How many T206s to a row? (revisited I'm sure)

[t206hound]

I've been trying to figure out this row size thing for a while. There has to be some explanation why the 17 card scenario fits well for some issues, but seems forced for others. There are a few back subsets that are relatively small, in which all of the subjects have been confirmed that lead me to a theory.

The Brown Hindu subset size of 136 which includes 34 Southern Leaguers definitely leads one to believe that the row size was 17. The "Sheet Mystique" article on T206Resource.com speculates that each image could have been repeated vertically three times for a total of 102 cards on a sheet (17 cards per row; 6 rows per sheet). The entire subset could be printed on four sheets. I can't find any reason to believe that this isn't the case for this print run.

But the Old Mill Southern League back has a series of 48 cards. According to Scot Reader in Inside T206,

"the 150/350 SL subjects and 350-only SL subjects are available in similar quantities with the Old Mill Southern back, which is a 350 series back type. This suggests that both the 150/350 SL group and the 350-only SL group completed a full print with the Old Mill Southern back."

If the 48 cards with OMSL backs were distributed in equal numbers, how could the row size be 17? I think the explanation may be staring us in the face. Based on the subset size of 48, is it POSSIBLE that ATC decided to go to 16 cards per row (just leaving a little extra space on each side of the sheet)? And maybe they only repeated each player twice vertically rather than three times? If that is possible, then all 48 players could have been on a single sheet, thus guaranteeing equal distribution.

I'll also look at the Sovereign 460 subset. It's comprised of 46 regular cards and six super prints for a total of 52 cards in the series. How can this possibly fit into the 16 or 17 cards per row scenarios? While we can speculate that there must have been some cards short printed, there does not appear to be a consensus on exactly which cards they were.

Some believe that the super prints are much tougher to find in this series, so let's assume (for the sake of argument) that they WERE short printed. Note that you can choose any six cards, but it's simpler to choose these six. If that's the case, then we are potentially looking at 16 cards per row again, with six rows per sheet. Thus, each sheet would contain all 46 of the regular cards and 2 of the super prints. Repeat twice to substitute in the remaining four super prints. This yields a three-to-one ratio of regular cards to super prints. This is obviously pure speculation, and I don't have hard facts to prove any card is a 3-1 short print. It's just a guess.

Is it POSSIBLE that the earlier print runs with the various backs were 17 cards per row, and that for some reason (based perhaps solely on the number of subjects to be printed), they switched to 16 cards? For me, at least right now, based on what I know (which granted isn't much), changing from 17 cards to 16 cards seems to be a reasonable theory. Did it really HAVE to be the EXACT SAME number of cards per row very EVERY print run?

Thoughts?