MBB: Duke 86, Notre Dame 78 Postgame Thread

to put things in perspective: if one were to compare just the games in which both Kon and Isaiah have played (ie., removing the 3 games which Isaiah didn't play from Kon's ledger) just so that we're comparing apples to apples at least as far as the teams played (not necessarily the quality of the defense seen by each player in each game), then Kon is shooting 44.2% from 3 on the year. Isaiah is shooting 47.3% on the year in those same games. In the last 3 games Kon is shooting 9-22 from 3 (40.9%) whereas Isaiah is shooting 0-5 (0%).
 
DukieInBrasil said:
to put things in perspective: if one were to compare just the games in which both Kon and Isaiah have played (ie., removing the 3 games which Isaiah didn't play from Kon's ledger) just so that we're comparing apples to apples at least as far as the teams played (not necessarily the quality of the defense seen by each player in each game), then Kon is shooting 44.2% from 3 on the year. Isaiah is shooting 47.3% on the year in those same games. In the last 3 games Kon is shooting 9-22 from 3 (40.9%) whereas Isaiah is shooting 0-5 (0%).
Click to expand...
Uh oh.
 
DukieInBrasil said:
to put things in perspective: if one were to compare just the games in which both Kon and Isaiah have played (ie., removing the 3 games which Isaiah didn't play from Kon's ledger) just so that we're comparing apples to apples at least as far as the teams played (not necessarily the quality of the defense seen by each player in each game), then Kon is shooting 44.2% from 3 on the year. Isaiah is shooting 47.3% on the year in those same games. In the last 3 games Kon is shooting 9-22 from 3 (40.9%) whereas Isaiah is shooting 0-5 (0%).
Click to expand...
How did Isaiah do against highly ranked Auburn?
 
SkyBrickey said:
Devil's advocate. We played a mid-tier ACC team that was scorching hot from 3 and they still couldn't beat us. I would bet the house that no other ACC team shoots 58% from 3 against our defense.

I think we've only got 4 ACC games left that are losable. @Wake, Clemson, 2xUNC.

I believe the computers are still predicting we drop 1 and go 19-1 in conference.
Click to expand...
OK, but what does "losable" mean mathematically? In an athletic event, there is almost always a 1 in 20 or at least 1 in 100 chance that the underdog team pulls off the upset. For example, Massey's computer ratings that have Duke currently at number 6 and Miami currently at number 161, still show Duke with just a 90% chance of winning at home against Miami game.

And if you play 20 games each for which you have a 90% chance of winning (9 of 10), you have just 12.16% chance of winning all 20.

If you play 20 games each for which you have a 95% chance of winning (19 of 20), you have just 35.85% chance of winning all 20.

If you play 20 game each for which you have a 96.66% chance of winning (29 of 30), you finally have a 50.76% chance of winning all 20.

That's just how probability works.
 
Stray Gator said:
I've given up trying to comprehend why some fans feel compelled to find fault with the coaches' strategy and personnel decisions, even when it's working fine. At the game yesterday, the fellow sitting next to us -- who seemed to be a huge Duke fan attending his first game in person -- turned to his buddy with a giant smile when timeout was called with about 4.5 minutes left and Duke leading by 18: "That's it -- Scheyer's putting in the subs." I turned to him and said, as politely as I could, that Scheyer wouldn't likely take out the starters just yet. When he looked puzzled, I explained that I would expect the coaches to use this opportunity to get the starters some valuable experience managing a late-game situation, and also to try keeping Notre Dame's scoring in check for the purpose of maintaining Duke's favorable defensive metrics with an eye towards qualifying for a #1 seed. After ND proceeded to go on a 14-1 run to narrow the margin to 4 points, our neighbor looked over at me and nodded, saying: "Now I see why he left the starters in."

No coach is infallible, of course. But no coach is prescient, either. There were plenty of times when I wanted to wail loudl,y and rend my garments at some of K's decisions about personnel and game strategy. But over time, I came to realize that he was making far more informed choices; and more often than not, his Duke team ultimately would end up winning, even if it wasn't always free of nervous, frustrating moments for us fans.

It's perfectly fine for us to express opinions about which players should be getting more or less playing time -- that's what message boards like this are intended to facilitate. But I think we should never lose sight of two facts: First, that the coaches know more about the players' day-to-day potential to contribute to the success of the team than any of us; and second, that none of us is more motivated to see Duke succeed than the players and coaches themselves. So far, I think these coaches and players are doing a fabulous job; and while we engage in speculation and debate about the way it's all unfolding and how it might be done differently, I hope that everyone who cheers for Duke will enjoy and appreciate the special season we're fortunate to be witnessing.
Click to expand...
Yes. I was hoping the strategy thread would take all of these "Duke should do this" stuff off the post-game threads. It didn't. I get really annoyed reading comments about things from people who definitely don't know. Also there is an announcer thread for people to bash Jay Bilas and other announcers.
 
stickdog said:
OK, but what does "losable" mean mathematically? In an athletic event, there is almost always a 1 in 20 or at least 1 in 100 chance that the underdog team pulls off the upset. For example, Massey's computer ratings that have Duke currently at number 6 and Miami currently at number 161, still show Duke with just a 90% chance of winning at home against Miami game.

And if you play 20 games each for which you have a 90% chance of winning (9 of 10), you have just 12.16% chance of winning all 20.

If you play 20 games each for which you have a 95% chance of winning (19 of 20), you have just 35.85% chance of winning all 20.

If you play 20 game each for which you have a 96.66% chance of winning (29 of 30), you finally have a 50.76% chance of winning all 20.

That's just how probability works.
Click to expand...
Right. And Torvik is dynamically running this model with these probabilities and is projecting us at 19-1. I haven't looked at the underlying math, but I would bet the model says there's an 80-90% chance we take that loss against UNC, Clemson or Wake.

Miami is 4-12 and #138 in KP and coming into Cameron. They don't have a 1 of 10 chance of winning this game. I would love to get those odds on a betting site!
 
Last edited:
ClemmonsDevil said:
Yes. I was hoping the strategy thread would take all of these "Duke should do this" stuff off the post-game threads. It didn't. I get really annoyed reading comments about things from people who definitely don't know. Also there is an announcer thread for people to bash Jay Bilas and other announcers.
Click to expand...
Have you considered starting a thread titled "Your opinion is worthless unless you have experience coaching"?

I'm glad you weren't my hiring manager at McKinsey. I mean, what experience did I have with Insurance and Manufacturing?
 
DukieInBrasil said:
to put things in perspective: if one were to compare just the games in which both Kon and Isaiah have played (ie., removing the 3 games which Isaiah didn't play from Kon's ledger) just so that we're comparing apples to apples at least as far as the teams played (not necessarily the quality of the defense seen by each player in each game), then Kon is shooting 44.2% from 3 on the year. Isaiah is shooting 47.3% on the year in those same games. In the last 3 games Kon is shooting 9-22 from 3 (40.9%) whereas Isaiah is shooting 0-5 (0%).
Click to expand...
Where do you see 44 for Kon?I see 35
 
SkyBrickey said:
Right. And Torvik is dynamically running this model with these probabilities and is projecting us at 19-1. I haven't looked at the underlying math, but I would bet the model says there's an 80-90% chance we take that loss against UNC, Clemson or Wake.

Miami is 4-12 and #138 in KP and coming into Cameron. They don't have a 1 of 10 chance of winning this game. I would love to get those odds on a betting site!
Click to expand...

Well, Torvik's ratings have a bigger gap between Duke and the rest of the ACC. For example, Torvik's system gives Duke a 99% chance of beating Miami rather than the 90% that Massey's ratings suggest. And Torvik's system also takes in account the fact that Duke is already 6-0.

Still, Torvik's own numbers give Duke just a 40% chance to go 14-0 on Duke's last 14 ACC regular season games and just a 28% chance to win out through the rest of the regular season, despite his system's favoring Duke by at least 7.8 points in every remaining ACC game.

And personally, I think Torvik's system's +30.3 point victory prediction for Tuesday's Miami game is just a bit optimistic in terms of Duke's chances. I mean, Miami lost to Arkansas by 2, lost to Tennessee by 13, and lost at Virginia Tech by 1. Miami's recent 15 point loss to FSU was their worst loss of the season to date in terms of point differential. My guess is that Miami would win more than 1 out of 100 games against Duke, even if all of these games were played in Cameron. Even clearly inferior teams can pull off big upsets 1 out of 50 games. Torvik's system also favors Duke +17.1 at home against UNC. And as much as would like to crush UNC by even more than 17 points, I would stay away from that line.
 
Last edited:
stickdog said:
Well, Torvik's ratings have a bigger gap between Duke and the rest of the ACC. For example, Torvik's system gives Duke a 99% chance of beating Miami rather than the 90% that Massey's ratings suggest. And Torvik's system also takes in account the fact that Duke is already 6-0.

Still, Torvik's own numbers give Duke just a 40% chance to go 14-0 on Duke's last 14 ACC regular season games and just a 28% chance to win out through the rest of the regular season, despite his system's favoring Duke by at least 7.8 points in every remaining ACC game.

And personally, I think Torvik's system's +30.3 point victory prediction for Tuesday's Miami game is just a bit optimistic in terms of Duke's chances. I mean, Miami lot to Arkansas by 2, Tennessee by 13, and at Virginia Tech by 1. My guess is that Miami would win more than 1 out of 100 games against Duke, even if all of these games were played in Cameron. Even the clearly inferior teams can pull off big upsets 1 out of 50 games. Torvik's system also favors Duke +17.1 at home against UNC. And as much as would like to crush UNC by even more than 17 points, I would stay away from that line.
Click to expand...
stickdog said:
Well, Torvik's ratings have a bigger gap between Duke and the rest of the ACC. For example, Torvik's system gives Duke a 99% chance of beating Miami rather than the 90% that Massey's ratings suggest. And Torvik's system also takes in account the fact that Duke is already 6-0.

Still, Torvik's own numbers give Duke just a 40% chance to go 14-0 on Duke's last 14 ACC regular season games and just a 28% chance to win out through the rest of the regular season, despite his system's favoring Duke by at least 7.8 points in every remaining ACC game.

And personally, I think Torvik's system's +30.3 point victory prediction for Tuesday's Miami game is just a bit optimistic in terms of Duke's chances. I mean, Miami lot to Arkansas by 2, Tennessee by 13, and at Virginia Tech by 1. My guess is that Miami would win more than 1 out of 100 games against Duke, even if all of these games were played in Cameron. Even the clearly inferior teams can pull off big upsets 1 out of 50 games. Torvik's system also favors Duke +17.1 at home against UNC. And as much as would like to crush UNC by even more than 17 points, I would stay away from that line.
Click to expand...
Thanks for sharing all this. I'm not sure what Massey is using, but the Torvik model is pretty sophisticated and based on lots of seasons of historical data across all of CBB. His lines tend to track pretty closely to the betting lines - more validation. A 99% chance of winning rings true to me.
 
SkyBrickey said:
Thanks for sharing all this. I'm not sure what Massey is using, but the Torvik model is pretty sophisticated and based on lots of seasons of historical data across all of CBB. His lines tend to track pretty closely to the betting lines - more validation. A 99% chance of winning rings true to me.
Click to expand...
I would agree to disagree that Duke should and will be more than a 30 point favorite against Miami. While the line hasn't opened yet, that would knock me for a loop.
 
stickdog said:
I would agree to disagree that Duke should and will be more than a 30 point favorite against Miami. While the line hasn't opened yet, that would knock me for a loop.
Click to expand...
No, they tend to track very closely in the 0-15 pt range. You can see the daily Torvik predictions here and compare them to the lines yourself:

barttorvik.com

Scheduled MBB games for 20250112 and - Customizable College Basketball Tempo Free Stats - T-Rank

T-Rank College Basketball Ratings and Rankings. And fun.
barttorvik.com

Things get a little wonky at these higher numbers, because you are kind of betting how long Jon will play the starters. I would bet we open as a 24-26 pt favorite.
 
You must log in or register to reply here.
Back
Top