I think the two of you are talking circles around each other, referring to different definitions of the word "Accurate".
Remember that in old days, powder was even more difficult to obtain than it is now. People "husbanded" powder, by using lighter charges in their guns, and relying on their tracking skills to find game they had shot but did not drop dead within sight of the shooter. The fact that a particular ROT shows up more in guns surviving in museums may only reflect that reality, and not that this was the "Best" ROT.
There are "games" we play where different kinds of barrels are used. For instance, it was After the Civil War, and well into the cartridge era, that long bullets, and various ROT barrels, and groove depths were tried out for long range shooting contests. Some of this work had been done before the war, as evidenced by some of the Sniper rifles used by both sides during that sad affair. But, most people had never seen such rifles before the war ended. As better rifling machines were made, and sold at prices people could afford, better barrels were made, and sold to the public.
With a Round Ball, and any ROT, a gun can shoot accurately PROVIDED that the patch is thick enough to seal the bore, and hold onto the ball to impart the fast ROT spin.
With conicals, You can go a bit slower than 1:48 and still stabilize some longer bullets. But, either paper patched bullets in shallow grooves must be used, nor a good base wad to seal the gases behind the bullet needs to be used so that you can get the velocity needed to stabilize these longer bullets in flight.
Round balls in slow twist barrels may require more powder-- MAY--- not "must have". It all depends on barrel harmonics.
The traditional teaching about slow ROT barrels is that they are more "forgiving" of changes in the amount of powder used, and will still shoot their balls to the same POI. That is NOT THE SAME THING as NEEDING more powder to achieve the same degree of tight groups( One definition of accuracy), as you may get from a barrel with a slightly faster( ie. 1:48") ROT.
Run The Greenhill formula( which is intended for Conicals, but will still give you a good idea of what works well for round balls) and you find that 1:48 ROT is terrific for .32, .36, .40 and .45 caliber guns. When you get to .50 and .54, that 1:66 begins to be less " finicky" with powder charges. Go on up to .58, .66, .69, and .72 calibers, and you find an even slower ROT will probably be more forgiving on variations of powder charges. ( ie. 1:70, 1:72, etc.)
IN those same museums, you will find guns with a ROT being 1:100" or slower! You have to question the quality control on the powders and the other components used in those guns.
I believe, from my own testing, that Roundball is right about the 1:48 ROT. It shoots ROUND BALLS as accurately as any slower ROT will, Provided, the right charge is found. But, that is true of all barrels. Barrel harmonics control that degree of accuracy. You do seem to have to be fairly exact in measuring that powder charge, to get small groups, compared to shooting RBs out of slower ROT barrels.
When it comes to shooting conicals, you are talking different considerations, shallower grooves, and the Greenhill formula dictates the use of even faster Rates of Twist in the rifling as the length of the bullet increases, to stabilize the bullets in air.
If you like FAST ROTs, the current US military battle rifle now comes rifled with a 1:7" ROT. Its also spitting long, 70+ grain bullets out at over 3000 fps. MV. :hatsoff:
50 years ago, when that gun was first introduced, it had a 1:10" ROT, shooting a shorter, 55 grain bullet at over 3300 fps. MV.