2025-03-30

The Data Set

  • For this project I have chosen to use college basketball data sets containing data for the 2022-2024 seasons
  • Variables for study include:
    • Team/Conference
    • Wins/Games
    • Offensive/Defensive Efficiency
    • Postseason Finish/Tournament Seed
  • Data Source: kaggle

Important Data Definitions

  • Important abbreviations used in the data set and code include:
    • ADJOE- Adjusted Offensive Efficiency: estimate of points scored per 100 possessions a team would have against average D1 defense
    • ADJDE- Adjusted Defensive Efficiency: estimate of points allowed per 100 possessions a team would have against average D1 offense
    • BARTHAG- Power Rating: chance of beating an average D1 team
    • TOR- Turnover Percentage
    • ADJ_T- Adjusted Tempo: estimate of possession per 40 minutes (tempo) a team would have against the team that wants to play at the average D1 tempo

Data Filtering 1

  • CSV files for 2022, 2023, and 2024 college basketball data were merged together
  • To narrow the focus of study the data was filtered to only include the 68 teams that qualified for the NCAA tournament each year
  • Additional variables were created for:
    • Win Percentage
    • Average Power Rating per Tournament Placement per Year
  • Data not used in analysis was removed from the data frame
  • The following slides contain the set-up code and a preview of the data frame

Data Filtering 2- R Code

#Data Sets:
CBB22_df = read.csv("cbb22.csv", sep = ",", header = TRUE)
CBB23_df = read.csv("cbb23.csv", sep = ",", header = TRUE)
CBB24_df = read.csv("cbb24.csv", sep = ",", header = TRUE)
CBB22_df <- CBB22_df %>%
  mutate(YEAR = 2022) %>%
  mutate(WIN = W/G * 100)
CBB23_df <- CBB23_df %>%
  mutate(YEAR = 2023) %>%
  mutate(WIN = W/G * 100)
CBB24_df <- CBB24_df %>%
  mutate(YEAR = 2024) %>%
  mutate(WIN = W/G * 100)
CBB <- bind_rows(CBB22_df, CBB23_df, CBB24_df)
CBB <- CBB %>% 
  filter(SEED != 'N/A') %>%
  select(-EFG_O, -EFGD_D, -TORD, -ORB, -DRB, -FTR, -FTRD, -X2P_O, -X2P_D,
         -X3P_O, -X3P_D, -EFG_D, -EFG., -EFGD., -G, -W, -WAB)
CBB$SEED = as.numeric(CBB$SEED)
#Create custom variables for further analysis
# Variable for average BARTHAG by seed
CBB <- CBB %>%
  group_by(YEAR,POSTSEASON) %>%
  mutate(avgBARTHAG = mean(BARTHAG)) %>%
  ungroup()
as.data.frame(CBB)
##                       TEAM CONF ADJOE ADJDE BARTHAG  TOR ADJ_T POSTSEASON SEED
## 1                  Gonzaga  WCC 120.3  89.9  0.9662 15.7  72.6        S16    1
## 2                   Kansas  B12 119.8  91.3  0.9580 17.3  69.1  Champions    1
## 3                   Baylor  B12 117.8  91.3  0.9491 17.9  67.6        R32    1
## 4                  Arizona  P12 118.0  93.7  0.9341 18.3  72.3        S16    1
## 5                     Duke  ACC 122.7  96.0  0.9440 15.0  67.4         F4    2
## 6                Villanova   BE 117.2  93.0  0.9346 15.5  62.2         F4    2
## 7                 Kentucky  SEC 119.4  95.4  0.9295 16.7  67.8        R64    2
## 8                   Auburn  SEC 111.7  90.0  0.9228 16.7  70.0        R32    2
## 9               Texas Tech  B12 110.5  85.4  0.9509 19.8  66.3        S16    3
## 10               Tennessee  SEC 111.3  87.1  0.9439 17.7  67.4        R32    3
## 11                  Purdue  B10 121.6  97.5  0.9267 17.5  66.0        S16    3
## 12               Wisconsin  B10 108.7  93.4  0.8511 13.0  66.4        R32    3
## 13                    UCLA  P12 115.8  92.2  0.9321 13.2  65.4        S16    4
## 14                Illinois  B10 111.9  93.0  0.8943 18.1  67.1        R32    4
## 15                Arkansas  SEC 108.7  90.7  0.8894 16.9  70.6         E8    4
## 16              Providence   BE 111.4  95.5  0.8554 17.0  65.3        S16    4
## 17                 Houston Amer 116.5  88.5  0.9595 16.9  63.7         E8    5
## 18                    Iowa  B10 121.1  97.2  0.9265 13.0  69.8        R64    5
## 19            Saint Mary's  WCC 109.8  90.5  0.9024 17.0  63.7        R32    5
## 20             Connecticut   BE 113.2  95.9  0.8709 17.3  65.0        R64    5
## 21                   Texas  B12 112.5  91.5  0.9152 17.8  63.5        R32    6
## 22                     LSU  SEC 106.3  89.6  0.8766 20.7  68.7        R64    6
## 23                 Alabama  SEC 115.2  98.3  0.8612 19.8  71.8        R64    6
## 24            Colorado St.  MWC 111.9  98.1  0.8190 15.2  65.6        R64    6
## 25                Ohio St.  B10 115.5  98.5  0.8612 17.1  65.1        R32    7
## 26            Michigan St.  B10 111.6  96.6  0.8405 18.8  67.6        R32    7
## 27              Murray St.  OVC 110.0  96.2  0.8241 17.1  66.0        R32    7
## 28                     USC  P12 108.6  96.2  0.8021 17.9  66.2        R64    7
## 29          North Carolina  ACC 114.4  93.5  0.9113 16.4  70.1        2ND    8
## 30           San Diego St.  MWC 102.4  86.9  0.8675 19.3  65.3        R64    8
## 31               Boise St.  MWC 107.6  92.9  0.8434 18.4  64.5        R64    8
## 32              Seton Hall   BE 105.4  93.4  0.8000 17.7  67.6        R64    8
## 33                 Memphis Amer 112.4  92.6  0.9028 22.5  70.0        R32    9
## 34                     TCU  B12 108.6  92.1  0.8698 21.4  65.9        R32    9
## 35               Creighton   BE 104.8  92.1  0.8151 20.4  67.0        R32    9
## 36               Marquette   BE 107.7  96.5  0.7787 17.2  71.2        R64    9
## 37           San Francisco  WCC 110.2  93.0  0.8748 17.9  69.0        R64   10
## 38                Miami FL  ACC 113.6  97.9  0.8467 14.0  67.7         E8   10
## 39          Loyola Chicago  MVC 106.9  92.3  0.8439 18.5  64.6        R64   10
## 40                Davidson  A10 115.8 101.1  0.8258 14.8  64.6        R64   10
## 41           Virginia Tech  ACC 115.6  95.2  0.9028 17.0  63.3        R64   11
## 42                Michigan  B10 113.9  96.2  0.8741 17.3  66.4        S16   11
## 43              Notre Dame  ACC 112.8  96.7  0.8542 15.6  65.4        R32   11
## 44                Iowa St.  B12 101.7  88.4  0.8346 20.4  66.3        S16   11
## 45                 Rutgers  B10 106.6  96.3  0.7629 17.3  65.0        R68   11
## 46                 Indiana  B10 107.7  91.6  0.8652 17.2  67.2        R64   12
## 47                     UAB CUSA 111.8  99.1  0.8001 16.4  68.9        R64   12
## 48                 Wyoming  MWC 107.9  97.3  0.7661 17.1  65.8        R68   12
## 49          New Mexico St.  WAC 106.0  96.0  0.7579 20.8  66.1        R32   12
## 50                Richmond  A10 106.9  99.1  0.7071 14.3  66.7        R32   12
## 51                 Vermont   AE 109.7  97.3  0.7984 14.7  64.8        R64   13
## 52             Chattanooga   SC 107.3  97.3  0.7557 17.0  64.6        R64   13
## 53        South Dakota St.  Sum 114.1 103.5  0.7542 15.3  69.6        R64   13
## 54                   Akron  MAC 105.5 101.7  0.6041 17.7  61.9        R64   13
## 55                 Colgate  Pat 107.0 103.2  0.6026 16.9  67.0        R64   14
## 56                Longwood BSth 105.7 103.7  0.5541 18.4  65.8        R64   14
## 57                    Yale  Ivy  99.4  97.9  0.5449 18.3  68.9        R64   14
## 58             Montana St. BSky 103.6 102.5  0.5308 18.7  67.1        R64   14
## 59           Saint Peter's MAAC  99.4  93.1  0.6786 20.1  65.6         E8   15
## 60                Delaware  CAA 106.5 103.2  0.5908 18.3  66.6        R64   15
## 61        Jacksonville St. ASun 104.5 102.2  0.5638 19.7  65.1        R64   15
## 62       Cal St. Fullerton   BW 102.0 103.1  0.4675 17.9  66.0        R64   15
## 63             Georgia St.   SB 101.9  98.4  0.5983 16.6  67.3        R64   16
## 64                  Bryant  NEC 102.9 104.1  0.4651 19.0  72.3        R68   16
## 65              Wright St. Horz 104.8 106.2  0.4613 17.7  68.8        R64   16
## 66          Texas Southern SWAC  96.3  98.5  0.4356 21.4  68.1        R64   16
## 67             Norfolk St. MEAC 100.7 103.0  0.4342 20.1  67.0        R64   16
## 68  Texas A&M Corpus Chris Slnd  96.4 102.8  0.3223 19.7  69.2        R68   16
## 69                  Kansas  B12 113.8  92.3  0.9177 17.5  69.0        R32    1
## 70                  Purdue  B10 117.9  93.7  0.9333 17.2  63.9        R64    1
## 71                 Alabama  SEC 116.1  89.0  0.9548 18.7  72.6        S16    1
## 72                 Houston Amer 117.3  90.3  0.9532 15.4  63.4        S16    1
## 73                   Texas  B12 116.1  91.8  0.9371 16.3  68.5         E8    2
## 74               Marquette   BE 118.7  96.0  0.9197 15.3  68.4        R32    2
## 75                 Arizona  P12 119.4  95.4  0.9292 18.4  72.2        R64    2
## 76                    UCLA  P12 114.9  88.5  0.9528 14.9  66.1        S16    2
## 77              Kansas St.  B12 112.6  94.7  0.8803 19.3  69.8         E8    3
## 78                 Gonzaga  WCC 123.2  99.0  0.9251 14.8  70.1         E8    3
## 79                  Baylor  B12 121.7 100.6  0.8991 17.9  66.9        R32    3
## 80                  Xavier   BE 118.0  98.5  0.8890 17.2  70.5        S16    3
## 81             Connecticut   BE 121.5  91.2  0.9643 18.9  66.7  Champions    4
## 82                 Indiana  B10 113.1  96.9  0.8557 16.8  67.9        R32    4
## 83                Virginia  ACC 110.0  94.0  0.8591 13.6  61.3        R64    4
## 84               Tennessee  SEC 111.2  87.5  0.9401 18.1  65.1        S16    4
## 85           San Diego St.  MWC 110.8  89.5  0.9205 17.3  65.7        2ND    5
## 86                Miami FL  ACC 119.1 100.9  0.8715 16.1  68.6         F4    5
## 87            Saint Mary's  WCC 112.1  90.5  0.9213 16.4  61.9        R32    5
## 88                    Duke  ACC 113.3  92.9  0.9072 18.3  64.9        R32    5
## 89               Creighton   BE 115.0  93.2  0.9177 16.6  67.7         E8    6
## 90                Kentucky  SEC 115.7  97.7  0.8738 17.4  65.8        R32    6
## 91                     TCU  B12 111.7  94.6  0.8704 16.7  69.7        R32    6
## 92                Iowa St.  B12 107.0  90.9  0.8678 19.6  64.0        R64    6
## 93            Northwestern  B10 109.7  93.4  0.8642 14.7  64.5        R32    7
## 94                Missouri  SEC 117.5 104.5  0.7939 15.8  68.9        R32    7
## 95               Texas A&M  SEC 113.8  95.7  0.8808 18.3  66.0        R64    7
## 96            Michigan St.  B10 114.4  96.4  0.8771 16.4  64.6        S16    7
## 97                Maryland  B10 112.6  95.8  0.8658 16.5  64.3        R32    8
## 98                    Iowa  B10 119.9 103.9  0.8385 14.0  69.6        R64    8
## 99                 Memphis Amer 115.4  95.0  0.9030 18.3  71.0        R64    8
## 100               Arkansas  SEC 111.9  93.3  0.8894 18.1  69.5        S16    8
## 101       Florida Atlantic CUSA 114.0  95.8  0.8815 16.9  67.5         F4    9
## 102                 Auburn  SEC 113.1  94.9  0.8829 17.7  67.6        R32    9
## 103          West Virginia  B12 117.2  97.5  0.8933 18.5  68.4        R64    9
## 104               Illinois  B10 108.9  95.4  0.8216 18.4  69.8        R64    9
## 105               Penn St.  B10 117.5 100.7  0.8547 13.7  64.3        R32   10
## 106                    USC  P12 111.8  97.4  0.8298 18.1  68.0        R64   10
## 107              Boise St.  MWC 110.1  95.6  0.8361 16.6  66.3        R64   10
## 108               Utah St.  MWC 115.4  98.0  0.8670 17.6  68.6        R64   10
## 109             Pittsburgh  ACC 113.1 101.6  0.7743 16.3  67.2        R32   11
## 110             Providence   BE 114.9 101.2  0.8128 16.8  66.8        R64   11
## 111     North Carolina St.  ACC 113.6  99.2  0.8272 13.5  69.2        R64   11
## 112            Arizona St.  P12 108.4  95.0  0.8207 16.8  68.1        R64   11
## 113        Mississippi St.  SEC 103.4  91.8  0.7985 19.4  63.6        R68   11
## 114                 Nevada  MWC 109.7  99.5  0.7542 15.3  65.9        R68   11
## 115                  Drake  MVC 108.8  95.6  0.8155 15.5  66.3        R64   12
## 116                    VCU  A10 105.5  93.4  0.8026 19.6  67.9        R64   12
## 117           Oral Roberts  Sum 111.8 101.6  0.7487 13.3  70.2        R64   12
## 118  College of Charleston  CAA 109.4  97.7  0.7865 16.7  71.0        R64   12
## 119                 Furman   SC 112.7 104.8  0.6984 15.7  68.2        R32   13
## 120    Louisiana Lafayette   SB 110.8 101.6  0.7299 17.7  68.4        R64   13
## 121                   Iona MAAC 110.3  98.3  0.7909 15.1  68.5        R64   13
## 122               Kent St.  MAC 107.0  97.0  0.7560 16.3  67.8        R64   13
## 123           Grand Canyon  WAC 111.3 106.0  0.6388 18.1  64.9        R64   14
## 124            Montana St. BSky 105.1  98.2  0.6871 17.8  66.3        R64   14
## 125           Kennesaw St. ASun 105.6 100.3  0.6428 17.8  68.4        R64   14
## 126       UC Santa Barbara   BW 110.0 105.7  0.6126 17.0  64.4        R64   14
## 127                Vermont   AE 109.7 102.5  0.6873 14.3  64.5        R64   15
## 128                Colgate  Pat 112.5 106.1  0.6607 15.0  67.5        R64   15
## 129          UNC Asheville BSth 102.3 103.5  0.4657 20.1  68.2        R64   15
## 130              Princeton  Ivy 109.1 101.0  0.7083 16.5  67.0        S16   15
## 131    Fairleigh Dickinson  NEC 106.1 114.7  0.2909 16.3  68.6        R32   16
## 132      Northern Kentucky Horz 102.6 101.4  0.5349 17.3  61.9        R64   16
## 133                 Howard MEAC 102.7 106.3  0.4031 22.8  69.0        R64   16
## 134 Texas A&M Corpus Chris Slnd 105.8 107.4  0.4566 17.7  68.9        R64   16
## 135         Texas Southern SWAC  95.4 105.4  0.2394 20.3  69.2        R68   16
## 136 Southeast Missouri St.  OVC  99.4 107.8  0.2823 17.3  72.0        R68   16
## 137                Houston  B12 119.2  85.5  0.9785 13.7  63.3        S16    1
## 138            Connecticut   BE 127.1  93.6  0.9712 14.9  64.6  Champions    1
## 139                 Purdue  B10 126.2  94.7  0.9644 16.5  67.6        2ND    1
## 140         North Carolina  ACC 116.8  93.2  0.9305 14.4  70.4        S16    1
## 141               Iowa St.  B12 113.6  86.5  0.9583 15.7  67.6        S16    2
## 142                Arizona  P12 121.5  93.6  0.9526 16.1  72.0        S16    2
## 143              Tennessee  SEC 115.6  91.2  0.9382 14.6  69.2         E8    2
## 144              Marquette   BE 118.9  94.6  0.9328 14.3  69.1        S16    2
## 145              Creighton   BE 120.6  96.5  0.9289 15.3  66.9        S16    3
## 146               Illinois  B10 125.3 101.1  0.9221 15.1  69.9         E8    3
## 147                 Baylor  B12 122.2  99.9  0.9103 17.7  65.7        R32    3
## 148               Kentucky  SEC 123.0 102.2  0.8934 14.2  72.5        R64    3
## 149                 Auburn  SEC 120.7  92.1  0.9573 14.9  69.8        R64    4
## 150                   Duke  ACC 120.8  96.8  0.9272 14.3  66.9         E8    4
## 151                Alabama  SEC 125.2 102.1  0.9126 16.0  72.5         F4    4
## 152                 Kansas  B12 113.5  92.7  0.9111 16.4  68.9        R32    4
## 153           Saint Mary's  WCC 115.8  93.7  0.9198 16.0  62.1        R64    5
## 154                Gonzaga  WCC 121.3  98.6  0.9151 14.0  69.1        S16    5
## 155              Wisconsin  B10 119.8  98.8  0.9018 14.6  65.0        R64    5
## 156          San Diego St.  MWC 111.5  93.9  0.8789 16.0  66.1        S16    5
## 157                    BYU  B12 120.6  99.0  0.9067 15.2  69.0        R64    6
## 158             Texas Tech  B12 117.5  98.4  0.8846 16.5  66.4        R64    6
## 159                Clemson  ACC 116.7  99.3  0.8647 14.9  66.2         E8    6
## 160         South Carolina  SEC 114.4  98.9  0.8420 15.5  63.0        R64    6
## 161                  Texas  B12 119.0  98.3  0.9006 16.6  67.2        R32    7
## 162                Florida  SEC 119.7  99.3  0.8951 15.8  71.6        R64    7
## 163         Washington St.  P12 112.8  96.6  0.8556 16.1  64.9        R32    7
## 164                 Dayton  A10 117.4 100.7  0.8547 15.8  64.0        R32    7
## 165               Nebraska  B10 115.7  95.6  0.8999 15.9  68.7        R64    8
## 166        Mississippi St.  SEC 112.6  94.9  0.8772 18.7  67.2        R64    8
## 167       Florida Atlantic Amer 117.3 103.0  0.8170 15.8  67.9        R64    8
## 168               Utah St.  MWC 114.1 100.4  0.8144 15.7  68.5        R32    8
## 169           Michigan St.  B10 112.7  93.3  0.8983 14.7  65.3        R32    9
## 170           Northwestern  B10 117.2  99.2  0.8717 13.3  63.7        R32    9
## 171                    TCU  B12 114.5  97.1  0.8690 17.9  69.9        R64    9
## 172              Texas A&M  SEC 115.0  98.9  0.8499 14.2  65.6        R32    9
## 173               Colorado  P12 117.2  98.0  0.8866 18.1  67.7        R32   10
## 174                 Nevada  MWC 115.1  98.0  0.8632 15.0  66.8        R64   10
## 175              Boise St.  MWC 114.0  97.8  0.8538 16.5  66.6        R68   10
## 176           Colorado St.  MWC 113.8  98.1  0.8460 14.8  65.8        R64   10
## 177                  Drake  MVC 114.8 100.2  0.8283 13.9  67.8        R64   10
## 178               Virginia  ACC 104.6  92.0  0.8147 13.2  60.0        R68   10
## 179             New Mexico  MWC 114.8  95.5  0.8927 13.9  72.8        R64   11
## 180                 Oregon  P12 113.3  99.6  0.8149 14.8  67.1        R32   11
## 181     North Carolina St.  ACC 114.0 101.2  0.7981 13.7  67.9         F4   11
## 182               Duquesne  A10 106.3  96.1  0.7612 17.5  66.3        R32   11
## 183          James Madison   SB 112.0 100.4  0.7780 14.7  69.8        R32   12
## 184           Grand Canyon  WAC 111.1  99.8  0.7733 17.8  68.0        R32   12
## 185            McNeese St. Slnd 111.3 101.0  0.7524 14.2  65.3        R64   12
## 186                    UAB Amer 112.7 106.3  0.6634 16.7  67.7        R64   12
## 187                   Yale  Ivy 109.8 100.1  0.7437 14.1  64.6        R32   13
## 188                Samford   SC 111.7 102.4  0.7315 18.1  72.6        R64   13
## 189  College of Charleston  CAA 113.3 105.2  0.7019 14.3  70.0        R64   13
## 190                Vermont   AE 104.5  98.7  0.6593 13.8  63.3        R64   13
## 191                  Akron  MAC 105.0 101.6  0.5953 17.2  65.7        R64   14
## 192           Morehead St.  OVC 106.7 103.4  0.5889 18.5  64.3        R64   14
## 193                Oakland Horz 107.8 104.6  0.5853 16.2  66.8        R32   14
## 194                Colgate  Pat 102.3 101.8  0.5128 16.3  66.2        R64   14
## 195       Western Kentucky CUSA 105.2 101.8  0.5938 18.3  75.1        R64   15
## 196       South Dakota St.  Sum 107.6 104.4  0.5836 16.4  67.9        R64   15
## 197         Long Beach St.   BW 105.4 108.0  0.4313 17.1  71.4        R64   15
## 198          Saint Peter's MAAC  96.8 100.2  0.4038 19.6  64.0        R64   15
## 199               Longwood BSth 104.0 104.8  0.4768 18.0  66.8        R64   16
## 200            Montana St. BSky 102.9 105.8  0.4193 16.8  68.5        R68   16
## 201                Stetson ASun 108.3 114.0  0.3559 15.8  66.0        R64   16
## 202                 Howard MEAC 104.5 112.9  0.2913 21.4  67.7        R68   16
## 203          Grambling St. SWAC  97.7 106.0  0.2805 20.5  64.9        R64   16
## 204                 Wagner  NEC  95.0 105.0  0.2399 15.9  61.3        R64   16
##     YEAR       WIN avgBARTHAG
## 1   2022  87.50000  0.9092625
## 2   2022  85.00000  0.9580000
## 3   2022  79.41176  0.8632188
## 4   2022  89.18919  0.9092625
## 5   2022  82.05128  0.9393000
## 6   2022  78.94737  0.9393000
## 7   2022  76.47059  0.7245094
## 8   2022  82.35294  0.8632188
## 9   2022  72.97297  0.9092625
## 10  2022  77.14286  0.8632188
## 11  2022  78.37838  0.9092625
## 12  2022  75.75758  0.8632188
## 13  2022  77.14286  0.9092625
## 14  2022  69.69697  0.8632188
## 15  2022  75.67568  0.8435500
## 16  2022  81.81818  0.9092625
## 17  2022  84.21053  0.8435500
## 18  2022  72.22222  0.7245094
## 19  2022  78.78788  0.8632188
## 20  2022  69.69697  0.7245094
## 21  2022  64.70588  0.8632188
## 22  2022  64.70588  0.7245094
## 23  2022  57.57576  0.7245094
## 24  2022  83.33333  0.7245094
## 25  2022  62.50000  0.8632188
## 26  2022  63.88889  0.8632188
## 27  2022 100.00000  0.8632188
## 28  2022  76.47059  0.7245094
## 29  2022  74.35897  0.9113000
## 30  2022  71.87500  0.7245094
## 31  2022  79.41176  0.7245094
## 32  2022  67.74194  0.7245094
## 33  2022  66.66667  0.8632188
## 34  2022  61.76471  0.8632188
## 35  2022  65.71429  0.8632188
## 36  2022  59.37500  0.7245094
## 37  2022  72.72727  0.7245094
## 38  2022  70.27027  0.8435500
## 39  2022  78.12500  0.7245094
## 40  2022  81.81818  0.7245094
## 41  2022  63.88889  0.7245094
## 42  2022  55.88235  0.9092625
## 43  2022  70.58824  0.8632188
## 44  2022  62.85714  0.9092625
## 45  2022  56.25000  0.5791000
## 46  2022  60.00000  0.7245094
## 47  2022  81.81818  0.7245094
## 48  2022  75.75758  0.5791000
## 49  2022  87.09677  0.8632188
## 50  2022  64.86486  0.8632188
## 51  2022  84.84848  0.7245094
## 52  2022  81.81818  0.7245094
## 53  2022  90.90909  0.7245094
## 54  2022  75.00000  0.7245094
## 55  2022  67.64706  0.7245094
## 56  2022  89.65517  0.7245094
## 57  2022  63.33333  0.7245094
## 58  2022  81.81818  0.7245094
## 59  2022  66.66667  0.8435500
## 60  2022  64.70588  0.7245094
## 61  2022  70.00000  0.7245094
## 62  2022  70.00000  0.7245094
## 63  2022  69.23077  0.7245094
## 64  2022  70.96774  0.5791000
## 65  2022  62.85714  0.7245094
## 66  2022  59.37500  0.7245094
## 67  2022  85.71429  0.7245094
## 68  2022  74.19355  0.5791000
## 69  2023  77.77778  0.8243750
## 70  2023  82.85714  0.7543719
## 71  2023  83.78378  0.8955875
## 72  2023  89.18919  0.8955875
## 73  2023  76.31579  0.9150500
## 74  2023  80.55556  0.8243750
## 75  2023  80.00000  0.7543719
## 76  2023  83.78378  0.8955875
## 77  2023  72.22222  0.9150500
## 78  2023  86.11111  0.9150500
## 79  2023  67.64706  0.8243750
## 80  2023  72.97297  0.8955875
## 81  2023  79.48718  0.9643000
## 82  2023  65.71429  0.8243750
## 83  2023  75.75758  0.7543719
## 84  2023  69.44444  0.8955875
## 85  2023  84.21053  0.9205000
## 86  2023  78.37838  0.8765000
## 87  2023  79.41176  0.8243750
## 88  2023  75.00000  0.8243750
## 89  2023  64.86486  0.9150500
## 90  2023  64.70588  0.8243750
## 91  2023  62.85714  0.8243750
## 92  2023  57.57576  0.7543719
## 93  2023  64.70588  0.8243750
## 94  2023  71.42857  0.8243750
## 95  2023  71.42857  0.7543719
## 96  2023  61.76471  0.8955875
## 97  2023  62.85714  0.8243750
## 98  2023  57.57576  0.7543719
## 99  2023  74.28571  0.7543719
## 100 2023  61.11111  0.8955875
## 101 2023  94.59459  0.8765000
## 102 2023  61.76471  0.8243750
## 103 2023  55.88235  0.7543719
## 104 2023  60.60606  0.7543719
## 105 2023  62.16216  0.8243750
## 106 2023  66.66667  0.7543719
## 107 2023  72.72727  0.7543719
## 108 2023  76.47059  0.7543719
## 109 2023  66.66667  0.8243750
## 110 2023  63.63636  0.7543719
## 111 2023  67.64706  0.7543719
## 112 2023  63.88889  0.7543719
## 113 2023  61.76471  0.5186000
## 114 2023  68.75000  0.5186000
## 115 2023  79.41176  0.7543719
## 116 2023  77.14286  0.7543719
## 117 2023  96.77419  0.7543719
## 118 2023  91.17647  0.7543719
## 119 2023  84.84848  0.8243750
## 120 2023  83.87097  0.7543719
## 121 2023  77.14286  0.7543719
## 122 2023  84.84848  0.7543719
## 123 2023  72.72727  0.7543719
## 124 2023  75.75758  0.7543719
## 125 2023  81.25000  0.7543719
## 126 2023  81.81818  0.7543719
## 127 2023  69.69697  0.7543719
## 128 2023  76.47059  0.7543719
## 129 2023  81.81818  0.7543719
## 130 2023  76.66667  0.8955875
## 131 2023  60.00000  0.8243750
## 132 2023  66.66667  0.7543719
## 133 2023  68.75000  0.7543719
## 134 2023  77.41935  0.7543719
## 135 2023  42.42424  0.5186000
## 136 2023  55.88235  0.5186000
## 137 2024  88.23529  0.9344500
## 138 2024  91.17647  0.9712000
## 139 2024  87.87879  0.9644000
## 140 2024  79.41176  0.9344500
## 141 2024  79.41176  0.9344500
## 142 2024  75.75758  0.9344500
## 143 2024  75.00000  0.9130500
## 144 2024  73.52941  0.9344500
## 145 2024  71.87500  0.9344500
## 146 2024  76.47059  0.9130500
## 147 2024  71.87500  0.8256000
## 148 2024  71.87500  0.7082844
## 149 2024  79.41176  0.7082844
## 150 2024  75.00000  0.9130500
## 151 2024  65.62500  0.8553500
## 152 2024  70.96774  0.8256000
## 153 2024  81.25000  0.7082844
## 154 2024  80.64516  0.9344500
## 155 2024  62.85714  0.7082844
## 156 2024  75.00000  0.9344500
## 157 2024  69.69697  0.7082844
## 158 2024  69.69697  0.7082844
## 159 2024  65.62500  0.9130500
## 160 2024  78.78788  0.7082844
## 161 2024  62.50000  0.8256000
## 162 2024  68.57143  0.7082844
## 163 2024  72.72727  0.8256000
## 164 2024  77.41935  0.8256000
## 165 2024  69.69697  0.7082844
## 166 2024  61.76471  0.7082844
## 167 2024  75.75758  0.7082844
## 168 2024  87.09677  0.8256000
## 169 2024  57.57576  0.8256000
## 170 2024  65.62500  0.8256000
## 171 2024  63.63636  0.7082844
## 172 2024  58.82353  0.8256000
## 173 2024  70.58824  0.8256000
## 174 2024  81.25000  0.7082844
## 175 2024  73.33333  0.5947750
## 176 2024  75.00000  0.7082844
## 177 2024  84.84848  0.7082844
## 178 2024  69.69697  0.5947750
## 179 2024  76.47059  0.7082844
## 180 2024  67.64706  0.8256000
## 181 2024  61.11111  0.8553500
## 182 2024  70.58824  0.8256000
## 183 2024  93.93939  0.8256000
## 184 2024  90.62500  0.8256000
## 185 2024 103.44828  0.7082844
## 186 2024  69.69697  0.7082844
## 187 2024  75.86207  0.8256000
## 188 2024  90.62500  0.7082844
## 189 2024  81.81818  0.7082844
## 190 2024  87.50000  0.7082844
## 191 2024  75.00000  0.7082844
## 192 2024  86.66667  0.7082844
## 193 2024  67.64706  0.8256000
## 194 2024  75.75758  0.7082844
## 195 2024  73.33333  0.7082844
## 196 2024  70.96774  0.7082844
## 197 2024  63.63636  0.7082844
## 198 2024  61.29032  0.7082844
## 199 2024  67.74194  0.7082844
## 200 2024  54.83871  0.5947750
## 201 2024  70.96774  0.7082844
## 202 2024  56.25000  0.5947750
## 203 2024  64.51613  0.7082844
## 204 2024  55.17241  0.7082844

Data Filtering 3- Data Frame

## # A tibble: 12 × 12
##    TEAM      CONF  ADJOE ADJDE BARTHAG   TOR ADJ_T POSTSEASON  SEED  YEAR   WIN avgBARTHAG
##    <chr>     <chr> <dbl> <dbl>   <dbl> <dbl> <dbl> <chr>      <dbl> <dbl> <dbl>      <dbl>
##  1 Gonzaga   WCC   120.   89.9   0.966  15.7  72.6 S16            1  2022  87.5      0.909
##  2 Kansas    B12   120.   91.3   0.958  17.3  69.1 Champions      1  2022  85        0.958
##  3 Baylor    B12   118.   91.3   0.949  17.9  67.6 R32            1  2022  79.4      0.863
##  4 Arizona   P12   118    93.7   0.934  18.3  72.3 S16            1  2022  89.2      0.909
##  5 Duke      ACC   123.   96     0.944  15    67.4 F4             2  2022  82.1      0.939
##  6 Villanova BE    117.   93     0.935  15.5  62.2 F4             2  2022  78.9      0.939
##  7 Longwood  BSth  104   105.    0.477  18    66.8 R64           16  2024  67.7      0.708
##  8 Montana … BSky  103.  106.    0.419  16.8  68.5 R68           16  2024  54.8      0.595
##  9 Stetson   ASun  108.  114     0.356  15.8  66   R64           16  2024  71.0      0.708
## 10 Howard    MEAC  104.  113.    0.291  21.4  67.7 R68           16  2024  56.2      0.595
## 11 Gramblin… SWAC   97.7 106     0.280  20.5  64.9 R64           16  2024  64.5      0.708
## 12 Wagner    NEC    95   105     0.240  15.9  61.3 R64           16  2024  55.2      0.708

Plot 1- Percentage of Top 8 Seeds per Conference per Year

Plot 1- Analysis

  • The plots allow for analysis of which conference produced the lowest seeded, therefore better, teams over the last three years
  • Below is a breakdown of total teams seeded in the top 8 over the last three years:
##       [,1]  [,2]  [,3]  [,4] [,5]  [,6]  [,7]  [,8]  [,9]   [,10] [,11]
## CONF  "SEC" "B12" "B10" "BE" "ACC" "P12" "MWC" "WCC" "Amer" "A10" "OVC"
## total "19"  "17"  "16"  "11" " 8"  " 7"  " 6"  " 6"  " 4"   " 1"  " 1"
  • The top three conferences in terms of top 8 seeds the last three years were the SEC, Big-12 (B12), and Big-10 (B10) with the following percentages:
##            [,1]       [,2]       [,3]      
## CONF       "SEC"      "B12"      "B10"     
## Percentage "19.79167" "17.70833" "16.66667"

Plot 2- Average Power Rating by Postseason Finish by Year

Plot 2- Analysis

  • The plot shows the average power rating of the teams to finish at each round of the last three NCAA tournaments
  • A logical expectation of the data is for the teams that make the final 2 to have the highest power rating and teams that are eliminated in either the rounds of 32 and 64 or the first four to have lower power ratings (*)
  • The average power rating of teams over all three years at each tournament finishing position are presented below:
##            [,1]        [,2]        [,3]        [,4]        [,5]        [,6]       
## POSTSEASON "Champions" "2ND"       "F4"        "E8"        "S16"       "R32"      
## total      "0.9645000" "0.9320667" "0.8903833" "0.8905500" "0.9131000" "0.8377313"
##            [,7]        [,8]       
## POSTSEASON "R64"       "R68"      
## total      "0.7290552" "0.5641583"
  • This data supports the claim in (*), but an interesting trend is a higher average power rating for teams eliminated in the sweet sixteen compared to teams eliminated in the elite eight or final four
    • A potential explanation is there are 8 teams eliminated in the sweet sixteen, which provides a larger selection size of teams eliminated increasing the chance of an outlier with a higher power rating but earlier tournament elimination

Plot 3- Range of Turnover Rate by Seed by Year

Plot 3- Analysis

  • The intent of this plot was to investigate if teams seeded lower (1-4), hence the better teams, had a lower turnover rate compared to the higher seeded (13-16) teams
  • The plot reveals no substantial relation between turnover rate and team seed although it does appear overall turnover rate between all seeds was lowest in 2024

Plot 4- Win Percentage vs Adjusted Offensive and Defensive Efficiency

Plot 4- Analysis 1

  • The plot explores the relation between win percentage and both adjusted offensive efficiency and adjusted defensive efficiency
  • Linear regression can be used with both variables to analyze their relation to win percentage:
  • Model: y = a1x1 + a2x2 + b
## 
## Call:
## lm(formula = WIN ~ ADJOE + ADJDE, data = CBB_P4)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -23.5710  -6.5030  -0.2949   5.7684  30.8818 
## 
## Coefficients:
##             Estimate Std. Error t value Pr(>|t|)    
## (Intercept)  48.2148    19.8667   2.427 0.016110 *  
## ADJOE         0.3678     0.1092   3.368 0.000909 ***
## ADJDE        -0.1642     0.1318  -1.246 0.214066    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 9.557 on 201 degrees of freedom
## Multiple R-squared:  0.07407,    Adjusted R-squared:  0.06486 
## F-statistic: 8.039 on 2 and 201 DF,  p-value: 0.0004377

Plot 4- Analysis 2

  • The summary on the previous slides leads to the following statements and conclusions:
  • ADJOE:
    • The slope for adjusted offensive efficiency implies a 0.3678 increase in win percentage per one point increase in ADJOE
    • The 0.000909 p-value means ADJOE is statistically significant in predicting win percentage
  • ADJDE
    • The slope for adjusted defensive efficiency implies a 0.1642 increase in win percentage per one point decrease in ADJDE
    • The 0.214066 p-value means ADJDE is NOT statistically significant in predicting win percentage
  • The small adjusted R-squared value (0.06486) implies there are many more factors that influence win percentage
  • The statistics show that adjusted offensive efficiency is a better predictor of win percentage than adjusted defensive efficiency.

Plot 5- Offensive Efficiency vs Adjusted Tempo

Plot 5- Analysis 1

  • This last plot was a test to see if the tempo a team plays at influences their offensive efficiency
  • Test statistical significance with linear regression model: y = ax + b
## 
## Call:
## lm(formula = ADJOE ~ ADJ_T, data = CBB_P5)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -17.2230  -4.0078   0.5336   4.5026  16.4848 
## 
## Coefficients:
##             Estimate Std. Error t value Pr(>|t|)    
## (Intercept)  82.4192    11.3951   7.233  9.6e-12 ***
## ADJ_T         0.4365     0.1694   2.576   0.0107 *  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 6.264 on 202 degrees of freedom
## Multiple R-squared:  0.0318, Adjusted R-squared:  0.02701 
## F-statistic: 6.635 on 1 and 202 DF,  p-value: 0.01071

Plot 5- Analysis 2

  • The slope for adjusted tempo implies a 0.4365 point increase in adjusted offensive efficiency per one point increase in adjusted tempo
  • The p-value of 0.0107 implies there is statistical significance
  • These statistics lead to the conclusion that teams that play at a higher tempo are likely to have a higher adjusted offensive efficiency
    • However, the small R–squared value implies other factors also influence ADJOE

Summary

Summary of Plots
Plot Number Method Type Variables Statistical Analysis
1 plot_ly Pie Seed, Year, Conference Proportions
2 ggplot Bar Postseason, Year, BARTHAG Mean
3 plot_ly Box Seed, Year, TOR N/A
4 plot_ly 3D Scatter ADJOE, ADJDE, Year, Win Percentage, Postseason Linear Regression
5 ggplot Scatter ADJ_T, ADJOE, Postseason Linear Regression