That's probably from seeing the breakdowns Chris has done splitting the 460s into smaller groups. His groups are based more on the 34 theory but with substitutions making up smaller groups. Obviously the sheet would have to be either on a larger piece of stock, or have a less regular and unbalanced arrangement.
I lean more towards your theory of 12 subject sheets and tried to use the pop report numbers to disprove his groups. And as far as I'm concerned his groups held up very well. At some point I hope he shares what he's come up with more widely, but that's up to him.
The Hoe #5 did have a 19" track, and the floorplan shown in the old article did show several hoe #5 presses at ALC. I believe the article was in scientific American but their archive is behind a paywall now and I can't find the copy I thought I'd saved. I also couldn't find the Hoe press ad I'd seen.
But that was only a diagram of one floor of a very large firm. Most large printing plants have several sizes, unless they're doing huge volume of only one sort of item. A newspaper would only have one very large fast press, but a general business would have at least a few different sizes.
There is a lot of small detail that leads me towards believing a far more complex situation for the sheets.
1) There are groups that work perfectly with a 12 subject sheet.
BUT there are also groups that Just don't and many of those either are or are much closer to 17/34
2) The fragment of packing log that specifies "other than philadelphia area" implies a different group of cards for that area. The easy way is to make a different sheet so there's no concern about mixing up which players went where. (Alternately they could have pulled the players they didn't want to send to phillly, OR perhaps ALC packed them in stacks of one player. Short of some miracle, we'll probably never know. A sheet turning up is more likely than an intact boxful sent from ALC to any of the factories) *
3) There's a lot of evidence for each of the current groups being sent to press multiple times for the same brand even within a series.
Examples
Tinker hands on knees - Normal, with traces of Chicago showing underneath cubs, and with Chicago showing clearly. At least two distinct runs, probably three.
Dygert - Comes both with and without red lips. In roughly equal quantities, so it's not a missing color or fading.
Obviously the Demmitt and Ohara show that the sheets were redone at the very least for Polar Bear.
Conroy fielding and Ritchey - Both have differences that split clearly between 150 and 350 backs.
Wilson - Orange or yellow sky. They're actually quite different, and it ought to be a more recognized variation. Certainly it's more of a different plate situation than Nodgrass or Dopner.
There's more, most of it much less obvious.
4) If we assume a simple sheet with a straightforward layout, and the same sheet used for all brands those subjects were available with then the pop report numbers should have roughly the same distribution across brands. This is generally true for Piedmont and Sweet Caporal, but breaks down for the other brands. So either there are some odd patterns to what cards get sent in, or the sheets for some brands were entirely different.
5) Hoe made 2 color presses, and I see some indication that some T206s may have been printed on a 2 color press. Quite often when there's a small color shift two colors are shifted equally. The Hoe#5 wasn't two color.
6) The number of dual name cards compared to simple miscuts showing two of the same name I think supports at least some sheets having an unbalanced arrangement.
So I suppose both the 12 subject guys and the 17/34 subject guys can call me a heretic. I think what's likely is that for brands like Hindu the sheets were 12 subjects. And for Piedmont they were probably 17/34 maybe more.
So both camps are probably both right and wrong all at the same time. Confused yet
That's why I like the wide range of efforts. Tracking the double name cards, the plate scratches, the factory numbers in the margins, the cutting marks (Why the Heck are they on the back when the cutting was done from the front?!) All that and more will eventually give us a much better picture of the production.
I also think we need to redefine the print groups. (Sorry Scot) And to look at each brand and series as its own set. At the level of the basic subjects there's a lot of overlap, but when the small details are looked at there's probably a lot less, possibly none.(So a common subject front will probably have small but identifiable details with no crossover between brands. )
To me it's more about what's possible, what would make sense in a manufacturing context, and what part of that can be proven.
I know my own theories are pretty far out there, and may not be something I can prove or disprove within my lifetime. (There are people who have done the same thing for stamps and taken decades to chart a handful of plates for one stamp. And that's comparatively "easy" since the sheet size is known and usually there are blocks of stamps available)
Steve B
* But damn, can you imagine one of the packers or loading dock guys or even a janitor liberating an entire boxful of Wagners or Planks before it hit the dumpster?
