Introduction to these player comments
Index of player comments
Glossary of terms
Jerome Bettis career statistics

Among all above-the-baseline running backs last year, Bettis was the one who was most allergic to catching the forward pass. He had only 8 receptions for 48 yards. His fantasy points came 96.5 percent from rushing and 3.5 percent from receiving. Here's the full list; runners at the top, receivers at the bottom:

Name                PCTRSH
--------------------------
Jerome Bettis         96.5
Anthony Thomas        90.0
Stephen Davis         89.4
Antowain Smith        88.2
Dominic Rhodes        88.0
Curtis Martin         86.8
Stacey Mack           86.3
LaDainian Tomlinson   83.3
Corey Dillon          82.4
Shaun Alexander       82.3
Eddie George          81.6
Mike Alstott          81.5
Michael Pittman       81.3
Lamar Smith           79.0
Maurice Smith         78.5
Garrison Hearst       78.0
Ricky Williams        73.8
Priest Holmes         73.5
Ahman Green           73.0
Tiki Barber           65.7
Marshall Faulk        61.7
Charlie Garner        56.3
Duce Staley           49.3
Warrick Dunn          46.0

Question: is there a general tendency for rushing-dependent backs to hold value better than more versatile backs?

Answer: No.

We'll proceed with the usual plan: look at all RBs who finished above the baseline and then played at least 8 games the following year. Then divide that group into two halves: the rushing-heavy backs and the receiving-heavy backs. The two groups did almost equally well the next year:

• The rushing-dependent backs: 32 percent improved the next year (as measured by fantasy points per game) and 68 percent declined. Overall, they declined by 12 percent.
• The receiving-dependent backs: 35 percent improved the next year (as measured by fantasy points per game) and 65 percent declined. Overall, they also declined by 12 percent.

There is no general tendency for one-dimensional backs to hold value better or worse, on the whole, than versatile backs. Nothing to see here, folks. Move along.

Complete data set