Okay, I looked at the 2004 game logs, and I definitely think last season was a fluke. I still firmly believe that Johnson was very inconsistant last year, but in 2004 he had no such problems.
Johnson in 2004
0-10 points: 5 games (31.25%)
10-20 points: 9 games (56.25%)
20+ points: 2 games (12.5%
I'm OK with your logic. However, all I would say definitively is that Rudi was more consistent in 2004 than in 2005.What do you think of this statistic?
Excluding LJ, Standard deviation of the top 6 backs in 2005:
LT 13.23
SA 10.02
Tiki 8.81
Edge 7.5848
Rudi 7.5840
Portis 7.41
Suggests Rudi was more consistent than every back save Portis (due to Week 17 outlier as per my earlier posts)
Knowing standard deviation is friendly to lower means, expanding this analysis to normalize the data vs. total season points produces this category:
Standard Deviation/Total Points Ratio
SA 10.02/363.8 = 0.02754
Edge 7.5848/268.3 = 0.02827
Tiki 8.81/305.0 = 0.02888
Portis 7.41/244.1 = 0.03036
Rudi 7.5840/226.8 = 0.03344
LT 13.23/317.8 = 0.04163
Both studies show LT was the most inconsistent RB last season. Part of this most likely is attributed to him playing through injury. Oh wait! The same could be said of Rudi, who clearly was second to LT in this normalized statistic.
Interesting to me, but perhaps Rudi's early-season knee injury is responsible for weaker early games and hence more inconsistency than shown in 2004. By expansion, a larger theme is that injury may cause greater inconsistency among RBs. Makes sense to me.