Another good question - what to do with rovers? We treat them like centers in the ATD and in the HOH projects, and I think that's the right thing to do. But in this project, we take center - already the most crowded forward position - and add rovers to it, then we've really cluttered it. We'd have centers and rovers missing the team that would have been 1st teamers had they played LW or RW.
On the other hand, if we decide to just have a 1st, 2nd and 3rd team rover for the 1910-1922 seasons, being that there's no rover in the east, in all PCHA seasons except 1916 and 1917, every rover would be guaranteed an all-star team spot. I don't like that either, despite the fact that they were typically excellent players.
So... what do you think we do? The only thing I can think of that's even close to a solution, is to name three full teams and then "on the side" name a 1st and 2nd team rover.
Alternate "solution" - name rovers on the all-star teams, BUT the 3rd rover spot is reserved for either a rover or a center, whoever merits it more.
Last "solution" - make two spots on every AST that are "C/R" - whichever two centers or rovers most merit the spots, get them.
Fourth solution - last one for real this time. Just name three forwards on each team, including rovers, regardless of position. There was so much switching around that we are never going to have it 100% correct anyway. This era was disproportionately strong at C and Rover anyway, as opposed to wing. These are the drafted players from this year who played mostly in the pre-NHL-ASTs era, listed by predominant position:
best Centers and Rovers: Lalonde, T.Smith, Foyston, Nighbor, MacKay, Morris, Keats, Malone, Frederickson, Taylor, Boucher, H.Smith, Morenz
best right wingers: Pitre, Oatman, Broadbent, Walker, Dye,
best left wingers: Crawford, Roberts, Harris, Noble, Denneny, Hay, Joliat
I see upsides and downsides to each solution.