Another Comparison of RBs’ Best 4 Years

Last week, I posted this article comparing some of the top RBs in NFL history, this time looking at their career peaks in terms of a 4 year stretch. As quick review, I defined their top 4 year stretches as the period of 4 consecutive years where the player gained the most total yards from scrimmage in his career. In my last post, many commenters (both here at Checking the Numbers and at BTSC) suggested improvements both in analysis technique and in players to review, so I decided to go back and complete a followup analysis. Based on the feedback that I received, I added several players that were suggested, and I also made a couple changes to my ranking system.

There were suggestions to normalize the players’ numbers against the league average for rushing (and possibly receiving) or to normalize against the top 10 (top 5 pre-SB era) rushers in the league. After considering the direction to take, my first thought was to normalize against the top rushers in the league. However, to do that would have required pulling the numbers for top 10 in rushing yards, yards/attempt and TDs for something like 50 or 60 total years of data. Then, I would have had to average those players’ for each year, and then average the years for each 4 year period. That would have taken forever.

So, I decided to normalize against league averages instead. I realize that in this case, I would be normalizing players’ data against teams’ data. But, I think it will still be useful, because the trends in average team data gives an indication of the average rushing defensive strength in the league for that period. It also indicates trends in relying on the run or the pass. For example, in compiling this data, I saw that the league averages for rushing increased from the 1930s to the 1960s and 1970s, but declined in the 1980s and beyond. Most likely, this is due to improved schemes into the ’60s and then the rules changes favoring the passing game since ’78. So, for these reasons, and the fact that it would take considerably less time, I went with league averages, rather than top 10 or top 5 players’ averages.

The final change that I made to my analysis was inspired by BTSC member Steely McSmash. Although I didn’t follow his formula, I used his idea of normalizing to the leader in each category to assign a value to the players’ stats based on what percentage of the highest value they had. He assigned a weighting percentage, while I did not. However, since I added categories for how each player performed relative to league average in rushing, my analysis now adds more weight to the rushing statistics than it did previously (doubling it).

From commenters suggestions, I added the following players to the analysis: Shaun Alexander, Jim Brown (’58-’61), Earl Campbell, Roger Craig, Corey Dillon, Eddie George, Cookie Gilchrist, Priest Holmes, Leroy Kelly, Bronko Nagurski, John Riggins, Gale Sayers, Jim Taylor, Thurman Thomas, Herschel Walker, Ricky Watters, Brian Westbrook, OJ Simpson, and Tiki Barber. Barry Foster and Terry Allen were also suggested, but Barry only had a single 1000 yard season, and I left Terry out of the compilation. When I reviewed Terry Allen’s number, his best 4 year period was just average to below average for this list, so I did not think it worth the extra time to add one more player.

On to the analysis! Below, I have a total of 9 tables. Tables 1-3 contain the players’ raw statistics, (taken from Pro-Football-Reference.c0m). Tables 4 and 5 contain per game and per touch statistics. Tables 6 and 7 contain values for each players’ statistics normalized to the leaders in each category. And, Tables 8 and 9 contain the players ranked according to two different ranking systems that I completed.

First, we have the players’ total numbers.

Players Years G GS YScm TotTD Fmb
Bronko Nagurski 1932-1935 45 38 1944 13 0
Spec Sanders 1946-1948* 40 30 3172 36 1
Marion Motley 1947-1950 51 36 3840 28 5
Jim Brown 1 1958-1961 50 50 6512 53 22
Jim Taylor 1961-1964 54 0 5671 60 18
Jim Brown 2 1962-1965 56 56 7302 63 28
Cookie Gilchrist 1962-1965 56 0 5039 42 29
Gale Sayers 1965-1968 50 0 5031 40 26
Leroy Kelly 1966-1969 55 0 5614 59 15
OJ Simpson 1973-1976 56 7 7392 48 27
Franco Harris 1976-1979 59 59 5206 45 33
Walter Payton 1 1977-1980 62 62 7746 50 28
Earl Campbell 1978-1981 62 61 6802 55 31
Tony Dorsett 1978-1981 61 60 6604 33 39
John Riggins 1979-1983* 54 42 4068 52 13
Walter Payton 2 1983-1986 64 64 7829 41 22
Marcus Allen 1983-1986 61 57 7056 50 32
Eric Dickerson 1983-1986 62 62 7842 57 49
Roger Craig 1985-1988 62 60 6863 36 22
Herschel Walker 1986-1989 60 51 6537 38 22
Bo Jackson 1987-1990 38 23 3134 18 11
Thurman Thomas 1989-1992 63 63 7893 49 24
Emmitt Smith 1992-1995 61 60 7921 76 16
Barry Sanders 1994-1997 64 64 8122 45 10
Terrell Davis 1995-1998 61 61 7594 61 16
Ricky Watters 1995-1998 64 64 6724 41 18
Jerome Bettis 1996-1999 62 58 5804 30 17
Eddie George 1997-2000 64 64 6771 42 21
Marshall Faulk 1998-2001 60 59 8992 69 8
Curtis Martin 1998-2001 63 63 6920 35 11
Corey Dillon 1999-2002 63 63 6235 33 17
Ricky Williams 2000-2003 58 58 7104 43 28
Jamal Lewis 2000-2004* 60 57 6822 34 24
Priest Holmes 2001-2004 54 54 7645 76 10
Clinton Portis 2002-2005 60 56 7059 49 16
Shaun Alexander 2002-2005 64 63 7189 82 17
Edgerrin James 2003-2006 60 60 6801 40 16
LaDainian Tomlinson 2003-2006 63 63 8301 86 13
Tiki Barber 2003-2006 64 62 8290 34 18
Brian Westbrook 2004-2007 55 53 6768 39 5

Table 1: years, games, games started, yards from scrimmage, total touchdowns, and fumbles

*Spec Sanders missed 1949 and returned 1950 as a Safety, John Riggins missed 1980 due to a contract dispute, and Jamal Lewis missed 2001 due to injury

Next are the players’ Rushing stats.


Rushing
Players Att Yds TD Lng Y/A Y/G A/G
Bronko Nagurski 422 1822 13 0 4.3 40.5 9.4
Spec Sanders 540 2900 33 0 5.4 72.5 13.5
Marion Motley 556 3233 24 69 5.8 63.4 10.9
Jim Brown 1 1067 5521 48 71 5.2 110.4 21.3
Jim Taylor 998 4968 55 84 5.0 92.0 18.5
Jim Brown 1090 5849 49 80 5.4 104.4 19.5
Cookie Gilchrist 928 4010 37 67 4.3 71.6 16.6
Gale Sayers 719 3834 31 70 5.3 76.7 14.4
Leroy Kelly 888 4402 51 70 5.0 80.0 16.1
OJ Simpson 1221 6448 39 88 5.3 115.1 21.8
Franco Harris 1166 4558 44 71 3.9 77.3 19.8
Walter Payton 1 1358 6317 45 76 4.7 101.9 21.9
Earl Campbell 1404 6457 55 81 4.6 104.1 22.6
Tony Dorsett 1160 5263 28 75 4.5 86.3 19.0
John Riggins 1007 3767 49 66 3.7 69.8 18.6
Walter Payton 2 1340 5989 34 72 4.5 93.6 20.9
Marcus Allen 1129 4700 38 61 4.2 77.0 18.5
Eric Dickerson 1465 6968 55 85 4.8 112.4 23.6
Roger Craig 943 4197 28 62 4.5 67.7 15.2
Herschel Walker 971 4057 31 84 4.2 67.6 16.2
Bo Jackson 515 2782 16 92 5.4 73.2 13.6
Thurman Thomas 1169 5435 33 80 4.6 86.3 18.6
Emmitt Smith 1401 6456 73 68 4.6 105.8 23.0
Barry Sanders 1287 6989 40 85 5.4 109.2 20.1
Terrell Davis 1343 6413 56 71 4.8 105.1 22.0
Ricky Watters 1294 5033 40 57 3.9 78.6 20.2
Jerome Bettis 1310 5372 28 50 4.1 86.6 21.1
Eddie George 1428 5506 34 40 3.9 86.0 22.3
Marshall Faulk 1090 5441 43 71 5.0 90.7 18.2
Curtis Martin 1385 5468 32 60 3.9 86.8 22.0
Corey Dillon 1232 5261 29 96 4.3 83.5 19.6
Ricky Williams 1336 5470 39 63 4.1 94.3 23.0
Jamal Lewis 1239 5763 33 82 4.7 96.1 20.7
Priest Holmes 1156 5482 70 56 4.7 101.5 21.4
Clinton Portis 1258 5930 45 65 4.7 98.8 21.0
Shaun Alexander 1344 6186 73 88 4.6 96.7 21.0
Edgerrin James 1341 5472 39 43 4.1 91.2 22.4
LaDainian Tomlinson 1339 6257 76 85 4.7 99.3 21.3
Tiki Barber 1284 6256 29 95 4.9 97.8 20.1
Brian Westbrook 851 3979 20 71 4.7 72.3 15.5

Table 2: rushing attempts, yards, TDs, long rush, yards/attempt, yards/game, and attempts/game

Then, the players’ Receiving stats.


Receiving
Players Rec Yds Y/R TD Lng R/G Y/G
Bronko Nagurski 10 122 12.2 0 23 0.2 2.7
Spec Sanders 18 272 15.1 3 13 0.3 6.8
Marion Motley 46 607 13.2 4 78 0.9 11.9
Jim Brown 1 105 991 9.4 5 77 2.1 19.8
Jim Taylor 98 703 7.2 5 35 1.8 13.0
Jim Brown 141 1453 10.3 14 83 2.5 25.9
Cookie Gilchrist 96 1029 10.7 5 76 1.7 18.4
Gale Sayers 94 1197 12.7 9 80 1.9 23.9
Leroy Kelly 94 1212 12.9 8 68 1.7 22.0
OJ Simpson 71 944 13.3 9 64 1.3 16.9
Franco Harris 92 648 7.0 1 39 1.6 11.0
Walter Payton 1 154 1429 9.3 5 75 2.5 23.0
Earl Campbell 75 345 4.6 0 46 1.2 5.6
Tony Dorsett 148 1341 9.1 5 91 2.4 22.0
John Riggins 49 301 6.1 3 23 0.9 5.6
Walter Payton 2 184 1840 10.0 7 74 2.9 28.8
Marcus Allen 245 2356 9.6 12 92 4.0 38.6
Eric Dickerson 118 874 7.4 2 37 1.9 14.1
Roger Craig 315 2666 8.5 8 73 5.1 43.0
Herschel Walker 229 2480 10.8 7 84 3.8 41.3
Bo Jackson 40 352 8.8 2 27 1.1 9.3
Thurman Thomas 229 2458 10.7 16 74 3.6 39.0
Emmitt Smith 228 1465 6.4 3 86 3.7 24.0
Barry Sanders 149 1133 7.6 5 66 2.3 17.7
Terrell Davis 152 1181 7.8 5 35 2.5 19.4
Ricky Watters 213 1691 7.9 1 37 3.3 26.4
Jerome Bettis 74 432 5.8 2 26 1.2 7.0
Eddie George 141 1265 9.0 8 54 2.2 19.8
Marshall Faulk 337 3551 10.5 26 78 5.6 59.2
Curtis Martin 211 1452 6.9 3 34 3.3 23.0
Corey Dillon 126 974 7.7 4 31 2.0 15.5
Ricky Williams 201 1634 8.1 4 59 3.5 28.2
Jamal Lewis 110 1059 9.6 1 77 1.8 17.7
Priest Holmes 225 2163 9.6 6 67 4.2 40.1
Clinton Portis 141 1129 8.0 4 72 2.4 18.8
Shaun Alexander 139 1003 7.2 9 80 2.2 15.7
Edgerrin James 184 1329 7.2 1 56 3.1 22.2
LaDainian Tomlinson 260 2044 7.9 10 74 4.1 32.4
Tiki Barber 233 2034 8.7 5 62 3.6 31.8
Brian Westbrook 301 2789 9.3 19 62 5.5 50.7

Table 3: receptions, receiving yards, yards/reception, TDs, long reception, receptions/game, and yards/game

Below are the per game and per touch rushing stats.

  Rushing Rsh Nrm-League Avg
Players Y/A Y/G TD/G Y/A Y/G TD/G
Bronko Nagurski 4.3 40.5 0.29 1.29 0.33 0.44
Spec Sanders 5.4 72.5 0.83 1.32 0.5 0.66
Marion Motley 5.8 63.4 0.47 1.34 0.41 0.34
Jim Brown 1 5.2 110.4 0.96 1.24 0.8 0.87
Jim Taylor 5.0 92.0 1.02 1.21 0.72 1
Jim Brown 2 5.4 104.4 0.88 1.33 0.84 0.86
Cookie Gilchrist 4.3 71.6 0.66 1.07 0.62 0.7
Gale Sayers 5.3 76.7 0.62 1.36 0.63 0.65
Leroy Kelly 5.0 80.0 0.93 1.25 0.65 1
OJ Simpson 5.3 115.1 0.7 1.31 0.8 0.69
Franco Harris 3.9 77.3 0.75 0.98 0.54 0.74
Walter Payton 1 4.7 101.9 0.73 1.18 0.74 0.73
Earl Campbell 4.6 104.1 0.89 1.15 0.78 0.88
Tony Dorsett 4.5 86.3 0.46 1.13 0.65 0.45
John Riggins 3.7 69.8 0.91 0.94 0.54 0.91
Walter Payton 2 4.5 93.6 0.53 1.11 0.75 0.56
Marcus Allen 4.2 77.0 0.62 1.03 0.62 0.66
Eric Dickerson 4.8 112.4 0.89 1.18 0.9 0.94
Roger Craig 4.5 67.7 0.45 1.12 0.55 0.49
Herschel Walker 4.2 67.6 0.52 1.06 0.56 0.58
Bo Jackson 5.4 73.2 0.42 1.35 0.62 0.48
Thurman Thomas 4.6 86.3 0.52 1.16 0.77 0.64
Emmitt Smith 4.6 105.8 1.2 1.19 0.98 1.61
Barry Sanders 5.4 109.2 0.63 1.41 1.01 0.8
Terrell Davis 4.8 105.1 0.92 1.22 0.95 1.16
Ricky Watters 3.9 78.6 0.63 0.99 0.71 0.79
Jerome Bettis 4.1 86.6 0.45 1.04 0.79 0.59
Eddie George 3.9 86.0 0.53 0.96 0.77 0.69
Marshall Faulk 5.0 90.7 0.72 1.24 0.82 0.96
Curtis Martin 3.9 86.8 0.51 0.98 0.78 0.68
Corey Dillon 4.3 83.5 0.46 1.05 0.75 0.59
Ricky Williams 4.1 94.3 0.67 0.99 0.82 0.84
Jamal Lewis 4.7 96.1 0.55 1.12 0.83 0.67
Priest Holmes 4.7 101.5 1.3 1.14 0.88 1.62
Clinton Portis 4.7 98.8 0.75 1.14 0.85 0.91
Shaun Alexander 4.6 96.7 1.14 1.12 0.83 1.38
Edgerrin James 4.1 91.2 0.65 0.99 0.79 0.81
LaDainian Tomlinson 4.7 99.3 1.21 1.13 0.86 1.51
Tiki Barber 4.9 97.8 0.45 1.18 0.84 0.57
Brian Westbrook 4.7 72.3 0.36 1.14 0.63 0.45

Table 4: Rushing per touch and per game stats and same stats normalized to league average for that time period

The left three columns show per game and per touch rushing stats, while the right three columns shows those stats divided by the league average for each player over his best 4 year period.

Next, we have the receiving and total per game and per touch statistics.


Receiving Total
Players Y/R R/G Y/G TD/G Y/G TD/G Fmb/T
Bronko Nagurski 12.2 0.2 2.7 0 43.2 0.29 0.000
Spec Sanders 15.1 0.3 6.8 0.08 79.3 0.9 0.002
Marion Motley 13.2 0.9 11.9 0.08 75.29 0.55 0.008
Jim Brown 1 9.4 2.1 19.8 0.1 130.24 1.06 0.019
Jim Taylor 7.2 1.8 13.0 0.09 105.02 1.11 0.016
Jim Brown 2 10.3 2.5 25.9 0.25 130.39 1.13 0.023
Cookie Gilchrist 10.7 1.7 18.4 0.09 89.98 0.75 0.028
Gale Sayers 12.7 1.9 23.9 0.18 100.62 0.8 0.032
Leroy Kelly 12.9 1.7 22.0 0.15 102.07 1.07 0.015
OJ Simpson 13.3 1.3 16.9 0.16 132 0.86 0.021
Franco Harris 7.0 1.6 11.0 0.02 88.24 0.76 0.026
Walter Payton 1 9.3 2.5 23.0 0.08 124.94 0.81 0.019
Earl Campbell 4.6 1.2 5.6 0 109.71 0.89 0.021
Tony Dorsett 9.1 2.4 22.0 0.08 108.26 0.54 0.030
John Riggins 6.1 0.9 5.6 0.06 75.33 0.96 0.012
Walter Payton 2 10.0 2.9 28.8 0.11 122.33 0.64 0.014
Marcus Allen 9.6 4.0 38.6 0.2 115.67 0.82 0.023
Eric Dickerson 7.4 1.9 14.1 0.03 126.48 0.92 0.031
Roger Craig 8.5 5.1 43.0 0.13 110.69 0.58 0.017
Herschel Walker 10.8 3.8 41.3 0.12 108.95 0.63 0.018
Bo Jackson 8.8 1.1 9.3 0.05 82.47 0.47 0.020
Thurman Thomas 10.7 3.6 39.0 0.25 125.29 0.78 0.017
Emmitt Smith 6.4 3.7 24.0 0.05 129.85 1.25 0.010
Barry Sanders 7.6 2.3 17.7 0.08 126.91 0.7 0.007
Terrell Davis 7.8 2.5 19.4 0.08 124.49 1 0.011
Ricky Watters 7.9 3.3 26.4 0.02 105.06 0.64 0.012
Jerome Bettis 5.8 1.2 7.0 0.03 93.61 0.48 0.012
Eddie George 9.0 2.2 19.8 0.13 105.8 0.66 0.013
Marshall Faulk 10.5 5.6 59.2 0.43 149.87 1.15 0.006
Curtis Martin 6.9 3.3 23.0 0.05 109.84 0.56 0.007
Corey Dillon 7.7 2.0 15.5 0.06 98.97 0.52 0.013
Ricky Williams 8.1 3.5 28.2 0.07 122.48 0.74 0.018
Jamal Lewis 9.6 1.8 17.7 0.02 113.7 0.57 0.018
Priest Holmes 9.6 4.2 40.1 0.11 141.57 1.41 0.007
Clinton Portis 8.0 2.4 18.8 0.07 117.65 0.82 0.011
Shaun Alexander 7.2 2.2 15.7 0.14 112.33 1.28 0.011
Edgerrin James 7.2 3.1 22.2 0.02 113.35 0.67 0.010
LaDainian Tomlinson 7.9 4.1 32.4 0.16 131.76 1.37 0.008
Tiki Barber 8.7 3.6 31.8 0.08 129.53 0.53 0.012
Brian Westbrook 9.3 5.5 50.7 0.35 123.05 0.71 0.004

Table 5: Receiving and Total per touch and per game stats

This table shows the per game and per touch stats for receiving and total (rushing and receiving combined). The big change here is in the fumble calculation. Last time, I had fumbles/game, while now I have calculated fumbles per touch. So, I divided the number of fumbles by the sum of rushing attempts and receptions.

Finally, having compiled all of these numbers, I sorted by each of the per game/per touch categories, and then I normalized by dividing every player’s value by the highest value for that category. For fumbles, the best ranking was zero, so I couldn’t divide by that. In order to follow the premise that the best ranking would yield a value of 1, I normalize with the following formula: 1-10*(Fmb/T). This didn’t give an exact percentage, as the other values did, but it did give a range where the more fumbles, the lower the value.

In the tables below, the player with a 1 for a given category had the highest value in that category. For example Marion Motley had the highest yards/attempt, but Barry Sanders had the highest yards/attempt normalized against league average. Priest Holmes had the highest rushing TDs/G, both in quantity and when normalized against league average. Emmitt Smith had 92% of the rushing TDs/G that Priest Holmes had in quantity, but when normalized against the league average, he was 99% of Priest’s value.

  Rushing Rsh Nrm-League Avg
Players Y/A Y/G TD/G Y/A Y/G TD/G
Bronko Nagurski 0.74 0.35 0.22 0.91 0.33 0.27
Spec Sanders 0.93 0.63 0.63 0.93 0.5 0.41
Marion Motley 1 0.55 0.36 0.95 0.4 0.21
Jim Brown 1 0.89 0.96 0.74 0.88 0.8 0.54
Jim Taylor 0.86 0.8 0.78 0.86 0.72 0.61
Jim Brown 2 0.93 0.91 0.67 0.95 0.83 0.53
Cookie Gilchrist 0.75 0.62 0.51 0.76 0.62 0.43
Gale Sayers 0.92 0.67 0.48 0.96 0.62 0.4
Leroy Kelly 0.85 0.7 0.71 0.89 0.65 0.62
OJ Simpson 0.91 1 0.54 0.93 0.79 0.42
Franco Harris 0.67 0.67 0.57 0.7 0.53 0.45
Walter Payton 1 0.8 0.89 0.56 0.84 0.74 0.45
Earl Campbell 0.79 0.9 0.68 0.82 0.77 0.54
Tony Dorsett 0.78 0.75 0.35 0.8 0.64 0.28
John Riggins 0.64 0.61 0.7 0.67 0.54 0.56
Walter Payton 2 0.77 0.81 0.41 0.79 0.75 0.35
Marcus Allen 0.72 0.67 0.48 0.73 0.61 0.41
Eric Dickerson 0.82 0.98 0.68 0.84 0.9 0.58
Roger Craig 0.77 0.59 0.35 0.79 0.55 0.3
Herschel Walker 0.72 0.59 0.4 0.75 0.56 0.36
Bo Jackson 0.93 0.64 0.32 0.96 0.61 0.3
Thurman Thomas 0.8 0.75 0.4 0.82 0.76 0.4
Emmitt Smith 0.79 0.92 0.92 0.84 0.97 0.99
Barry Sanders 0.94 0.95 0.48 1 1 0.5
Terrell Davis 0.82 0.91 0.71 0.86 0.94 0.72
Ricky Watters 0.67 0.68 0.48 0.7 0.7 0.49
Jerome Bettis 0.71 0.75 0.35 0.74 0.78 0.36
Eddie George 0.66 0.75 0.41 0.68 0.77 0.42
Marshall Faulk 0.86 0.79 0.55 0.88 0.81 0.59
Curtis Martin 0.68 0.75 0.39 0.7 0.77 0.42
Corey Dillon 0.74 0.73 0.35 0.74 0.74 0.37
Ricky Williams 0.71 0.82 0.52 0.7 0.81 0.52
Jamal Lewis 0.8 0.83 0.42 0.79 0.82 0.41
Priest Holmes 0.82 0.88 1 0.81 0.87 1
Clinton Portis 0.81 0.86 0.58 0.81 0.85 0.56
Shaun Alexander 0.79 0.84 0.88 0.79 0.83 0.85
Edgerrin James 0.7 0.79 0.5 0.7 0.78 0.5
LaDainian Tomlinson 0.81 0.86 0.93 0.8 0.85 0.93
Tiki Barber 0.84 0.85 0.35 0.84 0.83 0.35
Brian Westbrook 0.81 0.63 0.28 0.81 0.63 0.28

Table 6: Normalized values for rushing per touch and per game stats & same stats normalized to league avg. for that time period

The table above contains the normalized values for rushing numbers, and the table below constains them for receiving and total numbers. They are normalized by dividing the players quantitative statistic by that of the leader in that category.


Receiving Total
Players Y/R R/G Y/G TD/G Y/G TD/G Fmb/T
Bronko Nagurski 0.81 0.04 0.05 0 0.29 0.2 1
Spec Sanders 1 0.05 0.11 0.17 0.53 0.64 0.98
Marion Motley 0.87 0.16 0.2 0.18 0.5 0.39 0.92
Jim Brown 1 0.63 0.38 0.33 0.23 0.87 0.75 0.81
Jim Taylor 0.48 0.32 0.22 0.21 0.7 0.79 0.84
Jim Brown 2 0.68 0.45 0.44 0.58 0.87 0.8 0.77
Cookie Gilchrist 0.71 0.31 0.31 0.21 0.6 0.53 0.72
Gale Sayers 0.84 0.34 0.4 0.42 0.67 0.57 0.68
Leroy Kelly 0.85 0.31 0.37 0.34 0.68 0.76 0.85
OJ Simpson 0.88 0.23 0.28 0.37 0.88 0.61 0.79
Franco Harris 0.47 0.28 0.19 0.04 0.59 0.54 0.74
Walter Payton 1 0.61 0.44 0.39 0.19 0.83 0.57 0.81
Earl Campbell 0.3 0.22 0.09 0 0.73 0.63 0.79
Tony Dorsett 0.6 0.43 0.37 0.19 0.72 0.38 0.7
John Riggins 0.41 0.16 0.09 0.13 0.5 0.68 0.88
Walter Payton 2 0.66 0.51 0.49 0.25 0.82 0.45 0.86
Marcus Allen 0.64 0.72 0.65 0.45 0.77 0.58 0.77
Eric Dickerson 0.49 0.34 0.24 0.07 0.84 0.65 0.69
Roger Craig 0.56 0.91 0.73 0.3 0.74 0.41 0.83
Herschel Walker 0.72 0.68 0.7 0.27 0.73 0.45 0.82
Bo Jackson 0.58 0.19 0.16 0.12 0.55 0.34 0.8
Thurman Thomas 0.71 0.65 0.66 0.59 0.84 0.55 0.83
Emmitt Smith 0.43 0.67 0.41 0.11 0.87 0.88 0.9
Barry Sanders 0.5 0.42 0.3 0.18 0.85 0.5 0.93
Terrell Davis 0.51 0.44 0.33 0.19 0.83 0.71 0.89
Ricky Watters 0.53 0.59 0.45 0.04 0.7 0.45 0.88
Jerome Bettis 0.39 0.21 0.12 0.07 0.62 0.34 0.88
Eddie George 0.59 0.39 0.33 0.29 0.71 0.47 0.87
Marshall Faulk 0.7 1 1 1 1 0.82 0.94
Curtis Martin 0.46 0.6 0.39 0.11 0.73 0.39 0.93
Corey Dillon 0.51 0.36 0.26 0.15 0.66 0.37 0.87
Ricky Williams 0.54 0.62 0.48 0.16 0.82 0.53 0.82
Jamal Lewis 0.64 0.33 0.3 0.04 0.76 0.4 0.82
Priest Holmes 0.64 0.74 0.68 0.26 0.94 1 0.93
Clinton Portis 0.53 0.42 0.32 0.15 0.79 0.58 0.89
Shaun Alexander 0.48 0.39 0.26 0.32 0.75 0.91 0.89
Edgerrin James 0.48 0.55 0.37 0.04 0.76 0.47 0.9
LaDainian Tomlinson 0.52 0.74 0.55 0.37 0.88 0.97 0.92
Tiki Barber 0.58 0.65 0.54 0.18 0.86 0.38 0.88
Brian Westbrook 0.61 0.98 0.86 0.8 0.82 0.5 0.96

Table 7: Normalized values for receiving and total per touch and per game stats

Let’s take a look at a couple of items of note from these tables before moving on to the final rankings. The first thing that jumps out at me is that Barry Sanders is at the top in yards/attempt and yards/game normalized against the league averages. Emmitt Smith and Terrell Davis are the only ones within 90% of Sanders in yards/game while 7 other players are in yards/attempt. This is notable to me because I added this portion in as way to determine how the players looked against the rest of the league during their time. Sanders dominated a 4 year period in terms of pure running like no other player. His touchdown numbers were way down, but in the period that this analysis covers, he played in a run and shoot offense. Other than Herman Moore, he had smaller receivers (generally less effective blocking downfield), and he did not have a lead blocker during this period. Imagine his production with the Cowboys or 49ers during this period! In terms of scoring TDs, nobody was better over a 4 year period than Priest Holmes. Emmitt was right up there, though. And, of course, as we saw last time, Marshall Faulk just dominates all receiving categories (except Y/R). Spec Sanders and Marion Motley were obviously very dynamic players, with such high rankings in yards/attempt and yards/reception. Finally, I expected Jim Brown to dominate the rankings in normalization against league averages, but he didn’t. Perhaps the run game was so dominant then that no team used a single back as pre-dominantly as the ’80s and ’90s.

With that said, let’s see what the final rankings came out to be. For this analysis, I completed the rankings differently than in the past. Instead of ranking each player 1 through 40 for each category and averaging the ranking, I took the sum of the the normalized values for rushing, rushing numbers normalized against league average, receiving, and total numbers. This way, the largest number represented the highest normalized value based on all categories. The results of this process are in Table 8.

Rk Players Normalized Total
1 Marshall Faulk 10.94
2 Priest Holmes 10.56
3 LaDainian Tomlinson 10.12
4 Emmitt Smith 9.7
5 Jim Brown 2 9.4
6 Shaun Alexander 8.98
7 Brian Westbrook 8.96
8 Terrell Davis 8.87
9 Jim Brown 1 8.8
10 Thurman Thomas 8.76
11 OJ Simpson 8.64
12 Leroy Kelly 8.57
13 Barry Sanders 8.53
14 Marcus Allen 8.2
15 Jim Taylor 8.19
16 Clinton Portis 8.14
17 Tiki Barber 8.13
18 Eric Dickerson 8.12
19 Walter Payton 1 8.12
20 Ricky Williams 8.03
21 Gale Sayers 7.96
22 Walter Payton 2 7.91
23 Roger Craig 7.82
24 Herschel Walker 7.73
25 Edgerrin James 7.54
26 Spec Sanders 7.53
27 Jamal Lewis 7.37
28 Ricky Watters 7.37
29 Eddie George 7.34
30 Curtis Martin 7.33
31 Earl Campbell 7.27
32 Cookie Gilchrist 7.06
33 Tony Dorsett 7.01
34 Corey Dillon 6.85
35 Marion Motley 6.71
36 John Riggins 6.57
37 Bo Jackson 6.49
38 Franco Harris 6.44
39 Jerome Bettis 6.33
40 Bronko Nagurski 5.22

Table 8: Players Ranking

The top of this list looks similar to that of the last post, with Faulk at the top. Because of this, and the discussion that perhaps the receiving portion is weighted to high, I ran another ranking. In this case, I weighted the rushing portions double that of the receiving and total values. I did this by multiplying the sum of the receiving/total values by 0.5. That’s why this ranking has smaller numbers.

Rk Players Normalized Total
1 Priest Holmes 7.97
2 Marshall Faulk 7.71
3 LaDainian Tomlinson 7.65
4 Emmitt Smith 7.57
5 Jim Brown 2 7.11
6 Shaun Alexander 6.98
7 Terrell Davis 6.92
8 Jim Brown 1 6.8
9 Barry Sanders 6.69
10 OJ Simpson 6.62
11 Leroy Kelly 6.5
12 Eric Dickerson 6.46
13 Jim Taylor 6.41
14 Thurman Thomas 6.35
15 Clinton Portis 6.3
16 Walter Payton 1 6.2
17 Brian Westbrook 6.19
18 Tiki Barber 6.09
19 Ricky Williams 6.05
20 Gale Sayers 6.01
21 Marcus Allen 5.91
22 Walter Payton 2 5.89
23 Earl Campbell 5.89
24 Spec Sanders 5.78
25 Edgerrin James 5.76
26 Jamal Lewis 5.73
27 Roger Craig 5.58
28 Herschel Walker 5.55
29 Ricky Watters 5.55
30 Curtis Martin 5.52
31 Eddie George 5.52
32 Cookie Gilchrist 5.37
33 Tony Dorsett 5.31
34 Corey Dillon 5.26
35 John Riggins 5.15
36 Bo Jackson 5.12
37 Marion Motley 5.1
38 Franco Harris 5.02
39 Jerome Bettis 5.01
40 Bronko Nagurski 4.02

Table 9: Players Ranking with rushing values weighted double the receiving values

In this ranking, the dual threat players, like Brian Westbrook, Roger Craig, and Thurman Thomas drop pretty significantly. Faulk only drops to second behind Priest Holmes, though, and the rest of the top 6 stay in the same position.

Conclusions: Like BTSC member steeler.lifer said, the best rise to the top, however you analyze it. On the other hand, there are players that are on this list (and high on this list) that will not be in the HOF. Some performed at historic levels over a four year period but will be argued against due to lack of longevity. There are others in the HOF that weren’t even considered. The oldest player looked at on this list, Bronko Nagurski, didn’t dominate statistically against league averages the way that Barry Sanders and Priest Holmes did. In fact, he’s the lowest on either overall ranking list. He may be hurt by the passing numbers, though. But, either way, these players were certainly elite in their time, if only for a short time for some of them.

Well, I’m anxious to hear what you guys think about this one. Let me know your thoughts.

Tags: , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , ,

11 Responses to “Another Comparison of RBs’ Best 4 Years”

  1. Geoffrey B Says:

    I was surprised by how well Jim Taylor held up, you don’t hear much about him when they gush on those old Packer teams.

    Stats are always disappointing in some area, and here you see guys ranked better than others that you just can’t believe were actually better.

    Jerome Bettis in his best seasons was an elite RB, to see him rank behind Shaun Alexander in every category just really points out how stats can never tell the whole tale. I used those two because both had great O-Lines and were the focus of their team. I don’t think you can watch film of the two and say Shaun was better.

    On the other hand, Priest Holmes for those years was so good, he deserves his ranking.

    Marshall Faulk was a great running back, and as a receiver he was the best RB possibly ever, and the stats show that.

    Great work again Wolf Pack. If you had a full-time job doing this and some interns to run numbers all day we could add in percentages of teams offense and scoring. I’d like to see that because some teams score tons of points and others don’t, and I think the disparity in TD numbers would look a little different, esp for Jim Brown and some other old timers.

    Emmitt scored tons of TD’s, but his team scored at a high rate, so that ups his numbers without upping his percentage.

  2. Geoffrey B Says:

    I ran a small sample on just total TDs as a percentage of team TD’s:

    Emmitt Smith: 43.9%

    J. Brown1: 36.6%

    J. Brown2: 39.1%

    Tomlinson: 44.1%

    B. Sanders: 28.5%

    Bettis: 23.4%

    Holmes: 38%

    I don’t know how you would normalize that with the league average, or if it would just be worth a different look to the equation, Jerome would move up as the Steelers scored 64% as many TD’s as Holmes’ Chiefs for example, I don’t know if any others would move that significantly.

    You just need some stats interns to sit around and run all these numbers.

    • wolfpacksteelersfan Says:

      Hey, thanks for that, Geoffrey. I agree, there are some things that could be added, for sure. For example, none of this considers the average defense that players faced, or whether they played on a more prolific offense, like you said. Or, what kind of QB did they play with? Or, was the play calling so predictable that the defense knew the run was coming, but the guy still gained 6-7 yards because he was just that powerful? I think that may have been true in Bettis’s prime, particularly in the second half of games that the Steelers led.

  3. jeff Says:

    Dude,
    You…..are a dork. I bet you have a pocket protector with a slide rule in it.
    Gimme a break. Who (but dorks) cares about this?

    Football is all about Sundays, BBQ, Beer, and watchin’ the Almighty Steelers kick arse. I mean, ….FRAK! You want to turn this into A Beautiful Mind.

    Gimme a frakkin break. You need to unwind. If you’re in CA come over to my house. I’ll BBQ, we’ll drink some beer. We’ll scream, holler, curse, and cheer the Stillers, and you’ll find out what football is really supposed to be about.

    LambertsLunatic

    • wolfpacksteelersfan Says:

      Look, if you don’t like this, don’t read it. No need for name calling or personal attacks. I, and several other Steelers and other football fans, enjoy these analyses. And, no, I don’t have a pocket protector or slide rule. Give me a break! If you want to write on here, don’t be a jerk.

    • Navy_Steeler Says:

      If you think football is just an excuse to throw back a few beers on Sunday, more power to you. However, I think that most people who are serious fans of the game find it interesting to compare the best players from different eras against each other. If this type of analysis drives you to childish name-calling, perhaps you shouldn’t spend your time reading about it in the first place.

  4. Mr.GJG Says:

    Jim Brown was the most dominant (relative to his peers)RB of all time.
    You should compare players based on their peers, it will yield different, but
    more accurate results.
    Otherwise interesting stuff.
    Jeff, you’re a miserable little shit.

    • wolfpacksteelersfan Says:

      I did do a normalization to the team rushing averages for those years that I analyzed. It’s not quite as accurate as comparing to other players, but the amount of work required to pull data for all players for each year analyzed would have been enormous.

Leave a reply to Geoffrey B Cancel reply