Page 1 of 2 12 LastLast
Results 1 to 20 of 37
  1. #1
    Join Date
    Feb 2008
    Location
    Raleigh, NC

    Duke and the "live by the 3" cliche/myth

    When Duke loses games this year, I often hear people say "well if Duke doesn't hit their 3s, you can beat them," and "Duke lives by the 3."

    I think that's a common misconception. I'm not real sure where it started either, because Duke doesn't really shoot a ton of threes.

    This year, Duke shot 1772 FG attempts so far. Of those, 586 were 3pt attempts. That's 33% of all attempts.

    Last year, Duke shot 1915 FG attempts. 739 were 3pt attempts. That's 38.6%.

    The 2010 championship team took 2206 shots. 35% of those were threes.

    So compared to the previous two years, Duke shoots much fewer threes.

    Compared to the top of the ACC, Duke's 3pt attempts are in line with the others. Below is a list of the % of shots taken from three...

    Miami 34.5%
    Maryland 30%
    UNC 30%
    NCSU 23.3%
    UVA 26.6%

    The nation's top teams also compare similarly.

    Indiana 32%
    Kansas 29.6%
    Gonzaga 31%
    Louisville 30.4%
    Georgetown 32.5%
    Syracuse 30.6%
    Michigan 34.1%
    Michigan St 27%
    Florida 40.5%

    When you look at number of shots, Duke compares to Indiana (33% on 1772 shots vs 32% on 1719 shots). Georgetown has taken about 260 fewer total FGA, so their % of 3PA actually is more impactful, as they play at a slower overall pace.

    However, while Duke takes 33% of their shots from 3pt range, they hit around 41.6% of those shots. IU hits around 41.8%. Georgetown hits around 38%. It might just be why these teams are pretty good. And if any of those teams are missing 3s, they're likely also missing other shots, which eventually leads to games they may lose. That's how it works.

    I know it's a real rough way to look at it, but I don't think it's accurate at all to say Duke is a 3 pt shooting team this year. If that's the case, then nearly every team in the country is a 3pt shooting team since many of them hover in the 30-35% of all attempt range. Florida, a top 15 team, took 40.5% from 3.

    But if you want a REAL glimpse at a "live by the 3" team, check out Iowa St.

    They have attempted 845 3pt shots out of 1936 total shots. That's 43.6% of their shots.

    However, Iowa St goes to the FT line about as often as Duke. So either Duke shoots a lot of jumpers or they just never get fouled.

  2. #2
    Interesting work.

    This thread is NOT a substitute postgame thread. If you want to discuss 3-point shooting in general, go for it.

  3. #3
    Join Date
    Feb 2008
    Location
    Raleigh, NC
    Quote Originally Posted by hurleyfor3 View Post
    Interesting work.

    This thread is NOT a substitute postgame thread. If you want to discuss 3-point shooting in general, go for it.
    Thanks... didn't want it to de-rail.

  4. #4
    Join Date
    Apr 2008
    Location
    Greenville, SC
    These numbers bear out what I've seen with my eyes this year. Given that we have Mason in the post, we're just as likely to give him an opportunity down low. I don't mind the perception though. Teams seem to close so hard on our 3pt shooters that they have ample opportunity to jab step, pump fake, or drive by their defender--which immediately puts the defense at a disadvantage. Of course if you can't buy a 3, it's tough to win a close game or narrow a big deficit. That's true regardless of how much you depend on the 3 in general.

    -c

  5. #5
    Before tonight's game, we ranked 170th in the country in terms of three-point attempts to field goal attempts. So, broadly speaking, I think you're right, this is not a particularly three-dependent team.

    I think it's an interesting question why the perception that Duke shoots a ton of threes every year continues to exist despite the evidence that it's not really true. I would argue that at least one reason is the incredibly low percentage of our opponents' field goal attempts that are threes--honestly, I think it's probably the most consistent aspect of our defense. Since 2003, here are the percentages of opponent FGAs that are threes, along with the national rank:

    03 - 25.0 (6)
    04 - 25.2 (4)
    05 - 20.8 (1)
    06 - 21.3 (1)
    07 - 24.3 (2)
    08 - 25.0 (2)
    09 - 25.9 (5)
    10 - 25.4 (11)
    11 - 24.5 (5)
    12 - 24.1 (3)
    13 - 26.4 (21)

    Keep in mind, there have been something like 340 teams in the country over this span, so to be so consistently very close to the top in this metric is really incredible. I think when you watch Duke games, it's easy to be fooled into thinking it's Duke shooting a lot of threes, when in fact it's far more the case that it's our opponents who aren't shooting threes.

  6. #6
    Join Date
    Feb 2008
    Location
    Raleigh, NC
    Quote Originally Posted by vick View Post
    Before tonight's game, we ranked 170th in the country in terms of three-point attempts to field goal attempts. So, broadly speaking, I think you're right, this is not a particularly three-dependent team.

    I think it's an interesting question why the perception that Duke shoots a ton of threes every year continues to exist despite the evidence that it's not really true. I would argue that at least one reason is the incredibly low percentage of our opponents' field goal attempts that are threes--honestly, I think it's probably the most consistent aspect of our defense. Since 2003, here are the percentages of opponent FGAs that are threes, along with the national rank:

    03 - 25.0 (6)
    04 - 25.2 (4)
    05 - 20.8 (1)
    06 - 21.3 (1)
    07 - 24.3 (2)
    08 - 25.0 (2)
    09 - 25.9 (5)
    10 - 25.4 (11)
    11 - 24.5 (5)
    12 - 24.1 (3)
    13 - 26.4 (21)

    Keep in mind, there have been something like 340 teams in the country over this span, so to be so consistently very close to the top in this metric is really incredible. I think when you watch Duke games, it's easy to be fooled into thinking it's Duke shooting a lot of threes, when in fact it's far more the case that it's our opponents who aren't shooting threes.
    It's the design of the defense... sometimes to my chagrin. I hate watching players drive at will when we close out on shooters so much.

  7. #7
    Quote Originally Posted by FerryFor50 View Post
    It's the design of the defense... sometimes to my chagrin. I hate watching players drive at will when we close out on shooters so much.
    Oh I agree--for better or for worse, it's clearly what K wants to do. My point was that one partial answer to the implied question of where the idea that Duke is always a heavily three-dependent team comes from is that we generally shoot far more than our opponents--but this is driven primarily from our defensive choices.

  8. #8

    The three

    What I have learned about stats is that they are good for discussions.
    However, being a rather simple minded individual, in the coliseum tonight, it felt like we died by the three and not-so-smart defense.
    Maryland was 8-20 from the three (40%) - they won. We were 4-25 (16%) - we lost. Tonight, almost 40% of our shots came from the three. We hit 40% from the three - we win.

  9. #9
    Quote Originally Posted by vick View Post
    Since 2003, here are the percentages of opponent FGAs that are threes, along with the national rank:

    03 - 25.0 (6)
    04 - 25.2 (4)
    05 - 20.8 (1)
    06 - 21.3 (1)
    07 - 24.3 (2)
    08 - 25.0 (2)
    09 - 25.9 (5)
    10 - 25.4 (11)
    11 - 24.5 (5)
    12 - 24.1 (3)
    13 - 26.4 (21)
    Thanks for these numbers. This is actually part of K's brilliance. 3PT shots are inherently less consistent than 2PT shots. So when you're a better team, you don't want your opponent shooting 3's. Otherwise they can get hot from long range and win in an upset.

    The opposite is true on offense. Ideally you would want all your shots to be short jump shots or layups. These are more consistent and it's harder to have a really bad shooting night if you solely shoot 2's. UNC has traditionally done this well, so when they have a great team they are less prone to upsets via bad shooting nights. (Though they're not as good at denying the 3).

    One interesting thing about Duke's defense that has been written many times is that it takes away the two efficient shots in basketball: the layup (via charges down low) and the 3 (via aggressively guarding and closing in on 3PT shots).

  10. #10
    Join Date
    Mar 2008
    Location
    New Orleans, Louisiana
    I didn't see tonight's game (2013 ACC Tournament loss to Maryland, in case you're reading this in this distant future*). I read the since-closed postgame thread and noticed a renewed despair that Duke lives and dies by the 3. Thanks to Ferryfor50's post -- which should become a DBR Sticky, get added to Throatybeard's Commandments, or both -- I wonder if the sentiment is simply being misstated. Maybe Duke doesn't live or die by the 3. Instead, maybe Duke does live or die by single chance possessions, a high percentage of which are missed 3-point shots with no offensive rebounds. (Do stats at least support that phrasing?)

    I'm no coach, but over the years I've trained my eyes to watch the other 4 players when a shooter fires from beyond the perimeter. Generally, their body language tells me that they either believe that the shot is going in, or have made peace with themselves that the possession had to end this way. What I don't see is a healthy distrust of the shooter. Maybe this is counterproductive to the idea of teamwork, and at that moment those 4 players are simply proud to let the shooter represent them. Maybe the forward movement required in offensive rebounding has to be sacrificed to retain valuable seconds of defensive positioning at the other end. I don't know.

    I've come to terms that Duke's all-or-nothing approach, or whatever you want to call it, isn't going away, and I consider it part and parcel of being a Duke fan. But I am curious to hear what others think about the complaint that never seems to go away.

    * To readers from the future: Yes, there was once a University of Maryland in the town of College Park. Up until 2014, they were members of the Atlantic Coast Conference. Then they moved to the Big Ten, and it all went to hell. Forced to compete with the research giants of the Midwest, and quickly running out of viable options, they placed all their eggs into one basket: building meth labs, sponsored by Under Armour. In 2018, the men's basketball team lost in conference to previously winless Eastern Washington, and a postgame riot created a giant fireball, destroying everything in sight. The Big Ten replaced Maryland by unanimous vote with Cuba's top university, the Universidad de La Habana, so that they could keep their eight-division format.

  11. #11
    Quote Originally Posted by brevity View Post
    Instead, maybe Duke does live or die by single chance possessions, a high percentage of which are missed 3-point shots with no offensive rebounds. (Do stats at least support that phrasing?)
    [/SIZE]
    Miles Plumlee and Zoubek were great on the offensive glass. The 2010 title team was one of the best o-rebounding teams in the nation. Other than that team, Duke has been okay to bad on the offensive glass over the last ten years.

    I imagine the Brand, Boozer, Battier years were better, but I don't have any numbers to prove it.

  12. #12
    Join Date
    Feb 2007
    Location
    Raleigh, NC
    We die by poor defense, not missed 3's. It's just that when the 3's don't go down we die miserably.

  13. #13
    Join Date
    Feb 2008
    Location
    Raleigh, NC
    Quote Originally Posted by billoz View Post
    What I have learned about stats is that they are good for discussions.
    However, being a rather simple minded individual, in the coliseum tonight, it felt like we died by the three and not-so-smart defense.
    Maryland was 8-20 from the three (40%) - they won. We were 4-25 (16%) - we lost. Tonight, almost 40% of our shots came from the three. We hit 40% from the three - we win.
    Tonight 38% of Duke's shots were from 3. 39% of Maryland's shots were from 3. And they led the entire game.

    Duke threw up about 5-7 forced 3pt shots in the final 2 min to try to come back, so they shot an even fewer % of 3s than the final would indicate (similarly to how a losing team has to foul and force a team to shoot FTs at the end of a game can skew the FTA numbers).

    If *any* team in the country knocks down 40% of their 3s, they likely win.

    In Duke's losses, this is their 3pt %:

    ACCT Maryland 16% (38% were 3pt attempts)
    UVA 32% (47% were 3pt attempts)
    Maryland 31.6% (30% were 3pt attempts)
    Miami 17% (31% were 3 pt attempts)
    NCSU 30% (29.8% were 3pt attempts)

    So in the losses, only 2 games Duke shot abysmally from 3. In only one game did they "live/die by the 3" (UVA at 47%).

    The real reason Duke loses games is FG% allowed vs FG % for and an opposing player (or several, like against Miami) goes off on us.

    In the losses:

    ACCT Duke 41.5% Maryland 51% Wells scores 30
    UVA 46% Duke 39.6% Harris scores 36
    Maryland 60% (!!!!) Duke 47.6% Len scores 19 and shuts down Plumlee
    Miami 56.9% Duke 29.7% (!!!!) Everyone on Miami not named Reggie Johnson destroys us
    NCSU 50.8% Duke 44.8% Leslie wakes up and scores 25

    There have been games that Duke has won where they shot poorly, but their opponent also shot poorly. It all comes down to defense, just like it has with every Duke team with championship aspirations...

  14. #14
    Join Date
    Feb 2008
    Location
    Raleigh, NC
    Quote Originally Posted by ns7 View Post
    Miles Plumlee and Zoubek were great on the offensive glass. The 2010 title team was one of the best o-rebounding teams in the nation. Other than that team, Duke has been okay to bad on the offensive glass over the last ten years.

    I imagine the Brand, Boozer, Battier years were better, but I don't have any numbers to prove it.
    This year's team would be even more dominant with a Zoubek... a guy who cleans up the glass, plays as big as he is and plays within himself and doesn't need shots to be effective.

  15. #15
    Join Date
    Nov 2007
    Location
    Delaware
    To some extent, I think that the reputation comes from selection bias and the chucking factor.

    As has already been pointed out, Duke shot no better than 32% in any of its losses. So while only 2 losses had terrible 3 point shooting, being sub par on 3's is common to the losses. The selection bias comes in because Duke has had bad 3 point shooting nights in wins as well. For example, against VCU, Duke only shot 26.6% from 3 and against Louisville, Duke shot only 25% from 3. People tend to overlook those kinds of stats because Duke won the games. That's where the selection bias comes in.

    The chucking factor happens because in losses, teams will take more 3's in an effort to come back. Tonight for example, 6 of Duke's last 8 shots were from 3. Duke made 2 of them. In fact, before the end of the game, Duke was shooting 3's at a rate right around their season average (19/57 for 33%). I didn't look at the play by play for every loss, but Duke's overall 3 point rate in losses is 34.7%. In the second half of losses, that number jumps way up to 43.7%. In the VCU and L'ville wins, Duke's 3 point rate was 33.9%, not far off from the losses. In the second half of those games, Duke only shot 31% of their shots from 3. They were in the lead an thus no need to chuck.

    In some ways, the 3 point myth becomes a self fulfilling prophesy. In wins, teams will be more selective from 3 and often shoot better (and when they don't it gets ignored because of the W). In losses, teams are forced to take tougher 3's to try and come back, which pumps up the 3 point rate, and holds down the 3 point percentage.

  16. #16
    Join Date
    Feb 2007
    Location
    Norfolk, VA
    Quote Originally Posted by FerryFor50 View Post
    When Duke loses games this year, I often hear people say "well if Duke doesn't hit their 3s, you can beat them," and "Duke lives by the 3."

    I think that's a common misconception. I'm not real sure where it started either, because Duke doesn't really shoot a ton of threes.
    My take on how to analyze the "Live by the 3, Die by the 3" label is not to look at how many 3 PT FGs we attempt as a percentage of total FGs attempted or to compare how many we take compared to how many other teams take, but to look at how many we make when we lose compared to how many we make when we win.

    I compare bad Duke (losses) to good Duke (wins):

    In five losses this season, the numbers are 28-112 (25%) on 3 PT FGs.

    1/12/13: [State 84 - Duke 76] 6-20 (30%) on 3 PT FGs
    1/23/13: [Miami 90 - Duke 63] 4-23 (17.4%) on 3 PT FGs
    2/16/13: [Maryland 83 - Duke 81] 6-19 (31.6%) on 3 PT FGs
    2/28/13: [Virginia 73 - Duke 68] 8-25 (32%) on 3 PT FGs
    3/15/13: [Maryland 83 - Duke 74] 4-25 (16%) on 3 PT FGs

    In our 25 wins (I am way too lazy to list the numbers for 25 games), the numbers are 203-461 (44%). So in our five losses, our 3 PT FG% is 19 percentage points lower than it is in our 25 wins. That's significant.

    Perhaps the numbers are skewed by jacking up multiple 3s late in the game during comeback attempts but I don't know that to be true. To prove a skew exists, the rate 3 PT FGs are attempted over the last five minutes or six minutes (arbitrary number) of the game would need to be compared to the rate over the first 35 or 34 minutes. I am not going to do that for two important reasons: 1) I am lazy; 2) Multiple posters with a strong math background and Duke statistics classes under their belt would pipe in an explain why my calculations are inaccurate/false due to principles I don't understand.

    For me, the "Live by the 3, Die by the 3" label isn't important because we live a lot more than we die. It is what it is.
    Bob Green

  17. #17
    Quote Originally Posted by Bob Green View Post
    ...but to look at how many we make when we lose compared to how many we make when we win.
    But isn't this sort of thing true for every team? At least with regard to shooting, not necessarily three-point shooting. Teams that shoot well tend to win and teams that shoot poorly tend to lose. If we took fewer threes, maybe the correlation would be for overall shooting percentages (or two-point shooting percentages), but so what? Then we'd "live and die by the two."


    (I'm not at all picking on you, Bob; I'm just saying the whole "live and die" meme is flawed.)

  18. #18
    Join Date
    Feb 2007
    Location
    Norfolk, VA
    Quote Originally Posted by Kedsy View Post
    But isn't this sort of thing true for every team? At least with regard to shooting, not necessarily three-point shooting. Teams that shoot well tend to win and teams that shoot poorly tend to lose.
    Absolutely.

    Quote Originally Posted by Kedsy View Post
    (I'm not at all picking on you, Bob; I'm just saying the whole "live and die" meme is flawed.)
    No problem. I'm not feeling picked on. I'm too depressed to feel picked on. However, as an East Coaster I can confirm the sun came up this morning.
    Bob Green

  19. #19
    Quote Originally Posted by Kedsy View Post
    But isn't this sort of thing true for every team?
    Maybe, but it still might be true there's a higher correlation between Duke losses/wins and 3-point shooting than for other teams. I honestly have no idea if that's the case, but it seems reasonable to me that some teams could be more sensitive to it than others (even if it affects everyone to some degree). It would be interesting to see what Bob's 19% differential of 3 point percentage between wins and losses is for other ranked teams, or other ACC teams.

  20. #20
    I for one have never thought Duke won because of taking more 3's. Duke has won because Duke's defense has always been about taking away two things, drives to the basket which result in lay ups and the three point shot. ADuke has been entirely willing to allow a team to attempt to beat it shooting from 12-19 feet. Duke and Coach Krzyzewski also realized that there is no point to shooting from 15 feet since the college 3 point shot is essentially not a long distance shot, but is in reality a mildly extended mid-range shot, and the reward vastly outpaces the difficulty. You get a full 50% more reward for hitting a shot that is only about 10% more difficult. Coach K realized that the shot was in essence too short and that the team who had a better time with it would always win.

    Every team in college basketball has lived and died on some level via the three point shot, whether it was hitting it, or taking it away. The shot was never long enough and wildly skewed the nature of basketball. It removed the importance of the big man and in essence made college basketball a jumpshooting contest and a mid-range one at best.

    20 feet is essentially a long range shot only for starters on the average high school basketball team. Or... 16 year old boys. Move the line back to 22' 6", where the shot becomes more of a shot for a grown man and you will see the risk/reward come back and the game will be back in balance. Think about it, the line was for a long time the same for both the women's and men's game. That is insane! And the difference now is almost negligible.


    I am not saying people don't enjoy watching the game played with the short three point line. I am saying it skews the game wildly. The shot was purposefully made too short to make the game seem more exciting to the casual observer and to enhance the ability of underdogs to pull an upset.
    Last edited by Spret42; 03-16-2013 at 11:52 AM.

Similar Threads

  1. Replies: 3
    Last Post: 01-21-2013, 08:36 PM
  2. Replies: 15
    Last Post: 12-05-2012, 03:54 PM
  3. Replies: 0
    Last Post: 11-10-2009, 05:00 PM
  4. "LIVE STATS" Question
    By Bluedawg in forum Elizabeth King Forum
    Replies: 4
    Last Post: 11-12-2007, 12:59 PM

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •