I prefer deploying everything out the nosecone, since the rocket body is stronger. I also consider it more reliable. If the drogue and main parachutes are sized such that the nosecone descent rate is slightly slower than the rocket body descent rate, it is extremely reliable.
I've included a quick picture of the recovery concept, for those who have not used deployment bags and deployed everything out the top. This is the simplest version.
Note, I'm using CO2 and a tether, so I don't have to worry about black powder blowing holes in chutes or damaging lines.
The nosecone is popped off via CO2 system, and the drogue is mostly stuffed into the nosecone coupler, in a way that it falls out effortlessly. The 2' drogue keeps the rocket strung out vertically during descent, or at least keeps the nosecone above the rocket body.
The nosecone is stuck right under the drogue, and the 1/4" Kevlar drogue line goes right beside the nosecone down to the tether's releasable eyepin.
The drogue line has an attachment point at the top of the deployment bag. This keeps the dbag pulled down into the top of the rocket body tube until the main is to be deployed. The deployment bag is taking none of the drogue loading. That is important.
The 5/8" tubular nylon main line is neatly zig-zag coiled in a layer right on top the upper electronics bay bulkhead, in that cup-like area on this rocket. The dbag is squished right on top of that, held down by the bottom section of the drogue line.
When the tether is released, the drogue line is now pulling only on the top of the deployment bag. The rocket falls away under the bag, initially pulling out the main line and then the main's shroud lines, and then the canopy itself. Because of the rocket falling away underneath, it is all pulled out in a vertical string, ideally oriented. The main then opens.
This is why it is important to have the drogue sized such that the nosecone descent rate under drogue is lower than the free-fall descent rate of the rest of the rocket. Otherwise there won't be any pull. Of course this is easily achieved, unless one has a very heavy nosecone with a small drogue, and a lightweight rocket body.
Ideally you string the rocket fairly vertical under the drogue, but still bring it down pretty quickly. On a larger rocket I did this with a descent rate of over 100fps routinely. Bringing it down fast helps to minimize drift on higher flights. This one won't be that fast. My drogue is perhaps slightly larger than necessary.
Once the tether has been released, the drogue chute is the nosecone's main. And the main for the rocket body is the bigger main. They come down separately. In practice they tend to drift apart a bit. That can make finding both parts a bit more... fun. So there is an optional very light line, more like a thread, that can connect the two descending units together.
I have not yet decided whether I'll thread the two descent units together. I've tended not to do so in the past. But it is an option.
If you want to see something more complex using a drogue and then separate main chutes for the nosecone and the rocket body, check out:
https://www.rocketryforum.com/threads/sprite-6-and-a-baby-o.37382/
Gerald