College Football Rankings of
Earl R. Jessen, season of 1954
1 Ohio State 10-0-0 2 UCLA 9-0-0 3 Oklahoma 10-0-0 4 Notre Dame 9-1-0 5 Navy 8-2-0 6 Wisconsin 7-2-0 7 Duke 8-2-1 8 Miami FL 8-1-0 9 Maryland 7-2-1 10 Georgia Tech 8-3-0 11 Mississippi 9-2-0 12 West Virginia 8-1-0 13 Minnesota 7-2-0 14 Auburn 8-3-0 15 USC 8-4-0 16 Army 7-2-0 17 Arkansas 8-3-0 18 SMU 6-3-1 19 Purdue 5-3-1 20 Rice 7-3-0 21 Penn State 7-2-0 22 Michigan 6-3-0 23 Georgia 6-3-1 24 Kentucky 7-3-0 25 Colorado 7-2-1 26 Virginia Tech 8-0-1 27 Boston College 8-1-0 28 Denver 9-1-0 29 Baylor 7-4-0 30 Wichita 9-1-0
Listed above in this column you see the top 30 teams of 1954 according to football historian Earl R. Jessen, who put them in that order at the conclusion of that season. In other words, these are not retroratings, but were made at the conclusion of the 1954 season. The following column is almost the same list, except that the won-lost records have been replaced with rating numbers.
In the 1960s and 1970s I used to receive the college football year in review from both Earl R. Jessen and Richard Poling. Both rated the teams every year as well as reviewing the just completed season. For the example on this page I used 1954 as the sample season and used the year in review by Mr. Jessen in order to try to make this devil's advocate point.
Let us divide into four sections the way college football teams can be put in order.
1) A person can list the teams in order of subjective judgment, such as an individual voter in a weekly or final AP poll.
2) Teams can be rated by a math formula rating system which is usually the product of one individual - the inventor of the system. The annual book of NCAA Football lists several.
3) A poll can be taken of a collection of those who use step one. The AP poll does this.
4) A poll can be taken of a collection of those who use step two. (This is an enjoyable January pastime for some of us, and is now used in the BCS formula.)
Looking at the top 30 in the above column, you can't tell if this list of teams is based on step 1 (subjective judgment) or step 2 (a math formula rating system). But looking at the same top 30 in the following column, you see those numbers to the right of the teams and conclude that Mr. Jessen used a math formula rating system.
Does the presence of those rating numbers next to the teams in the following column (88.02, 87.98, etc.) cause you to think differently about the rankings of Mr. Jessen, compared to the same top 30 list with only won-lost records? Is a top 30 with a rating system superior to a top 30 without math, but put in order subjectively?
Okay, here's the joker in the deck, and the point to all this, and you could see it coming. Mr. Jessen never in his life used a math formula rating system, and those numbers to the right of the teams in the following column were made up with the help of a list of random numbers. Does this mean his ability to put teams in order is somehow inferior to those who use math ratings to rank teams? Do you somehow mentally respect more a list of teams if numbers are shown to the right of the teams? If Mr. Jessen had simply made up numbers every year (as I did in this example) to make people think he had a rating system, would we respect his ratings more?
College Football Ratings of
Earl R. Jessen, season of 1954
1 Ohio State 88.02 2 UCLA 87.98 3 Oklahoma 87.43 4 Notre Dame 86.91 5 Navy 86.25 6 Wisconsin 85.44 7 Duke 85.19 8 Miami FL 84.59 9 Maryland 83.92 10 Georgia Tech 83.87 11 Mississippi 83.46 12 West Virginia 82.99 13 Minnesota 82.66 14 Auburn 82.16 15 USC 81.34 16 Army 81.06 17 Arkansas 80.21 18 SMU 79.92 19 Purdue 79.73 20 Rice 79.50 21 Penn State 78.85 22 Michigan 78.70 23 Georgia 78.07 24 Kentucky 77.34 25 Colorado 76.48 26 Virginia Tech 75.59 27 Boston College 75.52 28 Denver 74.79 29 Baylor 74.46 30 Wichita 73.54
Would these ratings now be included in the annual book of NCAA Football if he had assigned numbers after each team? And how would anyone know if the numbers were just made up or the real thing?