This article is very similar to the one published earlier on
Drafting Wide Receivers Based on Years of Experience. In this
article, I will compare the previous year’s statistics with average
draft position the next year in order to determine which running backs
the average manager assumes will improve the most. Once we know which
running backs the average manager expects to improve, we can determine
which running backs should be targeted by looking at their statistics in
the next season.
In this paragraph I will explain which techniques I used to complete
this analysis but it is a little complicated and not required to
understand the major part of this analysis. The main difficulty
encountered when making this analysis was to find a way to compare draft
position and statistics from every year. I decided to take the previous
year’s statistics and make a regression based on average draft position
in order to determine an “expected” number of fantasy points for the
next season simply based on where the wide receivers were drafted. This
therefore allowed me to compare how many points the average manager
expects a wide receiver to get compared to the number of points he had
in the previous year. All the numbers used in this article are based on
the WCOFF scoring system with one point per reception, are in terms of
points per 16 games and only include wide receivers who played more than
eight games in that season. When comparing with the next season’s
statistics I also used points per 16 games but if a player did not play
enough games he was given 100 as his points total for the season.
Since there are
fewer running backs than there are wide receivers I decided to only look
at the top 30 running backs and not break them down into various groups
as I had done for wide receivers. The first column of the following
table contains the years of experience in the NFL each wide receiver has
and is followed by the number of wide receivers that are included in
each category from 2000 to 2007. The next column contains the average
number of expected points subtracted by the average number of points in
the previous year. The fourth column contains the average number of
points wide receivers improved in the next season and the final column
is the difference between the two where the higher number means a larger
overpayment and higher risk.
|
Yrs |
Nb |
Avg
Manager |
Next Season |
Difference |
|
2 |
33 |
31.5 |
18.1 |
13.4 |
|
3 |
33 |
5.7 |
-25.3 |
30.9 |
|
4 |
32 |
-15.0 |
-44.3 |
29.3 |
|
5 |
32 |
7.9 |
-23.8 |
31.7 |
|
6 |
27 |
-14.9 |
-4.3 |
-10.6 |
|
7 |
26 |
-33.2 |
-59.9 |
26.6 |
|
8 |
21 |
-20.5 |
-32.5 |
12.0 |
|
9 |
18 |
-43.4 |
-92.6 |
49.2 |
In that chart, the higher the number
the bigger the overpayment for a player is. It is no surprise that
second year running backs improve the most and the average manager
expects them to improve the most but to my surprise they are generally
better bargains than third, fourth and fifth year backs. This is however
a small sample again and if we focus only on the last five years instead
of the last seven the numbers are very similar than for running backs in
seasons three through five. I thought it would however be interesting to
group them by even and odd years of experience as we had done for wide
receivers:
|
Yrs |
Nb |
Avg
Manager |
Next Season |
Difference |
|
Even |
113 |
-2.4 |
-14.3 |
11.9 |
|
Odd |
109 |
-11.1 |
-44.2 |
33.2 |
Once again there is an interesting
difference between the two with players with an even number of years of
experience being undervalued by about 21 points compared to those with
an odd number of years of experience. This is becoming an interesting
trend as we had found an even bigger difference for wide receivers but
in that case it was the receivers with odd experience years that were
undervalued. That means these are not statistical errors so there are
two ways to explain this. The first possibility is that the samples are
too small but we have over 100 backs in each group and a difference of
21 points seems to be too significant to be simply explained by a small
sample size. The other possibility is that running backs and wide
receivers improve at different rates and the average manager does not
account appropriately for those. To be honest however I am not yet
convinced enough about this theory to use it in the projections or my
drafts but it is something that is worth keeping in mind. If you’re
hesitating between two players you might want to give the edge to the
wide receiver with an even number of years of experience or the running
back with an odd number of years of experience.
Follow
us on Twitter
RSS
Feed