Performance in Season After Breakout
Since Davis finished as RB14 last season (per pro-football-reference.com), I decided to look at a large sample of players that had finished as RB14 or higher in their first feature season (rookie season or otherwise). Players who finished lower in their first feature seasons had a lot more room for improvement, thus making the comparison apples and oranges (or at least that is my operating theory here).
Incidentally, this criteria eliminates Ricky Williams and Travis Henry from the discussion, as neither finished in the top 14 in their rookie seasons (in which each was his team's feature back).
I used per game averages to allow for fluctuations in games played. Percent change in fantasy points per game from first feature season to next:
Emmitt Smith: +42.1%
Tomlinson: +39.1%
Terrell Davis: +22.8%
Portis: +16.6%
Abdul-Jabbar: +9.9%
Lewis: +8.7% (using third season instead of second, which was lost to injury)
James: +7.1%
Green: +5.1%
Robert Smith: +4.3%
Alexander: +3.7%
Watters: +0.4%
Jamal Anderson: +0.0%
Sanders: -0.8%
Bettis: -5.1%
Faulk: -5.1%
Martin: -5.4%
McAllister: -8.3%
George: -8.7%
Dillon: -13.7%
Holmes: -14.2% (missed 8 games, coaching decision or injury?)
Barber: -15.5%
Stephen Davis: -15.5%
Thomas: -18.1%
Dunn: -19.9%
Kaufman: -30.6%
Taylor: -34% (missed 6 games, per game numbers affected by injuries?)
Gary: -37.9% (only played 1 game in next season, so really not useful here)
Levens: -45.8% (missed 9 games, per game numbers affected by injuries?)
Mike Anderson: -67.4% (Davis return = RBBC, so really not useful here)
I pulled these names by looking at the top 14 in random seasons within the past several years, so hopefully there is no bias here. I don't have a database, so this was painfully manual... perhaps someone else could actually perform a more complete study...
Observations:
1. There is no clear trend, at least not to me. IMO that suggests that Davis could go either way... it is no given that his numbers will go up.
2. This is essentially the cream of the RB crop over the past decade. I am not at all certain that Domanick Davis belongs in this group, talent-wise. So he may lag the performance of this group.
3. Only 4 players out of the 29 named above achieved a 10% gain or better (LT, Portis, Emmitt, TD). I definitely don't see Davis belonging in this group. (Abdul Jabbar was right there at 9.9%, so that helps Davis's cause.)
4. 17 of 29 players declined, while only 11 gained. It is true that a number of the decliners suffered injuries or major changes in situations. But even if you weed out the injuries and major situation changes, the odds still don't seem favorable for improvement.
The bottom line is that any improvement seems fairly optimistic and a decline seems just as likely, if not moreso, no matter what the reason ultimately is for the decline.
Scaling Up
Many have chosen to first scale up Davis's numbers from last season and then apply predicted improvement to that.
There are a number of the players above who also did not claim the feature back role until partway through their breakout season and maintained essentially the same role & situation in their next season: Dillon, Lewis (one season removed to avoid the injury), Green, Alexander, and Portis. Again, a group of very talented backs, likely more talented than Davis. It seemed similarly promising to scale up the statistics of each of these guys, just as people are doing for Davis.
But here's the rub: These other backs averaged a gain of only 4.1% in fantasy points per game the season following their breakout season, and that is from their straight numbers, not from scaled up numbers.
I would say this shows that scaling up is probably too optimistic to begin with, much less applying any improvement on top of that.
I would be interested in seeing someone post success stories of players similar to Davis in their breakout season, particularly if the breakout season was their rookie season. The ones I posted above are the most similar I am aware of, and they don't suggest there will be much improvement for Davis.
Touches
Davis had 285 touches last season and was unable to stay healthy. Of the 5 RBs listed above whose situations most resembled Davis's situation, Dillon's touches rose the most the following season... by 11.5%. The others: Green (+8.9%), Portis (+7.2%), Lewis (+5.7%), Alexander (0%... +1 touch). So it seems that there is some history to suggest his touches aren't likely to rise above 320.