NFL Statistical Study Question

[Buffalo Bob]

I’d like to have my approach to a study I performed critiqued from the standpoint of a math expert.

My question was how my favorite NFL team was “budgeting” or “allocating” their first and second round draft selections over the past 7 seasons versus the same budgeting by the teams who have qualified for the Super Bowl over the past 5 seasons, all from a standpoint of the players positions.

I broke down the positions into 8 categories (Defensive Line, Wide Receiver, etc.).

Since I had exact data available from the results of those drafts, I needed to quantify the positions by where they were taken in the rounds in order to gain a “budgeted” quantification. Since there are 32 picks in round one, and 32 picks in round two plus any compensatory 2nd round picks added at the league’s discretion, I used the formula of giving pick #1 in the first round a score of 64, pick #2 a score of 63, etc. If the first pick in the first round was a Quarterback, that teams “budget” was credited with 64 points in the Quarterback position. There are only limited compensatory picks, so I simply chose to count those as 1 point each, just like the 64th pick in the draft. I realize this slightly diminishes the overall accuracy, but it was to such a minor degree that I proceeded.

After accruing the quantification at each position for every team in the study, I added the total quantification for each position of the Super Bowl teams together and came up with percentages of the average that those teams allocate to the different positions in the study. I then simply did the same quantification for the team I follow and matched the results, giving me what appears to be notable differences in the positions that my team uses their “budget” on versus the averages of the Super Bowl Teams.

I’ll include the spreadsheet here for you to look at. I want to get a feeling for whether there are obvious flaws that diminish the results of the study, or whether the approach seems reasonable and sound. Any help is greatly appreciated.


Giants Colts Steelers Patriots* Bears Seahawks Eagles Panthers
Cum Score %
DL 82 0.21 DL 77 0.24 DL 46 0.11 DL 157 0.35 DL 123 0.26 DL 49 0.13 DL 143 0.37 DL 84 0.19 761 0.23 DL
DB 110 0.28 DB 76 0.24 DB 76 0.19 DB 70 0.16 DB 61 0.13 DB 123 0.32 DB 52 0.14 DB 95 0.21 663 0.20 DB
WR 54 0.14 WR 68 0.21 WR 43 0.11 WR 50 0.11 WR 83 0.18 WR 56 0.15 WR 70 0.18 WR 23 0.05 447 0.14 WR
TE 51 0.13 TE 41 0.13 TE 35 0.09 TE 77 0.17 TE 34 0.07 TE 37 0.10 TE 4 0.01 TE 0.00 279 0.086 TE
LB 0.00 LB 0.00 LB 114 0.28 LB 0.00 LB 0.00 LB 20 0.05 LB 12 0.03 LB 94 0.21 240 0.07 LB
OL 31 0.08 OL 23 0.07 OL 35 0.09 OL 50 0.11 OL 36 0.08 OL 87 0.23 OL 75 0.19 OL 72 0.16 409 0.13 OL
RB 0.00 RB 35 0.11 RB 0.00 RB 44 0.10 RB 88 0.19 RB 11 0.03 RB 0.00 RB 80 0.18 258 0.08 RB
QB 61 0.16 QB 0.00 QB 54 0.13 QB 0.00 QB 43 0.09 QB 0.00 QB 29 0.08 QB 0.00 187 0.06 QB

Total 389 320 403 448 468 383 385 448 3244
Offense 0.487

Bills Lions Cardinals Texans

DL 79 0.156 DL 65 0.11 DL 174 0.29 DL #DIV/0! DL #DIV/0! DL 168 0.35 DL #DIV/0! 486 0.22 DL
DB 101 0.20 DB 29 0.05 DB 57 0.10 DB #DIV/0! DB #DIV/0! DB 55 0.12 DB #DIV/0! 242 0.11 DB
WR 91 0.18 WR 240 0.40 WR 121 0.20 WR #DIV/0! WR #DIV/0! WR 94 0.20 WR #DIV/0! 546 0.25 WR
TE 0.00 TE 0.00 TE 0.00 TE #DIV/0! TE #DIV/0! TE 24 0.05 TE #DIV/0! 24 0.01 TE
LB 31 0.06 LB 115 0.19 LB 16 0.03 LB #DIV/0! LB #DIV/0! LB 70 0.15 LB #DIV/0! 232 0.11 LB
OL 61 0.12 OL 62 0.10 OL 147 0.25 OL #DIV/0! OL #DIV/0! OL 0.00 OL #DIV/0! 270 0.12 OL
RB 102 0.20 RB 0.00 RB 21 0.04 RB #DIV/0! RB #DIV/0! RB 0.00 RB #DIV/0! 123 0.06 RB
QB 43 0.08 QB 84 0.14 QB 55 0.09 QB #DIV/0! QB #DIV/0! QB 64 0.13 QB #DIV/0! 246 0.11 QB

508 595 591 0 0 475 0 2169
Offense 0.56
Offense 0.585 0.65 0.58
0.33

Bills Succe 48 O

RB 20 8
DB 20 21
WR 18 14
DL 16 24
OL 12 13
QB 8 6
LB 6 7
TE 0 8

 

[stapel]

Unfortunately, your copy-n-paste did not translate well, so your "table" is somewhat difficult to read...? Since we have no mathematical definition of "budgeting" or precise instructions regarding how to measure this, it is difficult to know whether you have evaluated the data correctly. Since apparently many of your processes resulted in division by zero, and since division by zero is not allowed, this would suggest that there may be something wrong with some aspect of some formula or process, or with some portion of the data. (And not being familiar with the industry-specific needs or practices, I can't say anything regarding the set-up, such as whether the percentages measure anything "meaningful".)

Please consult with your instructor regarding more-specific definitions and instructions. Or, if this is not school-related (so not tutoring-related), you might want to consult with an expert in the field, who can advise you regarding known-useful measures, formulas, and processes, and / or who could usefully critique what you've developed.

Thank you!

Eliz.