Game Predictions | March Madness 2019

The tables below provide our predictions for all games in the 2019 NCAA Tournament. The tables will be updated after each round of the tournament, with new games/predictions added as they become available. A brief description of how to interpret the predictions is provided above each table, and a summary of the historical accuracy of each model is provided at the end of the post.

Win Probability & Spread Predictions

Win Probabilities – For each game, “Team1 Win Prob” represents our model’s probability that Team1 (the better-seeded team) wins the game. Probabilities below 50 indicate that Team2 is more likely to win the game.

Spreads – The “284 Spread” is our model’s prediction of how many points Team1 will win by. For both Vegas and 284 Spreads, negative values indicate Team1 is favored, while positives indicate Team2 is favored. The “Team1 Cover Prob” is our model’s probability that Team1 covers the Vegas Spread (a Cover Prob below 50 indicates that Team2 is more likely to cover than Team1). Note that our 284 Spread and Cover Prob models are generated separately, so there will be games where they disagree (e.g., one predicts Team1 covers and the other predicts Team2 covers).

Team1Team2RdTeam1
Win Prob
Vegas
Spread
284
Spread
Team1
Cover Prob
1 Virginia3 TX Tech668-1.5-4.168
2 Mich St3 TX Tech569-3-3.462
1 Virginia5 Auburn541-5.5+0.526
1 Duke2 Mich St461-2.5-0.442
1 Gonzaga3 TX Tech485-4.5-10.166
1 Virginia3 Purdue470-4-4.353
2 UK5 Auburn439-4+323
1 Duke4 VA Tech375-7-3.539
2 Mich St3 LSU362-6+0.248
1 Gonzaga4 FL St367-7.5-3.727
2 Mich3 TX Tech379-2-4.168
1 Virginia12 Oregon362-8.5-4.532
2 Tenn3 Purdue372-1-3.672
1 UNC5 Auburn337-5+1.520
2 UK3 Houston353-2.5-1.738
1 Duke9 UCF296-13-14.243
4 VA Tech12 Liberty254-9-6.430
3 LSU6 Maryland269-3-6.256
2 Mich St10 Minn272-10-7.534
1 Gonzaga9 Baylor293-14-14.632
4 FL St12 Murray St290-4.5-11.680
3 TX Tech6 Buffalo244-3.5+0.936
2 Mich10 FL281-6.5-4.842
1 Virginia9 Oklahoma290-11.5-11.245
12 Oregon13 UC Irvine264-5-4.937
3 Purdue6 Villanova235-3.5+3.427
2 Tenn10 Iowa282-8-8.355
1 UNC9 Wash266-11.5-3.823
4 Kansas5 Auburn228+2+4.238
3 Houston11 Ohio St282-5.5-853
2 UK7 Wofford269-5-339
1 Duke16 NDSU197-27-20.733
8 VCU9 UCF167+1-6.672
5 Miss St12 Liberty162-6.5-7.945
4 VA Tech13 St Louis145-10.5-3.818
6 Maryland11 Belmont178-3-4.556
3 LSU14 Yale197-7.5-19.184
7 Louisville10 Minn132-5027
2 Mich St15 Bradley191-18-17.144
1 Gonzaga16 Fair Dick198-27.5-24.751
8 Syracuse9 Baylor174-2-2.848
5 Marquette12 Murray St184-4.5-6.961
4 Florida St13 Vermont183-9-10.653
6 Buffalo11 AZ St168-6.5-5.239
3 Texas Tech14 North KY183-14-8.531
7 Nevada10 Florida167-2-2.755
2 Michigan15 Montana193-15.5-12.430
1 Virginia16 G Webb190-23.5-18.363
8 Ole Miss9 Oklahoma165-1.5-5.364
5 Wisconsin12 Oregon138-1.5+3.226
4 Kansas St13 UC Irvine153-4.5-2.226
6 Villanova11 Sr Marys170-5.5-539
3 Purdue14 Old Dom174-12.5-9.431
7 Cincinnati10 Iowa161-3.5+0.145
2 Tennessee15 Colgate187-17.5-11.635
1 UNC16 Iona199-24.5-22.956
8 Utah St9 Wash133-2.5+6.435
5 Auburn12 N Mex St190-6-1055
4 Kansas13 Neastern178-6.5-6.655
6 Iowa St11 Ohio St181-5.5-7.950
3 Houston14 Georgia St191-12-10.641
7 Wofford10 Seton Hall172-2.5-5.958
2 Kentucky15 Abil Christ183-22-11.429
16 NDSU16 NC Central036-5+1.924
11 Belmont11 Temple041-3+0.944
16 Fair Dick16 Prair View046-1.5-0.839
11 AZ St11 St Johns043-1+0.737

Total Predictions

The Vegas Total can be compared to our “284 Total” (our model’s prediction of how many points will be scored in the game) as well as our “Over Prob” (our model’s probability that the game goes over the Vegas Total). An Over Prob below 50 indicates that an Under is more likely. Note that our 284 Total and Over Prob models are generated separately, so there will be games where they disagree (e.g., one predicts over and the other predicts under):

Team1Team2RdVegas
Total
284
Total
Over
Prob
1 Virginia3 TX Tech6118122.158.0
2 Mich St3 TX Tech5133135.253.8
1 Virginia5 Auburn5130.5135.965.0
1 Duke2 Mich St4150147.445.6
1 Gonzaga3 TX Tech4140141.653.3
1 Virginia3 Purdue4127131.358.1
2 UK5 Auburn4142142.753.1
1 Duke4 VA Tech3144.5144.348.6
2 Mich St3 LSU3149154.359.6
1 Gonzaga4 FL St3147.5142.537.1
2 Mich3 TX Tech3125126.450.4
1 Virginia12 Oregon3118126.365.8
2 Tenn3 Purdue3146.5150.461.3
1 UNC5 Auburn3163166.264
2 UK3 Houston3135.5132.339.6
1 Duke9 UCF2143.5145.952
4 VA Tech12 Liberty212612851
3 LSU6 Maryland2147146.353
2 Mich St10 Minn2142146.955
1 Gonzaga9 Baylor2148143.540
4 FL St12 Murray St2145145.351
3 TX Tech6 Buffalo2146149.462
2 Mich10 FL2123127.957
1 Virginia9 Oklahoma2127.5128.251
12 Oregon13 UC Irvine2123121.943
3 Purdue6 Villanova2137148.157
2 Tenn10 Iowa2156165.576
1 UNC9 Wash2147.5150.856
4 Kansas5 Auburn2147.5150.556
3 Houston11 Ohio St2131.5130.748
2 UK7 Wofford2139.5137.148
1 Duke16 NDSU1148.5151.556
8 VCU9 UCF1126.513352
5 Miss St12 Liberty1134136.645
4 VA Tech13 St Louis112612650
6 Maryland11 Belmont1148145.344
3 LSU14 Yale1156.515643
7 Louisville10 Minn1136137.351
2 Mich St15 Bradley1133.5139.854
1 Gonzaga16 Fair Dick1152.5152.449
8 Syracuse9 Baylor1131.5126.542
5 Marquette12 Murray St1149.5147.850
4 Florida St13 Vermont1133.5138.452
6 Buffalo11 AZ St1155.5161.965
3 Texas Tech14 North KY1137.5136.149
7 Nevada10 Florida1133140.866
2 Michigan15 Montana1130.5136.854
1 Virginia16 G Webb1130.5138.560
8 Ole Miss9 Oklahoma1142.5141.443
5 Wisconsin12 Oregon1117125.462
4 Kansas St13 UC Irvine1119123.562
6 Villanova11 Sr Marys1130.5127.450
3 Purdue14 Old Dom1126129.549
7 Cincinnati10 Iowa1138149.667
2 Tennessee15 Colgate1148157.361
1 UNC16 Iona1166169.358
8 Utah St9 Wash1135143.263
5 Auburn12 N Mex St1146140.538
4 Kansas13 Neastern1143148.951
6 Iowa St11 Ohio St1140.5136.348
3 Houston14 Georgia St1141.5150.870
7 Wofford10 Seton Hall1142.5141.348
2 Kentucky15 Abil Christ1132123.339
16 NDSU16 NC Central0133126.541
11 Belmont11 Temple0155.5159.354
16 Fair Dick16 Prair View0149147.447
11 AZ St11 St Johns0152.5149.848

Model Accuracy

Below is a summary of how each individual model performed on all 2017-2019 tournament games, with 2019 results being added to the tables as they become available. Note that spread/cover and total/over models are graded against the numbers as they are shown in the tables above. Additionally, links to detailed recaps of all predictions from the 2017 and 2018 tournaments can be found directly below:

Year Win Prob 284 Spread Cover Prob
Rec % Rec % Rec %
2017 47-15 0.76 34-26 0.57 42-21 0.67
2018 44-19 0.70 32-28 0.53 35-27 0.56
2019 45-18 0.71 24-36 0.40 27-33 0.45
Total 136-52 0.72 90-90 0.50 104-81 0.56

Year 284 Total Over Prob
Rec % Rec %
2017 39-22 0.64 42-21 0.67
2018 38-25 0.60 36-27 0.58
2019 32-29 0.52 32-29 0.52
Total 109-76 0.59 110-77 0.59