Who will win Harvard vs Duke?
The model gives Duke a >99% chance to beat Harvard, with a projected score of Duke 88, Harvard 56.
Half of games like thisFour in five
Calls of 90–100% have been right 95.4% of the time in 11,767 past games, and the typical miss on the margin is 8.8 points. After the final, the result is pinned to this strip. As of 11:23 AM ET on November 21, 2026. Updated three times a day until tip-off.
Players to watch
| Player | Pts | Reb | Ast | Last games | Chance of |
|---|---|---|---|---|---|
| Ngongba IIC · 24 min | 12.8 | 6.8 | 2.7 | 15+ 35% | |
| Boumtje BoumtjeC · 22 min | 11.1 | 6.2 | 0.8 | 10+ 56% | |
| BlackwellG · 24 min | 10.9 | 3.1 | 1.5 | 10+ 55% | |
| FosterG · 21 min | 10.0 | 3.3 | 2.4 | 10+ 48% | |
| Batties IIF · 29 min | 9.9 | 3.1 | 1.4 | 10+ 48% | |
| HintonG · 28 min | 9.7 | 2.7 | 1.2 | 10+ 46% | |
| BarbourG · 30 min | 9.6 | 3.1 | 0.9 | 10+ 45% | |
| SingletonG · 20 min | 7.2 | 2.0 | 1.3 | 5+ 66% | |
Bars: points in the player's latest games this season. Blue line: tonight's projection. Every regular's line
The game without one player
Without Patrick Ngongba II: Harvard 1%Duke 99%
The white mark is the call with him: Duke >99%.
| Player | Points a game he adds | Duke without him | Change | Projected score |
|---|---|---|---|---|
| Patrick Ngongba II | 5.3 | 99% | −1 | 58–85 |
| John Blackwell | 4.1 | 99% | −1 | 58–86 |
| Dame Sarr | 3.8 | >99% | 0 | 58–86 |
| Cayden Boozer | 3.8 | >99% | 0 | 58–86 |
| Thomas Batties II | 0.8 | >99% | 0 | 55–88 |
| Robert Hinton | 0.8 | >99% | 0 | 55–88 |
| Tey Barbour | 0.6 | >99% | 0 | 55–88 |
Change is in Duke's win chance, in points. Scores read Harvard, then Duke. A player's value is what the team is better by with him than with a replacement, from the player ratings.
Model against the line
Your call
All of today's gamesPlay-by-play
Team numbers
Form
Harvard 3-4LLLLW
- Nov 20W 77-65at Holy CrossCovered −11.5 · Over 137.5
- Nov 18L 69-74vs Boston CollegeDidn't cover +4.5 · Over 128.5
- Nov 15L 64-70at AmericanDidn't cover +1.5 · Over 130.5
- Nov 13L 57-85at G WashingtonDidn't cover +10.5 · Over 133.5
- Nov 10L 70-74at NortheasternDidn't cover −1.5 · Over 143.5
- Last five
- 1–4, scoring 67.4 and allowing 73.6
- Against the spread, this season
- 1–4
- Totals, this season
- 5 over, 0 under
Duke 3-2WWLWL
- Nov 17L 78-99vs IllinoisDidn't cover −8.5 · Over 167.5
- Nov 13W 90-47vs SIUECovered −42.5 · Under 148.5
- Nov 10L 73-79vs Michigan StDidn't cover −7.5 · Over 147.5
- Nov 5W 113-52vs CoastalCovered −40.5 · Over 149.5
- Nov 2W 84-50vs Army
- Last five
- 3–2, scoring 87.6 and allowing 65.4
- Against the spread, this season
- 2–2
- Totals, this season
- 3 over, 1 under
The spread and the total are the lines logged with each prediction. Form is what happened; the ratings above already count recent games more.
Offense against defense
| Harvard | Duke | |
|---|---|---|
| Points scored per 100 possessions | 103.2 217th | 121.7 9th |
| Points allowed per 100 | 109.4 186th | 89.8 2nd |
| Possessions a game | 67.3 361st | 71.1 92nd |
| Shooting (effective field-goal %) | 51.6% 118th | 50.4% 195th |
| Opponents' shooting | 51.0% 216th | 46.1% 9th |
| Turnovers, share of possessions | 18.3% 215th | 18.4% 233rd |
| Offensive rebounds, share of misses | 32.0% 185th | 35.4% 35th |
| Free throws per shot | 0.23 300th | 0.31 33rd |
The widest gap: Duke's defense (2nd) against Harvard's offense (217th), of 365 teams.
First half
- Projected halftime score
- Harvard 25, Duke 42
- First-half margin
- Duke by 17.5
- First-half points, both teams
- 67.3
- Projected second half
- Harvard 31, Duke 45
- Second-half points, both teams
- 76.1
- Tied at the half
- 4%
- Harvard, first halves so far
- Led 4 of 7 · 33.1 scored, 30.9 allowed
- Duke, first halves so far
- Led 3 of 5 · 37.4 scored, 26.4 allowed
Our own projection, not a sportsbook's first-half line. On 11,516 past games it had not seen, the halftime leader was right 67.2% of the time and the first-half total was off by 8.1 points on average. This season: 610 of 833 halftime leaders right. Scorecard
Player projections
Harvard
| Player | Min | Pts | Reb | Ast | 3s |
|---|---|---|---|---|---|
| Tey BarbourG | 30 | 9.6 | 3.1 | 0.9 | 1.6 |
| 10+ pts 45% · 3+ reb 56% · 2+ threes 45% | |||||
| Thomas Batties IIF | 29 | 9.9 | 3.1 | 1.4 | 1.3 |
| 10+ pts 48% · 3+ reb 56% · 2+ ast 40% | |||||
| Robert HintonG | 28 | 9.7 | 2.7 | 1.2 | 0.8 |
| 10+ pts 46% · 3+ reb 48% · 1+ threes 51% | |||||
| Ben EisendrathG | 27 | 4.4 | 2.5 | 2.4 | 0.5 |
| 5+ pts 40% · 3+ reb 42% · 3+ ast 42% | |||||
| Jason SingletonG | 20 | 7.2 | 2.0 | 1.3 | 0.4 |
| 5+ pts 66% · 2+ reb 53% · 2+ ast 35% | |||||
| Luca Ace-NasteskiF | 18 | 3.6 | 3.2 | 0.5 | 0.1 |
| 5+ pts 32% · 3+ reb 58% | |||||
| Frankie ManninoG | 11 | 1.5 | 1.2 | 0.5 | 0.1 |
| 2+ reb 33% | |||||
| Julian AdigunC | 11 | 2.0 | 2.0 | 0.2 | 0.1 |
| 5+ pts 15% · 2+ reb 54% | |||||
Duke
| Player | Min | Pts | Reb | Ast | 3s |
|---|---|---|---|---|---|
| John BlackwellG | 24 | 10.9 | 3.1 | 1.5 | 2.4 |
| 10+ pts 55% · 3+ reb 55% · 2+ ast 43% | |||||
| Patrick Ngongba IIC | 24 | 12.8 | 6.8 | 2.7 | 0.6 |
| 15+ pts 35% · 7+ reb 50% · 3+ ast 48% | |||||
| Dame SarrF | 22 | 8.9 | 3.4 | 1.8 | 0.7 |
| 10+ pts 40% · 2+ ast 51% · 1+ threes 48% | |||||
| Joaquim Boumtje BoumtjeC | 22 | 11.1 | 6.2 | 0.8 | 0.9 |
| 10+ pts 56% · 6+ reb 56% · 1+ threes 54% | |||||
| Caleb FosterG | 21 | 10.0 | 3.3 | 2.4 | 1.8 |
| 10+ pts 48% · 3+ ast 42% · 2+ threes 50% | |||||
| Cayden BoozerG | 21 | 7.6 | 3.3 | 4.0 | 0.5 |
| 5+ pts 69% · 3+ reb 59% · 4+ ast 54% | |||||
| Sebastian WilkinsF | 18 | 7.5 | 4.3 | 1.1 | 0.5 |
| 5+ pts 69% · 4+ reb 58% · 1+ threes 37% | |||||
| Deron Rippey Jr.G | 16 | 4.2 | 1.8 | 2.2 | 0.7 |
| 5+ pts 37% · 2+ reb 48% · 1+ threes 46% | |||||
Our own projections from each player's minutes and production, this opponent and the game's pace: not a sportsbook's lines. Under each name, the round numbers the player is about as likely as not to reach. How they have done
Win probability
Game chat
Loading the chat…