Sue,

That's an astute observation. The real answer is that it depends on the stove, the fuel, and the conditions. To really get it right takes some testing with monitoring equipment.

However, there are a few generalizations we can make:
1. You do want some clearance between the side of the pot and the windscreen. Normally a finger's width or so is fine. The ideal situation is where warm air is flowing up the sides of the pot between the windscreen and the pot. This warm air flow adds (some) heat to the pot and is referred to as a "flue effect." Generally, you do not want the windscreen to be tight up against the side of the pot.

2. You do want air to enter at the bottom of the windscreen such that the flame isn't starved. Most windscreens aren't so air tight that the flame gets starved, but it's actually not a bad idea to have a row of small holes punched in the windscreen along the lower edge of the screen. You can achieve the same effect by setting the windscreen on top of some small rocks. The thing to avoid is "sealing" the windscreen at the base by piling up dirt or snow or something. In the case of upright canister stoves, setting the windscreen (if it's the self standing kind) up on rocks has an additional benefit: Cool air entering at the base of the windscreen can help prevent canister overheating.

The above generalities will have to be tailored to your stove and your conditions. On a really windy day, you might have to really close up that windscreen tight and maybe even seal it with dirt along the base on the windward side. On a fairly still day, you might need to open up the windscreen a bit so you have two fingers width clearance around the pot and also set the windscreen on small rocks. Observing the flame can be helpful here. If you can see the flame getting buffeted by the wind, then you probably need to do more to exclude wind. If on the other hand the color of the flame looks as though the flame is starved, you'll need to open things up a bit.

HJ