Nice stuff DBB. You know what I really like to see someone do? (Obviously this is a hint for YOU to do such I report) I'd love to see the change in average across a span of years for certain positions.
Say for instance Runningback. On average how does a running back change from freshman to sophomore, SO to JR, JR to SR. What I think would be interesting is numbers that show what could be generally expected at certain positions. That we we could say (roughly) that this season Antonio Pittman should be expected to gain X total yards, X yards per carry, X fumbles, etc. Obviously, the report would need to cover a large group of RBs (or whatever position) to give us roughly accurate percentages. Food for thought.
Your instincts are amazing. When you asked for this, I wasn't really sure why, but I did it anyway (I actually like to do this stuff

Assumptions and Constraints
- I only went back to 1999 because the NCAA web-site doesn't have data going back farther that is available in a readily accessible format. It's still a lot of data.
- I only used the top 100 rushers for each year back to 2000, and only the top 50 for 1999; because that is the data the NCAA web-site had. It's still a lot of data.
- No filtering was done based on how many OL the team returned from the previous year. That's just too much work. I mean this is 215 different performances here. I don't know if you like spending your evening working through the web-sites of every Athletic Department on Earth but...:tic:
- The overall average improvement for those who had been ranked the previous year was 6.998 yards per game.
- The overall standard deviation was 27.579. For those of you who went to Notre Dame this means that there was a wide degree of variability in the results. For those of you who went to Michigan this means that some guys did a lot better and some guys did a lot worse.
- Sophomores who had been ranked the previous year averaged 5.59 more yards per game than they did as Freshmen.
- Juniors who had been ranked the previous year averaged 11.30 more yards per game than they did as Sophomores.
- Seniors who had been ranked the previous year averaged 4.75 more yards per game than they did as Juniors.
As you might expect, defenses keyed on guys who had been ranked the previous year. This is borne out by the fact that the average difference in yards per carry was -0.10 yards. That's right, guys who were ranked the previous year got 6.998 more yards per game but -0.10 fewer yards per carry. For those of you who went to Michigan, that means their coaches fed them the ball more and defenses focused on stopping them.
It doesn't take a rocket scientist to figure out that the Seniors get more attention from opposing coaches/defenses than do the underclassmen. But wouldn't a highly productive Sophomore also become a point of focus for teams he faced as a Junior?
To answer this question, I filtered the data to better match Antonio Pittman's situation. Averages were calculated including only those who:
- weren't QBs
- had > 100 yards per game
- had > 5 yards per carry
- had not missed a season (these data points skewed the results, as these were typically athletes returning from injury)
By comparison, Seniors who fit this description as Juniors averaged 12.65 FEWER yards per game and 0.70 FEWER yards per carry.
If you filter further for only those who had gained > 110 yards per game (like AP), you find that the Juniors averaged 0.33 yards FEWER per game and 0.54 FEWER per carry.
This suggests that Antonio Pittman will average 110.6 yards game and a little under 5 yards per carry. When you factor in AP's improvement over the course of the year last year, I stongly suspect that these numbers are lower than his actual production will be.
The real moral of the story is that there really isn't that much difference ON AVERAGE; and there is the chance that he could do much better and a chance that he could do much worse.
OK, show of hands: How many think that he'll do worse?