PDA

View Full Version : Bonussphere Bingo



MarkD83
02-07-2014, 06:29 AM
Once again I love the stats that Barry Jacobs shows. There is one additional calculation I do every game to let me know that Coach K's strategy is the right thing to do. Imagine a three point shooter takes a two instead of a three. Here are the shooting percentages that Duke would have to acheive to score as many pts from 2 as they score from 3. So you get approximately a 20% improvement in your scoring percentage by taking a 3. This is why I love it when Duke takes the 3s. (Sorry about the size of the table).

3870

roywhite
02-07-2014, 07:14 AM
The numbers are interesting, and Barry Jacobs makes a great point about Coach K seeming to know a secret about winning basketball games, but that path can be known by anyone who looks at the results. There has been some discussion in other threads about this, and I know Des Esseintes and others have made this same observation.

One qualification, and perhaps the quants can check into this -- seems to me that 2-point shot attempts draw more foul calls than 3-point shot attempts, and this does represent somewhat of a benefit to faithful users of the 2-point shot approach. Is that true; how much benefit does that provide? Do we have the resources to analyze that?

Bob Green
02-07-2014, 08:01 AM
I am a big fan of the 3 PT FG. Especially when the team works inside out with kick outs on offensive rebounds, to open shooters, being a prime example. The danger is in becoming one dimensional -- firing away from long distance early in the shot clock. As long as the team is creating open looks with ball movement (the extra pass), drive and kick out, plus the aforementioned offensive rebounds passed out to shooters, the 3 PT FG is both efficient and effective.

Mtn.Devil.91.92.01.10.15
02-07-2014, 08:07 AM
The numbers are interesting, and Barry Jacobs makes a great point about Coach K seeming to know a secret about winning basketball games, but that path can be known by anyone who looks at the results. There has been some discussion in other threads about this, and I know Des Esseintes and others have made this same observation.

One qualification, and perhaps the quants can check into this -- seems to me that 2-point shot attempts draw more foul calls than 3-point shot attempts, and this does represent somewhat of a benefit to faithful users of the 2-point shot approach. Is that true; how much benefit does that provide? Do we have the resources to analyze that?

Along these same lines, a three point shot provides a very different rebounding opportunity than an inside shot. A more distant shot frequently bounces much further off the rim/backboard affording opportunities for smaller players to follow the shot.

I have frequently noticed perimeter players collapsing to the free ball, and then kicking back out for another three pointer to take advantage of the extra space along the three point line and a quick shot.

Not sure how to quantify this, but definitely mitigates a deficiency under the basket.

Love how this team is really maximizing their strengths.

Go Duke

dyedwab
02-07-2014, 09:15 AM
...its should dispense with one of the most irritating tropes that comes about when debating our three point shooting.

....which is the very common idea that shooting threes is reliant on luck, or hot hands, or that if you shoot to many, you'll inevitably miss to many. What this article shows without any questions is that shooting in he high 30% from 3 is part of the DNA of the Duke offense. It's not random...it's designed.

And re, the points about fouling and rebounding (which I agree with, btw), there should be a fairly simple back of the napkin way to think about it. A three pointer is worth 1.5 what a two pointer is worth. Greater fouling on 2 should narrow that gap (as could differences in rebounding) but it I'm not sure that they account for a half a point difference per shot ( but I think they might, for some players - big men who shoot infrequently and get fouled a lot - see Amile Jefferson).

gus
02-07-2014, 09:31 AM
...its should dispense with one of the most irritating tropes that comes about when debating our three point shooting.

....which is the very common idea that shooting threes is reliant on luck, or hot hands, or that if you shoot to many, you'll inevitably miss to many. What this article shows without any questions is that shooting in he high 30% from 3 is part of the DNA of the Duke offense. It's not random...it's designed.

And re, the points about fouling and rebounding (which I agree with, btw), there should be a fairly simple back of the napkin way to think about it. A three pointer is worth 1.5 what a two pointer is worth. Greater fouling on 2 should narrow that gap (as could differences in rebounding) but it I'm not sure that they account for a half a point difference per shot ( but I think they might, for some players - big men who shoot infrequently and get fouled a lot - see Amile Jefferson).

One aspect that drawing fouls that is hard to quantify is foul trouble. Teams' play can be greatly affected when a key player get a 3rd foul early, or a fourth foul later in the game. (or obviously, if they foul out)

MChambers
02-07-2014, 09:38 AM
One aspect that drawing fouls that is hard to quantify is foul trouble. Teams' play can be greatly affected when a key player get a 3rd foul early, or a fourth foul later in the game. (or obviously, if they foul out)
It also helps if you can get in the bonus, and the double bonus, as soon as possible. Unless you are UNC and cannot make free throws.

Troublemaker
02-07-2014, 09:47 AM
This team, in particular, is the most comfortable I've ever felt about a Duke team shooting 3s. This team is just great at it. The '92 team might've had a higher shooting percentage but that mark was accomplished with a much lower volume of shooting, i.e. only shooting absolutely wide open 3s. Duke didn't really become a 3-heavy program until 1995. In any case, this 2014 Duke team is the best 3-pt shooting team Duke has ever had taking into account shooting percentage, volume, and consistency.

2014 probably has an historically great 3-pt attack among ALL Division-1 programs going back to when the 3-pt line was implemented. If this team continues shooting the way we have been going into April (!), I volunteer to perform a study to figure out just where this 3-pt attack ranks historically. The signature of this 3-pt attack has been its consistency. We've had only 1 truly poor shooting night: 29% against Arizona on low volume. We've also had two "meh" shooting nights of 32%, and then, every other game we've played, we've shot 34%+. That is pretty sick, and I suspect, historically consistent. The dorks definitely have tools to measure consistency, i.e. standard deviation. The study would have to consider 4 factors at least: consistency, shooting percentage, volume, and schedule adjustment. My hypothesis is this Duke team has a top-5 of all-time 3-pt attack if we maintain this rate of consistency, shooting percentage, and volume into April, against our quality of opponents.

dyedwab
02-07-2014, 09:49 AM
One aspect that drawing fouls that is hard to quantify is foul trouble. Teams' play can be greatly affected when a key player get a 3rd foul early, or a fourth foul later in the game. (or obviously, if they foul out)

Hard to quantify? Yes. Affects play greatly? Yes. I mean, we just saw it in the Syracuse game. But on basic math, shooting 50% from 2, which is a good shooting percentage is equivalent to shooting 33.3% from 3, or roughly six points less then Duke team's usually shoot from 3.

My point is that our strategy of shooting a lot of threes is not based on random, hot-handedness, but on a choice than can be based in statistical/mathematical data.

It also informs the idea that the very first thing our defense tries to do, even when we aren't a good defensive team (see this year and last year) is stop the 3.

Kedsy
02-07-2014, 11:36 AM
But on basic math, shooting 50% from 2, which is a good shooting percentage is equivalent to shooting 33.3% from 3, or roughly six points less then Duke team's usually shoot from 3.

Is 50% really a good shooting percentage? This season, shooting 50% on twos would rank a team approximately 130th among the 351 Division I teams. So, it's above average but I probably wouldn't characterize it as "good."

DevilYouthCoach
02-07-2014, 11:37 AM
Hard to quantify? Yes. Affects play greatly? Yes. I mean, we just saw it in the Syracuse game. But on basic math, shooting 50% from 2, which is a good shooting percentage is equivalent to shooting 33.3% from 3, or roughly six points less then Duke team's usually shoot from 3.

My point is that our strategy of shooting a lot of threes is not based on random, hot-handedness, but on a choice than can be based in statistical/mathematical data.

It also informs the idea that the very first thing our defense tries to do, even when we aren't a good defensive team (see this year and last year) is stop the 3.




In Youth Basketball a good 3-point shooter (and there are a lot of them out there) can completely take over a game. And they come in all sizes, shapes, and ages!

dyedwab
02-07-2014, 01:58 PM
Is 50% really a good shooting percentage? This season, shooting 50% on twos would rank a team approximately 130th among the 351 Division I teams. So, it's above average but I probably wouldn't characterize it as "good."

"Good" may be overstating. Overall point still holds. To use different numbers to better illustrate the point, a 3pt shooting percentage of 36% is the equivalent of a 2pt shooting percentage of 54%. A team shooting 54% from 2 would rank 25th among DI teams right now; Duke has shot below 36% from 3 exactly one since 1987 (2009, where we shot 34.9%)

Again, shooting threes is a strategic decision based on our talent and how we play, and there data to back up this strategy.

Kedsy
02-07-2014, 03:17 PM
"Good" may be overstating. Overall point still holds. To use different numbers to better illustrate the point, a 3pt shooting percentage of 36% is the equivalent of a 2pt shooting percentage of 54%. A team shooting 54% from 2 would rank 25th among DI teams right now; Duke has shot below 36% from 3 exactly one since 1987 (2009, where we shot 34.9%)

Again, shooting threes is a strategic decision based on our talent and how we play, and there data to back up this strategy.

I totally agree. In another thread I showed (or thought I showed, statistics not being my strongest suit) that as a strategy Duke shooting threes was more consistent than Syracuse shooting twos. With you, I was just nitpicking, and I apologize.

blUDAYvil
02-07-2014, 03:33 PM
2014 probably has an historically great 3-pt attack among ALL Division-1 programs going back to when the 3-pt line was implemented. If this team continues shooting the way we have been going into April (!), I volunteer to perform a study to figure out just where this 3-pt attack ranks historically. The signature of this 3-pt attack has been its consistency. We've had only 1 truly poor shooting night: 29% against Arizona on low volume. We've also had two "meh" shooting nights of 32%, and then, every other game we've played, we've shot 34%+. That is pretty sick, and I suspect, historically consistent. The dorks definitely have tools to measure consistency, i.e. standard deviation. The study would have to consider 4 factors at least: consistency, shooting percentage, volume, and schedule adjustment. My hypothesis is this Duke team has a top-5 of all-time 3-pt attack if we maintain this rate of consistency, shooting percentage, and volume into April, against our quality of opponents.

Not to play spoilsport, but Creighton's 3-pt production/efficiency has arguably been greater than Duke this year. They've averaged 11.0 made 3s a game vs. our 9.6. They're shooting at a 43.2% clip vs. our 41.6%. I haven't looked at standard deviations but their RPI is just a shade under ours (0.649 vs. 0.652).

UrinalCake
02-07-2014, 03:42 PM
According to the media,

Creighton shooting threes = smart play and a dangerous weapon
Duke shooting threes = risky, unsustainable, and dependent on luck

Kedsy
02-07-2014, 04:08 PM
According to the media,

Creighton shooting threes = smart play and a dangerous weapon
Duke shooting threes = risky, unsustainable, and dependent on luck

Is that from the media, or Duke's fan base?

azzefkram
02-07-2014, 04:14 PM
"Good" may be overstating. Overall point still holds. To use different numbers to better illustrate the point, a 3pt shooting percentage of 36% is the equivalent of a 2pt shooting percentage of 54%. A team shooting 54% from 2 would rank 25th among DI teams right now; Duke has shot below 36% from 3 exactly one since 1987 (2009, where we shot 34.9%)

Again, shooting threes is a strategic decision based on our talent and how we play, and there data to back up this strategy.

2P% are a bit misleading since they includes anything inside the arc from a dunk to a JH special. If it's a choice between a jumpshot from 8 to 20 ft away and a 3 ball, I definitely prefer the 3. If the choice is at the rim or a three ball, I'm leaning toward the dunk/lay-up. Dunks/Lay-ups have a pretty high success rate plus there is an increased likelihood of an O-rebound or foul.

Successful teams get a majority of their points from three places; at the rim, free throw and the 3 ball. If you are over-reliant on the 3, you have a tough time getting the other two which can limit your success. Duke is currently at about 40% in 3PA% (42nd). There are not a lot of quality and/or high major teams that shoot the 3 ball more frequently than us (Villanova and Creighton being the exceptions). I wouldn't mind if we shot a few less threes. Our FTR is middling and a few more dunks/lay-ups should help in that regard.

dyedwab
02-07-2014, 10:17 PM
2P% are a bit misleading since they includes anything inside the arc from a dunk to a JH special. If it's a choice between a jumpshot from 8 to 20 ft away and a 3 ball, I definitely prefer the 3. If the choice is at the rim or a three ball, I'm leaning toward the dunk/lay-up. Dunks/Lay-ups have a pretty high success rate plus there is an increased likelihood of an O-rebound or foul.

Successful teams get a majority of their points from three places; at the rim, free throw and the 3 ball. If you are over-reliant on the 3, you have a tough time getting the other two which can limit your success. Duke is currently at about 40% in 3PA% (42nd). There are not a lot of quality and/or high major teams that shoot the 3 ball more frequently than us (Villanova and Creighton being the exceptions). I wouldn't mind if we shot a few less threes. Our FTR is middling and a few more dunks/lay-ups should help in that regard.

We shoot a lot of 3s. We also have the 4th highest shooting percentage on 3s in the country (behind Creighton, Gonzaga, and Oregon State). We also are tied for 18th in the country in 3 attempted per game. So we shoot a lot of threes because we are good at shooting threes.

And right now, we have the most efficient offense in the country.

I'm repeating this, because, as Kedsy implied, a lot the the criticism of our three point shooting seems more about style than substance.

We shoot a lot more three than other teams, and make a lot more threes then other teams, because we have a team full of talented three point shooters. And that has led to the most efficient offense in the country. Sure, we can get better. But that's not about whether we take more twos or more threes. Thats about taking more good shots and fewer bad one. And for many of our player, 3s are very good shots.

Troublemaker
02-07-2014, 11:18 PM
Not to play spoilsport, but Creighton's 3-pt production/efficiency has arguably been greater than Duke this year. They've averaged 11.0 made 3s a game vs. our 9.6. They're shooting at a 43.2% clip vs. our 41.6%. I haven't looked at standard deviations but their RPI is just a shade under ours (0.649 vs. 0.652).

Yeah, I forgot to mention Creighton. They appear to also be an historically great 3-pt attack. Two in one year, though? Strange, but not unbelievable, imo.

Li_Duke
02-08-2014, 10:40 AM
Some have pointed out that 2-pointers have the advantage of netting more free throws and getting your opponent into foul trouble. I'll counter with if a team has an excellent reputation from 3, the opposing team can't clog up the middle, thus improving the chance of getting those high percentage 2-pointers (lay-ups and dunks) and forcing opponents to foul to avoid those lay-ups and dunks and thus getting opponents into foul trouble. :)

Conversely, if Duke has Shaq at his peak, his reputation as a great 2 point shooter and rebounder would force opponents to collapse into the paint and thus net more open 3s.

So you can't separate a team's 3 point success from their 2 point success from their free-throw success. They are all linked. You just lead with your best weapon (in our case, the three pointer) and go from there.