Capot
I'm sure you know all this - I'm expanding for those who may not have come across all these concepts:
Most modern airline SOPs are set up for the use of derates, usually the fine-grained control you get using assumed temperature. This is for engine life, maintenance issues, costs, noise, etc. I don't think they are mandatory anywhere - you can always use more thrust if you like - but I would hazard that the majority of pilots flying commercial jets today 'buy in' to the concept.
When you are preparing to take off from any runway you get the 'figures' (I hope

), either manually or through some sort of computerised performance calculation. Depending on who you work for, 'intersection data' may be available electronically or published in a manual.
What has been worked out for you, when you obtain a derate for a particular takeoff position, is the
lowest thrust setting that satisfies many criteria, including obstacle clearance, EO climb limits, WAT limits, Vmcg & Vmca problems, close-in turns, etc. As you increase the amount of runway available, the required thrust will reduce until you are limited by one or more of the above. On a 'normal' sized runway, if there are not too many significant obstacles or other extreme environmental conditions, the outcome of the calculation will often be to make ASDR = TODA, as that becomes the limiting factor.
*** My point is that if you are using full length and the derate from the last paragraph, you are not in any significant way increasing the margin for a V1 reject over someone who did the same calculations using an intersection and came up with less of a derate. ***
An interesting effect of this is that you only start to see a difference when using long runways or light weights: going for 2,500m over 2,200m will probably not affect the V1 case but taking 4,000m over 3,000m most likely will. Somewhat counter-intuitive as many think it's the other way round.
"why not maximise safety and use the same - higher - power setting you would have used from the intersection, but taking the full length?"
Why not *really* maximise safety and always use
full power? What's so special about the thrust needed from a random intersection? It's a bit like adding more fuel "for Mum": where do you stop? I think this is a key issue - either you have confidence in the aircraft performance or not; if it doesn't do what the book says, there should be an ADD with a performance decrement to apply or it should go back in the hangar to be fixed.
From a practical point-of-view, if I think there's a possibility of an intersection departure, then I'll set up for that then keep that configuration if I get sent for a longer intersection or the full length. This is more from a distraction management perspective than trying to increase margins. If you are derated, then you'll get a margin from the positive effects of real air density vs. assumed, anyway.
I'm not sure I really buy that. Most of the unknown variables are still present in the calculation.
Yes, they are but that's something we have to live with every day we go to work. If you think the aircraft is heavier than it should be, get it weighed or use more thrust, wherever you're taking off from...
(I'm hoping that JT, mutt or one of the other performance experts will step in and help out here, or at least correct my mistakes! C'mon guys, where are you?!?)