You make a fair point, Fly-by-Night, and at least it contributes to the discussion, unlike another comment!
In fact I did not really qualify my position very well, but it takes a long time to include all the ifs and buts , and it would all get a bit long-winded. The main point is, if there IS a problem with Airproxs at Coventry, one should carry out a risk assessment.
The comfort and convenience of pilots is not necessarily the most important thing, although I agree that it IS important. One has to analyse the risk. Are the long approaches a factor? (They may not be) Is it the intermediate approach that is causing the problem (it rather looks like it), could the aircraft be better protected by remaining within CA for longer? (yes) If so, how could they do this?
I still maintain that remaining as close as is reasonable to an airport located in Class G airspace is wise, for the reasons mentioned, and that a policy to try and achieve this in visual conditions would not be a bad idea. (All the airproxes would seem to have occured in visual conditions.)
At Coventry there is a lot of CA above and to the north of the airport. As many B737s (although there are only a few arrivals each day) come from the south or south east, could we not look at (for example) feeding them into the same approach path for BHX rwy 33, as they are on BHX's frequency anyway, and then pick them off over Leamington Spa, ie. before reaching HON, and take them RIGHT-HAND downwind for Coventry's 23, where they would remain in CA until they turned on to finals at (ideally) 4 miles. This would not cause BHX any problems. Bear in mind that aircraft departing CVT have to await release from Birmingham and effectively slot into their departures, so I don't see why they can't do the same on arrival, especially as we have been talking only 4 or 6 arrivals a day this winter. Different arrangements would apply when 15 was in use at BHX.
I am not PROPOSING this, per se, just saying that if the matter was thought through - by risk analysis - there just might be a better way found than the tradional very long finals in Class G (which requires one to fly to the 10nm intercept point through Class G as well as come back in on the approach path in Class G - maybe 20 miles in Class G for each approach to 23).
A good point was make by one poster who said that the only reason TFly and others do tightish visual circuits in Greece and other places is because of the mountains. Obviously it is done for safety reasons: it is safer to not fly close to mountains if one can help it. It is also safer, I suggest, to keep out of Class G as much as possible for the very same reason, and seeing Coventry's CA won't be around for a year or so, it wouldn't do any harm for Birmingham ATS, Coventry ATS, NATS and BY to sit round a table and see if there is a "better way".