Statistical Breakdown: Best Three-Point Shooters in the NCAA

With an insane amount of help from Kevin Hogan, a visual statistician and sports data analyst, we've (again, mostly Kevin) put together some statistical views of the best three-point marksmen in the NCAA - 14 to be exact.  The analysis included four players from Mid-Major Madness' preview last August (Armand, Bader, Baron, and Richmond); while Karvel Anderson and Tyler Harvey are new additions that round out our mid-major pool.

Some cruxes to this analysis are that these players play different positions, play in different offensive/defensive systems, and come about three-pointers in different ways.  I wanted to be able to separate the catch-and-shoot players from the off-the-bounce players as best I could.  So, I used a blend of the eye test and assisted three-point percentage (mostly) to create several buckets.  I would have liked to have gotten shot chart data, but alas, I don't have Synergy access.

One footnote before I dive into things: the data is as of the last game on 3/12.

The Division I average for assisted three-point percentage is 84.9%.  Without any preexisting benchmarks for the buckets I described above, I pulled back all the percentages and grouped accordingly - for three-point 'types'.

Can Create Off-the-Bounce:

PG - Billy Baron, Canisius - 46.6%

SG - Rodney Hood, Duke - 70.8%

SG - C.J. Wilcox, Washington - 73.3%

SG - Nik Stauskas, Michigan - 76.1%

SG - Tyler Harvey, Eastern Washington - 76.1%

SG - Karvel Anderson, Robert Morris - 78.3%

SG - Akeem Richmond, East Carolina - 79.7%

Blended:

G - Sean Armand, Iona - 86.2%

SG - Trevor Cooney, Syracuse - 88.2%

F - Doug McDermott, Creighton - 90.0%

Mostly Catch-and-Shoot:

G - Brady Heslip, Baylor - 91.0%

G - Phil Forte, Oklahoma St. - 92.6%

SG - Travis Bader, Oakland - 93.9%

F - Ethan Wragge, Creighton - 96.9%

The table below takes a look at each player’s respective three-point percentage, and pace-adjusted triples per game.  Players are ranked going left-to-right, top-to-bottom based on their projected makes from beyond the arc per 40-minutes.  The shading in each player's cell reflects their three-point percentage.

The first thing that jumps out to me is that the catch-and-shoot players rank in the top-5.  Conversely, the players who can create off the bounce, but also are athletic enough and/or smart enough to do more than just shoot from beyond the arc make the least amount of triples per 40-minutes.  Their also the four players who will hear their names called during the NBA Draft this June (presuming the underclassmen declare) - potentially all in the lottery.

Richmond and Bader are the only players who take an average of 10+ triples per game; their percentages are above the Division I average of 34.3%, but their certainly volume guys.  Wragge, Heslip, and Forte benefit from playing with great players, and take at least 3/4 of their shots from beyond the arc.

Anderson is an interesting case because he connects from beyond the arc per 40-minutes nearly as frequently as Forte and Heslip.  However, he is the focal point of the Robert Morris offense and often has to create his shot.  Baron falls into this same category, except he's a point guard.  His quick trigger off the dribble is impressive, and as a player who often has the ball in his hands (28.7% usage), he is seldom assisted on his triples - the lowest player of the group.

It appears as though Cooney simply doesn't belong among this 'elite' group.  He certainly has some outstanding teammates (Fair, Grant, Ennis), but isn't converting as many open looks as one might suspect.  As the lone deep threat on the team, opposing defenses key on him around the perimeter.  Still, one might guess he would be at least a 40% three-baller.

A majority of these players come in well above the Division I average of 54.2% TS%.  It shouldn't surprise anyone given that we are talking about some of the best three-point shooters.  Basically, if a majority of your shot attempts are from beyond the arc, and you convert at the rate these guys do, you're TS% is going to be pretty good - like Mr. Wragge.

Of the 211 shot attempts he's taken this season, 204 have been triples.  Couple that with the fact that he shoots a high percentage, and only attempted 34 free throws, and it's easy to see why he's the best three-point shooter of the group (in terms of %), and also leads the TS% metric by a fair amount.

Conversely, Cooney doesn't show well again.  His low FG% (41%) along with the lowest three-point percentage of the group doesn't do him any favors in this comparison.

Wilcox, Bader, Richmond, and Harvey all attempted more than 400 field goals and over 100 free throws.  That's simply too many attempts when compared against players like Wragge, Forte, and Heslip who mainly take three's and don't attempt many free throws.

Stauskas, Anderson, and McDermott are super efficient despite taking 50% or less of their shots from beyond the arc; McDermott at just 33.8% is simply mind-boggling.  It's not hard to figure out why they still rank high in this analysis if you look at their FG/3FG/FT percentage splits: Stauskas - 49/46/81, Anderson - 51/45/84, and McDermott - 53/46/86.  That kind of efficiency is indicative of a high TS% even if they take a lot of attempts.

Likewise, Armand benefits from an efficient FG% of 49.5%, and plays in a three-point friendly offense for the Gaels - 44.4% of their attempts are three's.  They lead all of the NCAA in triples made (330), and second in three-pointers attempted (834).

Lastly, every player has been plotted on a graph with offensive rating on the x-axis and defensive rating on the y-axis.  It's ideal to be in the bottom right quartile of the graph - high offensive rating, low defensive rating.

One thing to keep in mind here: none of these guys are defensive stalwarts, so it shouldn't be a surprise to see some less than stellar ones - my goodness Travis Bader.  The players who fall below the average defensive rating all come from teams ranked right around 100 or less in defensive rating.  Obviously, scheme matters a lot here as well.

The sortable chart in the bottom left highlights some nifty advanced metrics that mix individual metrics with team ones.

The players, who are the primary scoring options on their teams, obviously have high usage percentages and individual points produced.  The catch-and-shoot players are towards the bottom in both categories.  Not surprisingly, the lone point guard, Baron, leads the group in individual points produced, and the do-everything McDermott is second.

Again, the primary scoring threats ranked higher in offensive win shares, while the catch-and-shoot players along with secondary options didn't rank as highly.

Overall, there is a lot that can be analyzed here, and I only scratched the surface.  While all these guys are impressive three-point marksmen, I'd rank them like this:

Best Off-the-Dribble Shooter: Karvel Anderson

Best Blended: Doug McDermott

Best Catch-and-Shoot: Ethan Wragge

McDermott is the best scorer in the group; I mean he does have 3,046 career points on 1,101 made baskets.  I think Forte perhaps has the deepest range, or he at least attempts the most shots from at least five feet from beyond the arc on above the break triples.  I think he and Heslip are perhaps the two players who are the most similar, and they both play in the Big 12.

Anderson has one of the best shots off the dribble, but Baron is close.  Karvel can stop on a dime off two or three hard dribbles before raising up, while Billy lulls defenders to sleep when it appears he is scanning the floor for teammates only to rise up and 'flick' a triple for an uncontested basket.

Hood, Stauskas, and Wilcox are all athletic enough to drive the lane and finish in transition.  They can all put the ball on the floor, have good size for their position, and clearly have to be respected from beyond the arc.  It's why these are the guys who are on mock drafts.

The fact that Creighton can run high screen-and-rolls with Wragge or McDermott seems unfair.  Their ability to flare on screens and get to the three-point line while in transition is extremely impressive.

Again, a big thank you to Kevin for all his work on this, and the constant refreshes of the data.  All the graphs and tables above are interactive, so have at it!