I thought the main insurance a dual spring gives is it saves your engine when it does break. Not that it lasts longer. At .598 lift, whatever spring I use there is a chance of it breaking. But with a dual spring, even if it breaks it won't hurt any thing else.

 

Duals will not rev as high without additional spring tension over 918 and duals use friction between inner and outer to control harmonics and that shortens springlife as well. I dont see dual spring saving valve unless breakage happens at idle, this is less likely to happen at idle than at shift point rpm.
Read up on what im preaching here and you will see dual is old tech. and more likely to break.

 

https://ls1tech.com/forums/generation-iii-internal-engine/165867-c918-test-pressures-4800miles-tr224.html Link to my old post after checking my 918s @ TR224. I've since switched to the Patriot Gold's and will be testing them when I hit 10K miles. Some have stated the added weight on the valve of my Yella Terra rockers caused the springs to lose pressure quicker than with stock rockers. We'll find out for sure when I check the Patriots.

 

I have the same Tavia tester as you and it will vary 10psi if you test the same spring 3 times in a row! Seriously you would need to use Rimac tester to validate your springs tension loss. I agree Y.T. add weight on valve and will use up spring . Next 130lbs is overkill for 6500 shifts, LS6 motor uses 90lbs with similar lobe ramp rates to XE Comp profiles, i know LS6 has lighter valves but 918 can be used to 7500rpm.All springs drop pressure when broken in, even at 118lbs. spinning 6500 is no sweat.

 

I've got just shy of 30K miles beatin' on the car daily and seein' the shift light, 6700 rpm's, weekly. Running a Crane Valvetrain with duals. I'll change the springs this summer maybe.