MBB Nerd Polls 2024-25

But we covered both....not sure why his algorithm see this win as sub par.
I cant say exactly but things like getting called for 21 fouls with BC only converting 54% of its FTs cause the score to be closer than our defense deserved. We also only caused 7 turnovers which I’m guessing is below average.
 
Torvik heavily discounts late scoring in runaways. That might have something to do with it.
I get that but Torvik views the BC game as the same as the ND game. Against ND we were at home and didn't come close to covering Torvik's spread.....against BC we were on the road and did cover the spread.....just doesn't add up especially since we did get a bump in KenPom but lost points in Torvik.
 
I get that but Torvik views the BC game as the same as the ND game. Against ND we were at home and didn't come close to covering Torvik's spread.....against BC we were on the road and did cover the spread.....just doesn't add up especially since we did get a bump in KenPom but lost points in Torvik.
I haven’t studied the intricacies of Torvik’s formulas but here are a couple of possibilities:
First - our rank isn’t just a function of our play, it is also a determined by how yhe systems rate the play of other teams vying for the top spot. Maybe Torvik’s formulas liked HOU and AUB’s performances more than Kenpom did.

Second - both systems have our D ranked 2nd but Torvik has our offense rated 8th while Kenpom has it at 4th. So the diff in overall ranking comes from how they calc offensive efficiency. As I noted earlier Torvik’s lower OE ranking for Duke is likely due to that system’s heavy downweighting of runaway scores heavily. Duke has had several such games recently.

Finally, before the game Torvik may have had BC’s defense rated worse than KP did - if so then Torvik’s downward adjustment of our raw off efficiency would have been greater than KP’s adjustment.

None of the above is disputing your main point that it was surprising to see our Torvik rank drop after we toyed with BC on their floor. Just trying to figure out mathematically why it happened.
 
KenPom's Player of the Year metric now has Cooper with a big lead over Broome. If his current ranking holds, he would have the second highest ever for a season, just barely behind Frank Kaminsky.
 
Torvik heavily discounts late scoring in runaways. That might have something to do with it.
I think you're close, but it's slightly different. Torvik looks at a team's average lead throughout the game, so it rewards a team that has a big first half. If Duke's halves last night had been reversed, Duke's score would have been better:

"For the past two seasons I have produced separate ratings, the "Implied T-Rank," using the GameScript stat. What I do is use the GameScript stat—which represents a team's average lead or deficit during a game—to infer a final score, and then use this derived final score instead of the actual final score to create the ratings."
 
I think you're close, but it's slightly different. Torvik looks at a team's average lead throughout the game, so it rewards a team that has a big first half. If Duke's halves last night had been reversed, Duke's score would have been better:

"For the past two seasons I have produced separate ratings, the "Implied T-Rank," using the GameScript stat. What I do is use the GameScript stat—which represents a team's average lead or deficit during a game—to infer a final score, and then use this derived final score instead of the actual final score to create the ratings."
This seems like the reason Duke dropped a bit in Torvik’s ratings. Duke trailed early and BC kept it close until sometime around the 12 minute mark in the 2nd half. As a result Torvik calculated Duke’s average lead as 5.7 points. In contrast, Torvik calculated Duke’s average lead vs ND as 11.1 points, even though the final score was closer.

I’m not sure about the merits of using average margin vs final margin (which is what KenPom uses). I suspect that it mostly averages out over the course of the season, but it can produce differences in the way Torvik and KenPom evaluate individual games.
 
This seems like the reason Duke dropped a bit in Torvik’s ratings. Duke trailed early and BC kept it close until sometime around the 12 minute mark in the 2nd half. As a result Torvik calculated Duke’s average lead as 5.7 points. In contrast, Torvik calculated Duke’s average lead vs ND as 11.1 points, even though the final score was closer.

I’m not sure about the merits of using average margin vs final margin (which is what KenPom uses). I suspect that it mostly averages out over the course of the season, but it can produce differences in the way Torvik and KenPom evaluate individual games.
Calculating "average lead" means that starting hot is more important than ending hot. Hmm.
 
I think you're close, but it's slightly different. Torvik looks at a team's average lead throughout the game, so it rewards a team that has a big first half. If Duke's halves last night had been reversed, Duke's score would have been better:

"For the past two seasons I have produced separate ratings, the "Implied T-Rank," using the GameScript stat. What I do is use the GameScript stat—which represents a team's average lead or deficit during a game—to infer a final score, and then use this derived final score instead of the actual final score to create the ratings."
Thanks for the details. Using average lead effectively discounts late pull-aways - it’s just a novel way to do it.
 
PS - Ok I read the methodology in the link MChambers provided and it turns out Torvik discounts late runaways in 2 different ways:
1. By using the average lead rather than actual score late spurts count less than early spurts.
2. Torvik goes beyond this though. He actually completely ignores all scoring after leads are big enough to be rated “safe” using Bill James’ famous formula!

Torvik’s rationale for these adjustments include:
1. A team that empties its bench during garbage time isn’t weaker than a team that keeps its starters out there. However just looking at the final score would make the former look weaker.
2. Backtesting showed that average lead was better than final lead at predicting future performance.
 
This seems like the reason Duke dropped a bit in Torvik’s ratings. Duke trailed early and BC kept it close until sometime around the 12 minute mark in the 2nd half. As a result Torvik calculated Duke’s average lead as 5.7 points. In contrast, Torvik calculated Duke’s average lead vs ND as 11.1 points, even though the final score was closer.

I’m not sure about the merits of using average margin vs final margin (which is what KenPom uses). I suspect that it mostly averages out over the course of the season, but it can produce differences in the way Torvik and KenPom evaluate individual games.

To be clear, neither Torvik nor Pomeroy explicitly care about final margin. They both care about adjusted efficiency margins. Also to be clear, Torvik doesn’t just use average margin. His metric just includes average margin additionally as a way to adjust for running up the score late or a team storming back once the game was already “over.” It is a “game control” measure as an auxiliary input rather than the primary measure.

As for why Torvik viewed the game as equal to the ND game, it is a combination of the “game control” measure (we had much better game control in the ND game, whereas we didn’t in the BC game) and the fact that BC is much worse than ND.

That said, I wouldn’t worry too much over it. A 93 game score is still very good (roughly about 15th nationally). To rate about 15th nationally in ehat clearly wasn’t our A game is fine.
 
To be clear, neither Torvik nor Pomeroy explicitly care about final margin. They both care about adjusted efficiency margins. Also to be clear, Torvik doesn’t just use average margin. His metric just includes average margin additionally as a way to adjust for running up the score late or a team storming back once the game was already “over.” It is a “game control” measure as an auxiliary input rather than the primary measure.

As for why Torvik viewed the game as equal to the ND game, it is a combination of the “game control” measure (we had much better game control in the ND game, whereas we didn’t in the BC game) and the fact that BC is much worse than ND.

That said, I wouldn’t worry too much over it. A 93 game score is still very good (roughly about 15th nationally). To rate about 15th nationally in ehat clearly wasn’t our A game is fine.
Thanks for this and for all the other related replies to my question.

Makes sense.......now about those Evan Miyakawa rankings:)
 
Back
Top