to a road win vs number 59 (Providence) are NOT equivalent to each other because they are classified as “Quad 1.”

The fact that I even have to explain that is troubling.

The computer is still evaluating USC towards your strength of schedule, as only 8 spots better than where it stood before, and is only a single piece of data in a total context that now spans 24 games worth of results are just as equally weighted.

I understand most people didn’t like math in school, still don’t like math today, and call people who do like math nerds. Well the nerds on this board have been trying to explain the math to the masses for quite sometime. Math doesn’t factor in opinions, bias, perceptions, or prejudice. It smashes all blue tinted glasses. Other posters have even shared the link to the NCAA’s publication of what goes into the math. The only thing the NCAA haven’t shared is the exact formula.

Seton Hall was a bad basketball team over the course of its first 9 games Seton Hall then became a very good team over the next 13. So when you fail the first part of the exam and they score well (not perfect) on the second part of the exam, you don’t average out to an A or B. You’re still somewhere in the middle.

And that’s still what our NET number reflects. Regardless of USC winning to get them to 9-13 over an Oregon State team that is tied with them for last place in the PAC 12 because it moved our loss to a Q2 game. 🤦‍♂️ We need to stop with this type of baseless complaining.

 

[HallX2]

Because the computer doesn’t evaluate your Quad 1/2/3/4 records. Those buckets are fun categorization tools to carry a general conversation as to where the general quality of your wins have come from.

But make no mistake about it, at the end of the day when the committee of human individuals sits around a table to discuss your team sheet. A win vs the number 3 ranked NET team (UConn) when compared to a road win vs number 59 (Providence) are NOT equivalent to each other because they are classified as “Quad 1.”

The fact that I even have to explain that is troubling.

The computer is still evaluating USC towards your strength of schedule, as only 8 spots better than where it stood before, and is only a single piece of data in a total context that now spans 24 games worth of results are just as equally weighted.

I understand most people didn’t like math in school, still don’t like math today, and call people who do like math nerds. Well the nerds on this board have been trying to explain the math to the masses for quite sometime. Math doesn’t factor in opinions, bias, perceptions, or prejudice. It smashes all blue tinted glasses. Other posters have even shared the link to the NCAA’s publication of what goes into the math. The only thing the NCAA haven’t shared is the exact formula.

Seton Hall was a bad basketball team over the course of its first 9 games Seton Hall then became a very good team over the next 13. So when you fail the first part of the exam and they score well (not perfect) on the second part of the exam, you don’t average out to an A or B. You’re still somewhere in the middle.

And that’s still what our NET number reflects. Regardless of USC winning to get them to 9-13 over an Oregon State team that is tied with them for last place in the PAC 12 because it moved our loss to a Q2 game. 🤦‍♂️ We need to stop with this type of baseless complaining.

Why does it trouble you? Why not just explain it and leave the snark out?

 

[Piratz]

Seton Hall’s metrics and resume were very bad in the non-conference. This is going to plague us all year long and we better not pull in on the Bubble with that black mark. The wins over Connecticut and Marquette are high Q1 wins which may offset it as long as we don’t blow this. When we weren’t getting blown out, the lone highlight was a win over an SEC team that is 8-14, 0-9,

In addition to getting to at least 12-8 in the BE, SHU is going to need at least another quality road win too, IMO.

 

[TomD82]

I don't think that is what the poster was complaining about. After beating Depaul our net jumped like 12 spots. USC beat OSU who is better than DePaul by 28 even if it was at home and net improved by 3. Nobody is worse than DePaul by any metric. Was the computer not still evaluating SHU strength of schedule?

As Sami posted RPI is all math. So taking the math into consideration for both DePaul and OSU and their poor winning percentages there would not have been much of a net delta for both SHU and USC after those 2 wins.

 

[LeftCoastPirates]

I don't think that is what the poster was complaining about. After beating Depaul our net jumped like 12 spots. USC beat OSU who is better than DePaul by 28 even if it was at home and net improved by 3. Nobody is worse than DePaul by any metric. Was the computer not still evaluating SHU strength of schedule?

As Sami posted RPI is all math. So taking the math into consideration for both DePaul and OSU and their poor winning percentages there would not have been much of a net delta for both SHU and USC after those 2 wins.

Part of the math is that the NET puts significantly more value in wins on the road. Then neutral court victories and lesser value at home when looking at that component.

Should it be as significant as it is? That’s debatable and I am not defending it one way or another.

But we know this is the case. So why are we so shocked when one result doesn’t move the needle as far as the other?

 

[walshtrips]

Why does it trouble you? Why not just explain it and leave the snark out?

Agree. I happened to be very good at math, computer analytics are only as good as the programming. GIGO.

 

[LeftCoastPirates]

Why does it trouble you? Why not just explain it and leave the snark out?

If we had one thread on this board called “NET COMPLAINTS” I would probably be ok with it.

But every time a result doesn’t move the needle in favor of Seton Hall, we get a new thread about the injustice of the NET like that poster is pulling back the curtain on the fraud that is a mathematical system.

it would be more fun to debate Seton Hall’s position on the bubble based on this metrics and other contributing factors. But the lack of comprehension of a metric like this normally takes those conversations off the rails. When simply getting educated on the facts would improve that.

The frustration is we have discussed the components on how it works several times. But posters continue to ignore and just start new threads complaining how they don’t get it. 🤷‍♂️. Hence snarky.

 

[TomD82]

Part of the math is that the NET puts significantly more value in wins on the road. Then neutral court victories and lesser value at home when looking at that component.

Should it be as significant as it is? That’s debatable and I am not defending it one way or another.

But we know this is the case. So why are we so shocked when one result doesn’t move the needle as far as the other?

Look I get it, but you really have not explained anything and admitted that nobody knows the formula. What the poster and others have complained about and I could be wrong is things like this and something like Rutgers beats Michigan on the road by 5 on Sat. and their net went down by 5 spots. Now the last place team Michigan in the #2 ranked net conference is a lot better than than last place Oregon State by any metric. So the computer taking into account SHU SOS for USC and RU walshtrips expected to see SHU net move more positive.

 

[lloyde dobler]

He has explained in prior posts what goes into the formula. The NCAA does list the ingredients, they just don’t reveal the recipe.

Another simple thing that goes into the day-to-day movement is what day of the week a game falls on. For instance, the SHU win at DePaul was on a Tuesday, when the schedule is relatively light. There’s less traffic, so to speak, to prevent a nice one day bump in comparison to the rest of the field.

If that win happened on Saturday, where almost everyone plays, it’s obviously not going to affect SHU’s individual ranking, but it would be harder to make a 10-spot jump when other teams are also getting a bump for wins on the same day.

 

[NittanyPirate]

If we had one thread on this board called “NET COMPLAINTS” I would probably be ok with it.

But every time a result doesn’t move the needle in favor of Seton Hall, we get a new thread about the injustice of the NET like that poster is pulling back the curtain on the fraud that is a mathematical system.

it would be more fun to debate Seton Hall’s position on the bubble based on this metrics and other contributing factors. But the lack of comprehension of a metric like this normally takes those conversations off the rails. When simply getting educated on the facts would improve that.

The frustration is we have discuss the components on how it works several times. But posters continue to ignore and just start new threads complaining how they don’t get it. 🤷‍♂️. Hence snarky.

You’re setting yourself up to get trolled. Maybe you already are. Every time I see a post about NET I know you will soon be around to write a thesis about how it works and complain about people that don’t understand it.

If it drives you as crazy as you claim, you could always just skip over the thread and allow us simpletons to whine about it.

 

[Seton75]

Because the computer doesn’t evaluate your Quad 1/2/3/4 records. Those buckets are fun categorization tools to carry a general conversation as to where the general quality of your wins have come from.

But make no mistake about it, at the end of the day when the committee of human individuals sits around a table to discuss your team sheet. A win vs the number 3 ranked NET team (UConn) when compared to a road win vs number 59 (Providence) are NOT equivalent to each other because they are classified as “Quad 1.”

The fact that I even have to explain that is troubling.

The computer is still evaluating USC towards your strength of schedule, as only 8 spots better than where it stood before, and is only a single piece of data in a total context that now spans 24 games worth of results are just as equally weighted.

I understand most people didn’t like math in school, still don’t like math today, and call people who do like math nerds. Well the nerds on this board have been trying to explain the math to the masses for quite sometime. Math doesn’t factor in opinions, bias, perceptions, or prejudice. It smashes all blue tinted glasses. Other posters have even shared the link to the NCAA’s publication of what goes into the math. The only thing the NCAA haven’t shared is the exact formula.

Seton Hall was a bad basketball team over the course of its first 9 games Seton Hall then became a very good team over the next 13. So when you fail the first part of the exam and they score well (not perfect) on the second part of the exam, you don’t average out to an A or B. You’re still somewhere in the middle.

And that’s still what our NET number reflects. Regardless of USC winning to get them to 9-13 over an Oregon State team that is tied with them for last place in the PAC 12 because it moved our loss to a Q2 game. 🤦‍♂️ We need to stop with this type of baseless complaining.

I love math. But you can use math as mental masturbation.

Which is what this is.

 

[sami]

It’s almost time to stop complaining about single game net moves and start talking about this year’s bubble being particularly weak (as opposed to every other year’s weak bubble, hence… bubble).

 

[TomD82]

He has explained in prior posts what goes into the formula. The NCAA does list the ingredients, they just don’t reveal the recipe.

Another simple thing that goes into the day-to-day movement is what day of the week a game falls on. For instance, the SHU win at DePaul was on a Tuesday, when the schedule is relatively light. There’s less traffic, so to speak, to prevent a nice one day bump in comparison to the rest of the field.

If that win happened on Saturday, where almost everyone plays, it’s obviously not going to affect SHU’s individual ranking, but it would be harder to make a 10-spot jump when other teams are also getting a bump for wins on the same day.

Maybe for women's net it will take into account the time of the month. The NCAA net rankings they released on their own site are games through Feb. 4 and that is what the complaints are about.

Jesus, Mary and Joseph!

 

[PirateFan94]

If we had one thread on this board called “NET COMPLAINTS” I would probably be ok with it.

But every time a result doesn’t move the needle in favor of Seton Hall, we get a new thread about the injustice of the NET like that poster is pulling back the curtain on the fraud that is a mathematical system.

it would be more fun to debate Seton Hall’s position on the bubble based on this metrics and other contributing factors. But the lack of comprehension of a metric like this normally takes those conversations off the rails. When simply getting educated on the facts would improve that.

The frustration is we have discuss the components on how it works several times. But posters continue to ignore and just start new threads complaining how they don’t get it. 🤷‍♂️. Hence snarky.

I think the poeple here complain becasue the NET is such a stupid metric. Bottom line is win or lose. Style points shouldn't matter.

 

[LeftCoastPirates]

You’re setting yourself up to get trolled. Maybe you already are. Every time I see a post about NET I know you will soon be around to write a thesis about how it works and complain about people that don’t understand it.

If it drives you as crazy as you claim, you could always just skip over the thread and allow us simpletons to whine about it.

Don’t you think there is a difference with complaining about it vs people factually misrepresenting what it is. That’s more of my issue.

How many times have people said margin of victory should not be a component of this?

Well it’s not. That it is a factually inaccurate statement that keeps getting brought up. Offensively and defensive efficiency are. So one would expect a larger margin of victory to improve those, but it’s not tied directly to the score. The subs could be in for both teams down the stretch and the score difference remains the same with everyone missing. Defensive metrics up and offensive metrics down.

I love math. But you can use math as mental masturbation.

Which is what this is.

truth 💯

 

[lloyde dobler]

Maybe for women's net it will take into account the time of the month. The NCAA net rankings they released on their own site are games through Feb. 4 and that is what the complaints are about.

Jesus, Mary and Joseph!

No it’s not. The complaints mentioned SHU’s movement the day after DePaul vs that of USC and other teams, including RU vs Michigan.

 

[LeftCoastPirates]

I know I am beating this thing to a dead horse at this point, but here goes anyway.

Love this article…really informative.

https://www.ncaa.com/news/basketbal...05/college-basketballs-net-rankings-explained

Here are some highlights if you choose not to read it.

“The remaining factors include the Team Value Index (TVI), which is a result-based feature that rewards teams for beating quality opponents, particularly away from home, as well as an adjusted net efficiency rating. The adjusted efficiency is a team’s net efficiency, adjusted for strength of opponent and location (home/away/neutral) across all games played. For example, a given efficiency value (net points per 100 possessions) against stronger opposition rates higher than the same efficiency against lesser opponents and having a certain efficiency on the road rates higher than the same efficiency at home.”

Translation. You get more credit for playing efficiently against a top tier team than by doing so against a lowly major opponent. It also states that it’s more important to do that on the road.
**Correct me if I am reading that incorrectly.

Also there IS a formula that is published for the calculating a team’s NET efficiency. There is no published formula for a team’s Team Value Index (TVI)

Hang onto your hat to follow this:
NET efficiency = your offensive efficiency - defensive efficiency.
Offense:
Total Possessions = FG attempts - offensive rebounds + turnovers + .475 x Free throw attempts.

Total Points / Total Possessions = Offensive Efficiency

Work pretty much the same for defense:

Defense:
Total Opponents Possessions = FG attempts - offensive rebounds + turnovers + .475 x Free throw attempts.

Total Opponents Points / Total Opponent’s Possessions = Defensive Efficiency

 

[Piratz]

I love math. But you can use math as mental masturbation.

Which is what this is.

Is that a Pitino quote?

 

[LeftCoastPirates]

So let’s take it one step further and evaluate the NET efficiency of Seton Hall’s performance for 3 specific games.

Game 1: Win vs Missouri (93-87)

Offense:
Total Possessions (68.925) = FG attempts (56) - offensive rebounds (12) + turnovers (14) + .475 x Free throw attempts (23).

Total Points (93) / Total Possessions (68.925) = Offensive Efficiency (1.35)

Defense:
Total Opponents Possessions (74.45) = FG attempts (62) - offensive rebounds (10) + turnovers (12) + .475 x Free throw attempts (22).

Total Opponents Points (87) / Total Opponent’s Possessions (74.45) = Defensive Efficiency (1.17)

NET efficiency = +0.18

Game 2: Win vs DePaul (72-39)

Offense:
Total Possessions (63.7) = FG attempts (59) - offensive rebounds (13) + turnovers (12) + .475 x Free throw attempts (12).

Total Points (72) / Total Possessions (63.7) = Offensive Efficiency (1.13)

Defense:
Total Opponents Possessions (70.45) = FG attempts (50) - offensive rebounds (5) + turnovers (15) + .475 x Free throw attempts (22).

Total Opponents Points (39) / Total Opponent’s Possessions (70.45) = Defensive Efficiency (.55)

NET efficiency = +0.58

Game 3: Loss vs Rutgers (70-63)

Offense:
Total Possessions (64.825) = FG attempts (61) - offensive rebounds (16) + turnovers (7) + .475 x Free throw attempts (27).

Total Points (63) / Total Possessions (64.825) = Offensive Efficiency (0.97)

Defense:
Total Opponents Possessions (64.6) = FG attempts (56) - offensive rebounds (9) + turnovers (16) + .475 x Free throw attempts (22).

Total Opponents Points (70) / Total Opponent’s Possessions (64.6) = Defensive Efficiency (1.08)

NET efficiency = -0.21

All games count equally for this specific measure. So you could technically create a spreadsheet and plug all these numbers in for every game for every team and then rank them by NET efficiency.

 

[HallX2]

Isn’t it also the case that as the body of games builds into the season anyone game played by a previous opponent will have less impact?

 

[LeftCoastPirates]

Isn’t it also the case that as the body of games builds into the season anyone game played by a previous opponent will have less impact?

One would absolutely have to assume that.

The Team Value Index doesn’t share a specific formula like the NET efficiency rating does.

But yes, if each result is given a specific score then that should get weighted into the other results and start to balance out.

Hence you see a team like Gonzaga specifically try to beef up on its OOC schedule with whom they perceive to be very high end opponents so the lower conference opponents don’t put a complete drag on their TVI.

You also saw the WCC give their top two teams byes directly into the semifinals of their conference tournament, to prevent them from having to play a 1-8 or 2-7 game vs an opponent who would only hurt their TVI.

But conversely you still want to root for that opponent to keep winning so the value of that game remains as high as possible. It’s the team’s final body of work that counts l, not where they were when you beat them.

So USC, Missouri, and Rutgers absolutely playing poor basketball (that’s putting it kindly) is not helping the Hall at all right now.

 

[walshtrips]

If we had one thread on this board called “NET COMPLAINTS” I would probably be ok with it.

This thread didn't start as a complaint, I just stated that the NET metrics don't make sense to me. The metrics use arbitrary markers to determine what is most important in the rankings.

We've been told that Quad 3 and Quad 4 losses take a toll. We've been told that road wins help a lot.

I noticed that SHU dropped from a 53 NET to 66 after a road loss to #7 Marquette and the USC loss changing from Quad 2 to Quad 3. In the few days that follow, we beat DePaul on the road and the USC loss pops back up to a Quad 2 loss. Additionally, RU beats Michigan on the road. Our net showed no movement in the days that followed the Marquette loss, despite several positive occurrences.

Would you agree that some coaches (Sha) will not run up a score unnecessarily while others will? Yet, margin of victory makes a big difference in the metrics. There's no way to measure coaching philosophy and that's just one major flaw in the metrics.

 

[LeftCoastPirates]

This thread didn't start as a complaint, I just stated that the NET metrics don't make sense to me. The metrics use arbitrary markers to determine what is most important in the rankings.

We've been told that Quad 3 and Quad 4 losses take a toll. We've been told that road wins help a lot.

I noticed that SHU dropped from a 53 NET to 66 after a road loss to #7 Marquette and the USC loss changing from Quad 2 to Quad 3. In the few days that follow, we beat DePaul on the road and the USC loss pops back up to a Quad 2 loss. Additionally, RU beats Michigan on the road. Our net showed no movement in the days that followed the Marquette loss, despite several positive occurrences.

Would you agree that some coaches (Sha) will not run up a score unnecessarily while others will? Yet, margin of victory makes a big difference in the metrics. There's no way to measure coaching philosophy and that's just one major flaw in the metrics.

Well I can’t help if you noticed wrong.

Went from 54 to 60 with the loss @ home to CU. Not specifically because of the loss but because this occurred on a Saturday where other teams also improved ahead of us.

Went from 60 to 62 because other teams moved up ahead of our Providence game.

Went from 62 to 68 after the home Providence loss.

Went from 68 to 74 after the Marquette road loss.

Then improved from 74 to 64 after the DePaul win.

You do realize there is a website that tracks all this specifically for each school.

https://bracketologists.com/team/seton-hall-pirates

I am curious, can you explain to me how you saw Seton Hall go from 53 to 66 after the Marquette loss. That is just factually wrong.

😂 you were hoping Rutgers moving up to 104 in the NET by beating now 111 Michigan was going to improve Seton Halls resume and move the needle for us 22 games into the season? Yeah not sure what to say to that.

 

[Seton75]

Now I get it. This NET thing is a complicated ruse to get shuisnjsteam to root for Rutgers.

 

[lloyde dobler]

Another thing…. Don’t lose any sleep over not getting Missouri to return that game next year

 

[JoePeppitone2012]

Think of it this way SHU went up 6 spots in the NET with a home win against I think #7 Marquette but beat down one of the worst D1 teams we've probably ever seen from the Big East with DePaul on the road and up 10 spots

I wish the Marquette win was worth more because that's a much more difficult win in reality versus math but that's the way NET works. And it's bothersome that this is factor in your postseason seeding

 

[lloyde dobler]

Think of it this way SHU went up 6 spots in the NET with a home win against I think #7 Marquette but beat down one of the worst D1 teams we've probably ever seen from the Big East with DePaul on the road and up 10 spots

I wish the Marquette win was worth more because that's a much more difficult win in reality versus math but that's the way NET works

Without checking, it could have been worth more. Got to remember that your team rating and your rank compared to be the rest of the field are two totally different things.

 

[JoePeppitone2012]

Without checking, it could have been worth more. Got to remember that your team rating and your rank compared to be the rest of the field are two totally different things.

Correct but at the time that's all it was just like with DePaul etc.

 

[BostonBlackie]

If we had one thread on this board called “NET COMPLAINTS” I would probably be ok with it.

But every time a result doesn’t move the needle in favor of Seton Hall, we get a new thread about the injustice of the NET like that poster is pulling back the curtain on the fraud that is a mathematical system.

it would be more fun to debate Seton Hall’s position on the bubble based on this metrics and other contributing factors. But the lack of comprehension of a metric like this normally takes those conversations off the rails. When simply getting educated on the facts would improve that.

The frustration is we have discuss the components on how it works several times. But posters continue to ignore and just start new threads complaining how they don’t get it. 🤷‍♂️. Hence snarky.

Left Coast - You're the Most Valuable Snark on here. It's very frustrating to read posts from people who never bothered to read previous posts on that topic. Well said.

 

[Piratz]

Think of it this way SHU went up 6 spots in the NET with a home win against I think #7 Marquette but beat down one of the worst D1 teams we've probably ever seen from the Big East with DePaul on the road and up 10 spots

I wish the Marquette win was worth more because that's a much more difficult win in reality versus math but that's the way NET works. And it's bothersome that this is factor in your postseason seeding

But it’s one criteria. And it does a good job. Yes it’s slow on teams that change midstream in one way or another, but by the end of it it’s a pretty fair read of the overall body of work. Then you have a Committee bring in all the trends and superlatives that the math misses.

There are 362 teams. You need to have something that tries to organize them. And yes the actual number of relevant programs for NCAA Tournament every year is significantly less, but still, it helps to have some sort of tool to arrange it. There’s so much going on with that many teams playing each other for 4 months spanning 32 different conferences before you have to arrange 68 of them into singular bracket.

 

[Section112]

I always thought the "USC effect" wascollege QBs who are overrated in the draft and never live up to their reputation in the NFL - lol. Except maybe Carson Palmer?

Let's see if Caleb Williams changes that perception?

 

[JoePeppitone2012]

I always thought the "USC effect" wascollege QBs who are overrated in the draft and never live up to their reputation in the NFL - lol. Except maybe Carson Palmer?

Let's see if Caleb Williams changes that perception?

I thought it was where USC QBs go to the Jets and ruin their careers

 

[Section112]

I thought it was where USC QBs go to the Jets and ruin their careers

Any QB can go to the Jets and ruin their careers. The Jets seem to be good at that - lol.

 

[radecicco]

Just win five more games and all this goes away.

 

[SOpirate]

Bottom line here is every USC and Rutgers win puts Seton Hall in a better position to make the dance.

 

[Seton75]

I always thought the "USC effect" wascollege QBs who are overrated in the draft and never live up to their reputation in the NFL - lol. Except maybe Carson Palmer?

Let's see if Caleb Williams changes that perception?

Come on. Gary Beban went to UCLA. LOL

 

[mbraue]

Seton Hall’s metrics and resume were very bad in the non-conference. This is going to plague us all year long and we better not pull in on the Bubble with that black mark. The wins over Connecticut and Marquette are high Q1 wins which may offset it as long as we don’t blow this. When we weren’t getting blown out, the lone highlight was a win over an SEC team that is 8-14, 0-9,

In addition to getting to at least 12-8 in the BE, SHU is going to need at least another quality road win too, IMO.

Butler win in Indy is now big too. They are a projected 9 seed now by Wachtel.

 

[Piratz]

Seton Hall is also 0-8 against Creighton, Xavier, and Villanova under Holloway, teams they play 4x down the stretch.

 

[TheHall87]

I don't think that is what the poster was complaining about. After beating Depaul our net jumped like 12 spots. USC beat OSU who is better than DePaul by 28 even if it was at home and net improved by 3. Nobody is worse than DePaul by any metric. Was the computer not still evaluating SHU strength of schedule?

As Sami posted RPI is all math. So taking the math into consideration for both DePaul and OSU and their poor winning percentages there would not have been much of a net delta for both SHU and USC after those 2 wins.

The key numbers behind why we jumped 10 or 12 spots against DePaul (and they dropped similarly) are 39 and 22.

We allowed 39 points, a season-low for both us defensively and DePaul offensively. DePaul shot 22% on FGs. I haven't checked but I'm going to go with that likewise being a season low for us on defense and DePaul on offense.

Those numbers are far enough off the norm, they are going to cause larger than typical shifts in the Team Value Index on both sides -- and thus significantly affect each team's ranking.

Using the USC-Oregon State game as the example, the numbers on either side aren't outliers the way DePaul's were.