1. Ratio has to be 100% always for all trades. If its 100% for all trades and say, 125% for one particular trade, you are sending 25% more traffic to that trade. However the performance of all trades varies from time to time, because of thumbs rotation. That's why it is viser to leave it on 100% for all and let the algo dynamically decide which trade deserves the most.
2. Caps: depends on the skim setting, better to say: depends on how much (return) and what quality (productivity) traffic you are sending out. If your out traffic is super good quality, because you have friendly (high) skim, say, 70%, than you typically want a low 24 hours cap value, since your traffic is very good quality AND as such it is little quantity! So you don't want to waste it, in that case even a 125% daily cap is good. However if your skim is low, you are sending lots to trades but average quality, you want the caps to be high. Even 500% for hourly and daily is ok if your skim is say: 25%.
It depends on the site, yes. I love that sentence. It offers no explanation. ;-)
Basically if surfers
a, like the domain name (easy to remember, cool, unique)
b, like the design, the layout
c, love the content
d, you have trades that have more or less the same content (paysite)
e, you have a strong server
etc...
than your tgp will be a success
If anything from above is not met, you can have the best trade script and the best settings, your tgp will be a failure.
by which I am trying to say that MANY factors have to be good for success.
and of course: your rotator script and settings is very important since this is the 1st level in the whole pipeline.