3 point shooting per Ken Pom

[gofurman]

since the line has been moved back a foot this year:

as a percent of shots taken this year

3-Point Tracker (D-I vs. D-I only)
2009 to date... 3PA/FGA: 33.1%
2008 season... 3PA/FGA: 34.4%

so teams are being more selective in shooting the three

3-Point Tracker (D-I vs. D-I only)
2009 to date... 3PM/3PA: 34.3%
2008 season... 3PM/3PA: 35.1%

** despite being more selective in when to shoot the 3, they are averaging one percent point less in hitting them.

Quite significant to Duke and the mid-majors who rely heavily on the three.

[CDu]

since the line has been moved back a foot this year:

as a percent of shots taken this year

3-Point Tracker (D-I vs. D-I only)
2009 to date... 3PA/FGA: 33.1%
2008 season... 3PA/FGA: 34.4%

so teams are being more selective in shooting the three

3-Point Tracker (D-I vs. D-I only)
2009 to date... 3PM/3PA: 34.3%
2008 season... 3PM/3PA: 35.1%

** despite being more selective in when to shoot the 3, they are averaging one percent point less in hitting them.

Quite significant to Duke and the mid-majors who rely heavily on the three.

Interestingly, here's how Duke has looked:

3pa/fga
2008: 39.2%
2009: 33.7%

3pm/3pa
2008: 37.7%
2009: 33.7%

So despite all the talk that we would not be affected by the new three point line, it appears that maybe we HAVE been affected. I wonder if part of the problem is that, instead of just taking the same long threes as last year (which would simply be closer to the line), we're now taking them from even further out (in other words, from the same distance from the three point line, which is now further).

For whatever reason, we're shooting 4 percent lower on threes than we have at any point in the past four years.

[gofurman]

Interestingly, here's how Duke has looked:

3pa/fga
2008: 39.2%
2009: 33.7%

3pm/3pa
2008: 37.7%
2009: 33.7%

So despite all the talk that we would not be affected by the new three point line, it appears that maybe we HAVE been affected. I wonder if part of the problem is that, instead of just taking the same long threes as last year (which would simply be closer to the line), we're now taking them from even further out (in other words, from the same distance from the three point line, which is now further).

For whatever reason, we're shooting 4 percent lower on threes than we have at any point in the past four years.

right - and I just wanted to show for all of NCAA D1 , not just Duke, so that it wouldn't matter per a single shooter (say, loss of Reddick). More statistically relevant perhas.

So it has hurt and we take 20 threes a game ( 587 3s/ 29 games ytd) and play an average of 35 games a year so that is right at 700 3s attempted.

lets say this year is a little low for duke at 33.7 and last year was a little high at 37.7...
700 3s' attempted in a year times a historical (guess here) 36.5 = 256 made threes or 768 points..
vs going forward 700 times (guess here) 34.5 = 726 points

*a loss of 42 points a year. That is worth a few wins for sure.

Now the part I am not taking into account is the fact that for Duke as a single team is shooting far less threes as a part of our arsenal this year as CDu pointed out.

I don't know the net of all this - just that I worry a little that the 3line move was not good for Duke. A little more statistical analysis including hte higher % of 2 pointers taken is needed. Maybe someone can figure that out - I ma too lazy...

[CDu]

right - and I just wanted to show for all of NCAA D1 , not just Duke, so that it wouldn't matter per a single shooter (say, loss of Reddick). More statistically relevant perhas.

So it has hurt and we take 20 threes a game ( 587 3s/ 29 games ytd) and play an average of 35 games a year so that is right at 700 3s attempted.

lets say this year is a little low for duke at 33.7 and last year was a little high at 37.7...
700 3s' attempted in a year times a historical (guess here) 36.5 = 256 made threes or 768 points..
vs going forward 700 times (guess here) 34.5 = 726 points

*a loss of 42 points a year. That is worth a few wins for sure.

Now the part I am not taking into account is the fact that for Duke as a single team is shooting far less threes as a part of our arsenal this year as CDu pointed out.

I don't know the net of all this - just that I worry a little that the 3line move was not good for Duke. A little more statistical analysis including hte higher % of 2 pointers taken is needed. Maybe someone can figure that out - I ma too lazy...

There are a lot of factors that play into whether or not the three point line change has hurt offenses. The reduction in threes attempted mean more two-point attempts and potentially more fouls drawn. But missed threes are arguably easier to rebound because you get more long rebounds. Conversely, missed threes by the opposition may result in more fast break attempts, which create easy points.

The easiest way to examine the overall effect would be to simply look at offensive efficiency adjusted per possession. That would be the best measure of the overall impact of the change. Obviously we can't control for all variables, but that'd be the closest we could get.

[bradjenk]

since the line has been moved back a foot this year:

as a percent of shots taken this year

3-Point Tracker (D-I vs. D-I only)
2009 to date... 3PA/FGA: 33.1%
2008 season... 3PA/FGA: 34.4%

so teams are being more selective in shooting the three

3-Point Tracker (D-I vs. D-I only)
2009 to date... 3PM/3PA: 34.3%
2008 season... 3PM/3PA: 35.1%

** despite being more selective in when to shoot the 3, they are averaging one percent point less in hitting them.

Quite significant to Duke and the mid-majors who rely heavily on the three.

I think you may be over judging the significance. If you break this down to a per game basis you will find that in an average NCAA game: Last year the 2 teams combined to make about 14 out of 41 3's (34.1%). This year the combined teams are making about 13 out of 39 (33.3%). In other words, in an average game there is 1 less made 3 and 2 less 3 pt attempts. I would not call that very significant at all. It would be the equivalent of a team averaging .5 less turnovers per game or .5 more rebounds. It's a little different but not really by much.

[mehmattski]

since the line has been moved back a foot this year:

as a percent of shots taken this year

3-Point Tracker (D-I vs. D-I only)
2009 to date... 3PA/FGA: 33.1%
2008 season... 3PA/FGA: 34.4%

so teams are being more selective in shooting the three

3-Point Tracker (D-I vs. D-I only)
2009 to date... 3PM/3PA: 34.3%
2008 season... 3PM/3PA: 35.1%

** despite being more selective in when to shoot the 3, they are averaging one percent point less in hitting them.

Quite significant to Duke and the mid-majors who rely heavily on the three.

This is probably the scientist in me talking, but throwing around words like "significant" when talking about statistics made me perk my ears. Is it really significant? Consider the NCAA 3pt percentage over the past 10 years:

2009: 34.3%
2008: 35.1%
2007: 35.0%
2006: 35.0%
2005: 34.7%
2004: 34.6%
2003: 34.8%
2002: 34.6%
2001: 34.4%
2000: 34.4%
1999: 34.2%

Source

Certainly the difference between this season and last is the greatest difference between any two seasons. But do we really think that since 2001 the shooters have gotten "significantly" better? If we have 100 random seasons, will at least 95 of them be between 35.8% and 34.2%? Because if not, then you really can't say that there is a "significant" difference.

I also am not sure about the premise that in a single game it's going to matter much. One percentage point means that one out of a hundred treys is not going to fall. Will that make the difference for some three-heavy mid-major team in the tournament?

[CDu]

I think you may be over judging the significance. If you break this down to a per game basis you will find that in an average NCAA game: Last year the 2 teams combined to make about 14 out of 41 3's (34.1%). This year the combined teams are making about 13 out of 39 (33.3%). In other words, in an average game there is 1 less made 3 and 2 less 3 pt attempts. I would not call that very significant at all. It would be the equivalent of a team averaging .5 less turnovers per game or .5 more rebounds. It's a little different but not really by much.

And that's ignoring the offsetting increase in two pointers (or free throws) made. I'm guessing that, on a national aggregate, offensive efficiency may have gone down a tad, but not a lot. Some teams have taken bigger hits than others, obviously.

[bradjenk]

And that's ignoring the offsetting increase in two pointers (or free throws) made. I'm guessing that, on a national aggregate, offensive efficiency may have gone down a tad, but not a lot. Some teams have taken bigger hits than others, obviously.

It's also almost impossible to really tell what effect the 1 foot line change has on any given team. Unlike the NBA the rosters turn over so much that the offensive philosophy has to vary more. If you asked the Duke coaches I bet they will tell you the line change has not affected their shot selection as much as the lineup has. How else would you explain all the teams that are actually attempting more 3's this year? But I don't think any team that had good shooters at last year's line suddenly became reluctant shooters at this year's.

[CDu]

It's also almost impossible to really tell what effect the 1 foot line change has on any given team. Unlike the NBA the rosters turn over so much that the offensive philosophy has to vary more. If you asked the Duke coaches I bet they will tell you the line change has not affected their shot selection as much as the lineup has. How else would you explain all the teams that are actually attempting more 3's this year? But I don't think any team that had good shooters at last year's line suddenly became reluctant shooters at this year's.

You are correct that it would be REALLY hard (if not impossible) to tell exactly how much the three point line has made the difference, given different lineups, different opponents, different opponents' lineups, etc. With Duke, it's hard to say if it's the line or the players. Smith and Scheyer are down from their numbers last year, Henderson and Singler are up, and Paulus is way down (but that could be the injury and reduced role). We also lost Nelson and King, which hurt as well considering both had high percentages.

[bradjenk]

You are correct that it would be REALLY hard (if not impossible) to tell exactly how much the three point line has made the difference, given different lineups, different opponents, different opponents' lineups, etc. With Duke, it's hard to say if it's the line or the players. Smith and Scheyer are down from their numbers last year, Henderson and Singler are up, and Paulus is way down (but that could be the injury and reduced role). We also lost Nelson and King, which hurt as well considering both had high percentages.

You know I hadn't thought about Taylor King in a long time. But he certainly had an affect on the 3pt attempt ratio for sure.

One last thing on the significance issue, as it relates to the tradeoff from 3pt attempts. If the average team is attempting just 1 less 3 per game, what exactly can they impact with that ONE extra possession? How may more 2's can they attempt or make? How many more trips to the foul line can they get? How many more turnovers or held balls can they have? The answer to all of these still must add up to ... ONE. Therefore, on average there has been hardly any impact on the game at all. Of course individual teams may be impacted but I still think personel is the significant factor and the line change not much at all.

[gofurman]

It's also almost impossible to really tell what effect the 1 foot line change has on any given team. Unlike the NBA the rosters turn over so much that the offensive philosophy has to vary more. If you asked the Duke coaches I bet they will tell you the line change has not affected their shot selection as much as the lineup has. How else would you explain all the teams that are actually attempting more 3's this year? But I don't think any team that had good shooters at last year's line suddenly became reluctant shooters at this year's.

I agree - for one team it is more about who is taking the threes than the distance. Get Steph Curry on your team and number taken and percent made will go up. That is why I posted that the percent for all D1 teams is down one percentage point. I didn't want to focus on Duke or any other one team.

*However, Mattski gave the following for the 300+ D1 teams:

2009: 34.3%
2008: 35.1%
2007: 35.0%
2006: 35.0%
2005: 34.7%
2004: 34.6%
2003: 34.8%
2002: 34.6%
2001: 34.4%
2000: 34.4%
1999: 34.2%

Good data, perhaps I am mistaken. Also, I agree I am oversimplifying as for the increased number of two's taken / fouls incurred.