There are a number of scenarios regarding line twist which depend on
how your line was loaded onto the reel. Lets take them one by one.
You load the reel of the edge of the bulk spool: This puts the twist
on the reel. When you cast out the line is untwisted. When you wind it
back with no significant load, if the reel roller doesn't impart any
twist of its own (poor roller design is largely a thing of the past)
the same twist goes back on the reel and everything is as it was. If
the load is significant the line has a tendency to flatten slightly
making it more difficult for the twist to go over the roller and
back on to the spool.
This is why pump and wind is less likely to cause line twist than
winding in. Imagine the line as a piece of thin narrow tape. The twist
is going to have a really hard time passing over the roller. What
happens is that the tape keeps twisting between the roller and the
first ring until it's so twisted it can't do anything but go over the
roller and you have a section of the tape/line which is twisted to oblivion
on the reel. Next time you cast out, depending where that section was
it has the potential to cause a wind loop. It might or might not clear
up after a few casts.
Does pulling line off the drag cause twist in this scenario? It
doesn't actually twist the line. What it doesn't do is unwind the
twist that would have unwound had the line come off the reel as it
would when cast out. As the line didn't untwist it now means there's
twist in the cast out length. When you wind in it won't be 6 twists
(or whatever your reel naturally does) per meter, it will be 6 and a
bit twists per meter. Which may or may not be spread over the whole
length. Each time you pull off the drag it has the potential to build
up a bit more twist in the cast out length. But it isn't massive. It's
only for the length which is being pulled off which if it's 2 meters
in 50 isn't going to cause much of a problem. 50M~=300 twists as
opposed to ~312 twists. The closer in you are the bigger the potential
problem as it's the same 'extra' twist in a much shorter length. It
should sort itself out on the next cast although the extra twist will
still be there, it will just be spread over the whole length. Just be
careful! Backwinding prevents this happening.
There is a slight potential under load for the twist to build up under
pressure much as above between where the line leaves the spool and
meets the roller. This is such a short length I suspect any potential
issues are insignificant.
Winding against the drag definitely does put extra twist in the line.
Again, it's much more significant at close range.
You load the line off the face of the bulk spool: Assuming you spool
off the bulk in the right direction. The principle behind this method
is that it counteracts the twist induced by the reel resulting in twist
free line on the reel. There is however a problem with that in that
the reel spool and bulk spool are never the same diameter. Even if
they started at the same diameter it still wouldn't be twist free as
the reel spool diameter increases as the bulk spool decreases so we
end up with a bit of a mish mash of where and how much line twist
there is when cast out. It could, if the bulk spool was twice the
diameter result in half the twist being on the reel and half (in the
reverse direction) when cast out but for a 50mm diameter reel spool it
would require a 100mm bulk spool which AFAIK doesn't exist. For a
Shimano 4000 sized reel (~Daiwa 3000) which is about 50mm full and a
fairly normal bulk spool we'd have about 4t/M on the reel and 2t/M in
the cast out line. 2t/M isn't generally likely to cause problems so
long as it stays that way. I don't currently fish the rivers but I
could see there being a potential problem with smaller diameter reels
and floating lines. A 2000 sized reel might have a full spool diameter
of 35mm and cast out twist of ~4.5t/M which may or may not coil up on
the surface depending out how much memory the line has.
Under load you're still dependent on the twist going over the roller
only this time it untwists at 2t/M then twists at 4t/M.
If you could arrange for the reel and bulk spools to be the same
diameter there might be an advantage, assuming all the twist being on
the cast out line isn't a problem. When winding back under load you're
untwisting the line so the twist doesn't have to pass over the roller.
Loading line off the face of the bulk spool might help with poorly designed rollers.
My opinion is that loading off the face of the bulk spool is something
that was done in an attempt to mitigate the problems caused by poor
reel roller design. Shimano® explicitly say to load off the edge of
the bulk spool. Daiwa® don't explicitly recommend any particular
method but their bulk spool loader has the capability to adjust the
tension which wouldn't be necessary when loading off the face although
it could be used that way.
Food for thought! Of all the differing opinions you'll find on the net
there's one thing just about everyone agrees on. In order to remove line
twist, take the line out onto some grass and wind it back in. Isn't this
the same as loading of the edge of the bulk spool?
This isn't a recommendation, it's just the way I load fixed spool reels.
Make your own mind up and do whatever works for you. Soak the bulk spool
in water with a little washing up liquid overnight. Don't overdo it with
the washing up liquid, it could make the line sticky. Some WU liquids are
worse than others for this. I use Fairy. Put a mandrel through the spool
and place it on top of a pan full to the brim with warm water with a
little washing up liquid in it. The bottom half of the bulk spool should be
submerged. Thread the line through the butt ring of a rod and load the
line off the edge of the bulk spool with moderate tension on the line. Water
will fly about everywhere so make sure you're somewhere it doesn't matter!