A data-backed ranking of Division I men's basketball shooters from the 2014-15 through 2024-25 seasons, using Sports-Reference counting stats and ESPN shot locations where they exist.
Chris Collinsworth · · 10 min read
Loading post…
College basketball spent this decade stretching the floor and then arguing about whether anyone could actually shoot it. The argument is usually a name, a clip, and a single-season 3-point percentage. This is an attempt to do it the other way: take every Division I season from 2014-15 through 2024-25, keep the players who actually took a career's worth of threes and went to the line, and rank the ones who were accurate in more than one way.
199 players cleared 250 made threes and 200 free-throw attempts in that window. The 25 below are the ones who stacked volume-weighted 3-point percentage, free-throw percentage, and true shooting. Fletcher Magee is first. That will not shock anyone who watched Wofford in 2019. What might: Antoine Davis, who made 588 window threes, lands 82nd, and Buddy Hield — 240 makes in 2014-15 and 2015-16 — misses the cut by ten.1
The numbers are not a draft. Counting stats come from College Basketball Reference player pages when the slug resolved,2 and from ESPN's unofficial college statistics API otherwise.3 Shot maps come from ESPN play-by-play coordinates, and only for games that actually have them.4 The same window on the glass is in the 25 best college rebounders of 2015-2025.
The 25, ranked
Qualified Division I shooters, 2014-15 through 2024-25
1. Fletcher Magee, Wofford. 509 threes on 43.5 percent, 90.8 percent from the line, 64.3 percent true shooting in 134 games.5 Four seasons, all in the window: 91-of-190 as a freshman (47.9%), then 112, 148, and 158 makes. The composite likes him because nothing is a hole. He is first on the default weights, first if you ignore the 3P% shrink, first if you overweight free throws, second if you overweight threes or true shooting. That is not a ranking artifact.
2. Anthony Mathis, New Mexico / Oregon. 291 threes, 43.6 percent, 87.2 percent FT, 65.3 percent TS — the best true-shooting mark on the list.6 He is a reminder that the floor is 250 makes, not 400. Two quiet early seasons in Albuquerque, then 98 and 106 threes, then a graduate year at Oregon at 45.4 percent. He is first if you let TS% run the table.
3. Schadrac Casimir, Iona / Florida Gulf Coast. 279 threes at 44.5 percent, the highest raw 3P% in the 25.7 A 2014-15 Iona year (94-of-222) starts the window; 2018-19 at FGCU (74-of-155, 47.7%) ends it. He is first on a 3-point-heavy composite and third on the default. The shrink keeps a couple of tiny-sample Iona slivers from inflating him further.
4. Sam Hauser, Marquette. 309 threes, 43.9 percent, 88.0 percent FT, 63.8 percent TS.8 The 2017-18 season (95-of-195, 48.7%) is the spike; the other Marquette years still sit in the low-to-mid 40s. He does not need a single-year narrative. The ranking is the career in the window.
5. Sam Merrill, Utah State. 319 threes, 42.0 percent, 89.1 percent FT, 63.0 percent TS.9 Fifth on the default list, eighth if you throw 70 percent of the weight on 3P%. The free-throw mark is the hold.
Just outside that group: Markus Howard (434 threes, 42.7%, 88.2% FT) at sixth,10 and Terrell Gomez at seventh on ESPN-only totals — Sports-Reference never resolved a player page for him, so those 344 makes / 43.1% / 89.0% FT are ESPN season sums, cited as such.11 The rest of the 25, each tied to a player page: Jordan Dartis,12 Andrew Rowsey,13 Cassius Winston,14 Mike Daum,15 Cam Spencer,16 Ryan Kemrite,17 Ernie Duncan,18 Devin Cannady,19 Nate Navigato,20 Francis Alonso,21 Corey Kispert,22 Braden Norris,23 Hunter Cattoor,24 James McGee,25 LJ Cryer,26 Jordan Bohannon,27 MaCio Teague,28 and Tyler Hall.29
Eligibility is Division I men's basketball, regular season, seasons ending 2015 through 2025 (2014-15 through 2024-25). Earlier or later college seasons are ignored.
Two floors, chosen so a specialist who never got fouled and a low-volume marksman who never took enough threes both stay off the list:
at least 250 made threes in the window
at least 200 free-throw attempts in the window
Among players who clear both, three rates:
3P%, shrunk toward the qualified-group mean with a 120-attempt prior (empirical Bayes — a 48% season on 90 attempts is not treated like a 48% season on 400).
FT%, raw. The 200-FTA floor is the volume check.
TS%, which folds in twos and free throws so a player who only lives behind the arc does not automatically beat one who also finishes and draws fouls.
Each rate becomes a z-score inside the 199-player group. The default composite is 0.45 × z(shrunk 3P%) + 0.25 × z(FT%) + 0.30 × z(TS%). Ties break on made threes, then name.
TS% already contains 3P% and FT%. That is deliberate — it rewards complete shooters — and it is why the sensitivity table exists.
Discovery did not use Sports-Reference's Stathead player-season finder: that path returned a Cloudflare 403 from this environment.30 All-time career leader URLs 404'd after a /men/ redirect. Season leaderboards on Sports-Reference only publish a top ten, which would miss a four-year 70-make shooter who still clears 250. So the candidate pool is ESPN's unofficial by-athlete season totals for every Division I player, sorted by made threes, paginated until a page's lowest 3PM dropped under 40 (or eight pages, whichever first). Anyone with 200 ESPN-window makes was then matched to a Sports-Reference player page (/cbb/players/{slug}-n.html). When the page resolved, Sports-Reference counting stats and TS% win. When it did not, ESPN season totals are aggregated and TS%/eFG% are computed from FGM, FGA, 3PM, FTM, FTA, and points.
312 of 352 candidates matched a Sports-Reference page. The 40 misses are mostly suffix names (Jr., III) and a few mid-majors whose slugs we did not guess. Their ESPN totals stay in the pool; only Gomez of the 25 is in that group.
The bar is raw 3P% on window totals, not the shrunk rate used in the composite. The scatter is the same 25: attempts on the x-axis, accuracy on the y-axis. The first five are highlighted. Magee sits high and right — 1,169 attempts at 43.5%. Casimir is the high-efficiency, slightly lower-volume corner. Jordan Bohannon and Tyler Hall are the volume dots that do not quite climb the accuracy axis far enough to crack the top ten.
Free-throw percentage is the cheapest available proxy for "does this jumper stay on line when nobody is closing out." Andrew Rowsey (91.4%) is the far-top of that chart and jumps to second when free throws get 60% of the composite.13 Ryan Kemrite is 13th by default and 73rd if you overweight FT% — 76.7 percent on enough attempts to count, not enough to hang with this group.17
Each line is that player's 3P% in every window season with at least 20 attempts. Magee's freshman 47.9% is the early peak; he never falls out of the 41s. Mathis's first two New Mexico years are small-sample and omitted by the 20-attempt rule, which is why the shrink exists in the ranking.
1 is first. Default is 45/25/30 on shrunk 3P%, FT%, TS%.
1
Fletcher Magee
2
1
2
1
1
2
Anthony Mathis
4
5
1
2
2
3
Schadrac Casimir
1
4
4
4
3
4
Sam Hauser
3
3
3
3
4
5
Sam Merrill
8
6
5
5
5
6
Markus Howard
6
8
9
6
6
7
Terrell Gomez
5
7
15
9
8
8
Jordan Dartis
7
11
7
8
7
9
Andrew Rowsey
15
2
10
7
9
10
Cassius Winston
9
19
17
13
10
11
Mike Daum
17
17
6
10
11
12
Cam Spencer
11
10
14
12
12
13
Ryan Kemrite
10
73
8
16
13
14
Ernie Duncan
13
30
13
14
14
15
Devin Cannady
24
9
12
11
15
16
Nate Navigato
14
43
20
18
16
17
Francis Alonso
23
14
19
15
18
18
Corey Kispert
25
36
11
17
17
19
Braden Norris
16
74
16
24
19
20
Hunter Cattoor
18
54
21
21
20
21
James McGee
12
96
25
34
21
22
LJ Cryer
20
22
48
27
22
23
Jordan Bohannon
33
16
34
20
24
24
MaCio Teague
26
24
39
26
23
25
Tyler Hall
35
25
28
23
27
Magee does not move. Mathis and Casimir trade the 3P-heavy / TS-heavy crowns. Cassius Winston (10th default, 19th FT-heavy, 17th TS-heavy) is the "good enough at everything, not first at one thing" prototype.14 Mike Daum climbs to 6th when TS% is 60% of the weight — the South Dakota State big who actually shot it.15 James McGee is 21st default and 96th FT-heavy; do not cite him as a free-throw story.25
Antoine Davis is the stress test for the floors. 588 window threes is first among everyone we measured. 37.5 percent from three and 55.5 percent true shooting is why he is 82nd, not 8th. The list is not "most threes." It is "best shooter," and those are different jobs.31
ESPN play-by-play objects include shootingPlay and coordinate: {x, y} on a 50-foot-wide half court, basket at (25, 2), y = 0 at the baseline. Earlier testing found those coordinates starting around 2013-14 and missing for 2004 / 2008 / 2012. This article only requests them for a handful of the top 25, and only for each player's highest-3PA season in the window — not a career dump.
Coverage on the six maps we pulled: Hauser 191 shots in 16 of 34 games, Howard 311 in 16 of 34, Magee 96 in 6 of 35 (Wofford, 2018-19), Casimir 70 in 7 of 35, Merrill 83 in 5 of 35, Mathis 79 in 8 of 32 after dropping ESPN sentinel coordinates. That is the coverage, not a cleaned sample. Makes are filled dots; misses are the red X. The zones view is paint (inside 8 feet), seven mid-range wedges, and seven three-point wedges. Filter All / 2PT / 3PT; the bars are FG%, 2P%, 3P% on the same shots. The painted arc is the NCAA men's line in use since 2019-20 — 22 ft 1.75 in from the basket, hashes 3 ft in from each sideline — not the NBA 23.75. Every map below is a 2014-15 or 2018-19 season, so those shots were taken on the older 20 ft 9 in line; the overlay is today's college court for orientation. 2 vs 3 on the dots still comes from ESPN's play value, not from which side of the painted line a coordinate sits on. Smaller-conference games are unverified: if ESPN did not store a coordinate, the shot is not on the map.
Schadrac Casimir · 2014–15 (7 of 35 games with coordinates)
FG%
38.6%
2PT%
31.0%
3PT%
43.9%
70 field-goal attempts from ESPN play-by-play. Coverage is whatever ESPN logged — often thinner outside major-conference games.
Anthony Mathis · 2018–19 (8 of 32 games with coordinates)
FG%
41.8%
2PT%
38.9%
3PT%
42.6%
79 field-goal attempts from ESPN play-by-play. Coverage is whatever ESPN logged — often thinner outside major-conference games.
Fletcher Magee · 2018–19 (6 of 35 games with coordinates)
FG%
31.3%
2PT%
50.0%
3PT%
25.0%
96 field-goal attempts from ESPN play-by-play. Coverage is whatever ESPN logged — often thinner outside major-conference games.
Sam Hauser · 2018–19 (16 of 34 games with coordinates)
FG%
46.1%
2PT%
48.8%
3PT%
44.0%
191 field-goal attempts from ESPN play-by-play. Coverage is whatever ESPN logged — often thinner outside major-conference games.
Markus Howard · 2018–19 (16 of 34 games with coordinates)
FG%
43.1%
2PT%
42.0%
3PT%
44.2%
311 field-goal attempts from ESPN play-by-play. Coverage is whatever ESPN logged — often thinner outside major-conference games.
Sam Merrill · 2018–19 (5 of 35 games with coordinates)
FG%
41.0%
2PT%
44.0%
3PT%
36.4%
83 field-goal attempts from ESPN play-by-play. Coverage is whatever ESPN logged — often thinner outside major-conference games.
Window, not career. Hield's 2013-14 season is out; he finishes on 240 window makes and does not qualify. A 2025-26 season is also out.
Regular season. ESPN seasontype=2 and Sports-Reference regular-season player tables. The NCAA tournament can move a reputation more than these totals.
Transfers count. Every D-I season in the window is summed. A mid-major year equals a Power conference year.
TS% double-counts. Sensitivity rankings are the check. If someone is 4th on the default list and 18th when 3P% is 70% of the composite, say so when you cite them.
ESPN is unofficial. Responses are cached and never fetched from the live site. Counts can disagree with Sports-Reference by a make or two; when both exist, Sports-Reference is the published number.
Shot locations are not the ranking. They are one season's logged attempts. Mid-major coverage is whatever ESPN stored.
One 3-point line on the maps. NCAA men's distance was 20 ft 9 in through 2018-19 and 22 ft 1.75 in from 2019-20. The charts draw the later line only. A 2018-19 three just inside today's arc is still a three because ESPN marked the play, not because of the overlay.
College Basketball Reference, https://www.sports-reference.com/cbb/ — players_totals and players_advanced on resolved slugs. Cached under data/blog/raw/sr/. ↩
Sports-Reference Stathead player-season finder returned HTTP 403 (Cloudflare) from this fetch. We did not retry behind a browser or invent a combined table. ↩