I'm somewhat surprised at the idea of gas leakage around the O/P wadding causing a doughnut pattern rather than an irregular one, but I can see leakage THROUGH the wadding doing it. Anyway, concerning too-heavy wadding causing doughnuts, two of the possible contributing factors I've seen put forward are:
- If the gas pressure at the muzzle is sufficient, the thick wad column is still being accelerated for that 1/2"-3/4" after the shot has cleared the confines of the barrel, until the rear of the wads clear the muzzle. This would go along with the observation that it's easier to get tighter patterns with longer barrels in the absence of choke, shot cups, etc., than it is in shorter ones.
- Even if the wads are not being significantly accelerated at the muzzle, the powder gasses combined with aerodynamic drafting behind the still-compact shot charge could still cause the wads to bump the rear of the shot charge.
Whichever it is, it's consistent with the spark photographs of pre-shotcup cartridge loads fired from cylinder/unchoked barrels, with the rearmost pellets in the charge spreading laterally very close to the muzzle.
I cannot personally attest to the reasons, but I can to the effect in the 29" barrels of my double, with blown or inconsistent patterns using lube-saturated fibre cushion wads of 1/2" and even 1/4" thickness, and possibly with the 1/4" hard O/P wads combined with cushions. Lighter 1/4"± cushions (the rough bores NEED adequate lube) and substituting multiple O/S cards for the O/P one have made a significant difference in pattern consistency.
Back to powder gas blow-by, my 16ga has somewhat rough bores and I've found I need to use at least 3 and preferably 4 overshot cards to seal reliably (brown burn streaks on recovered paper shot protectors are very distinctive compared to black/grey fouling smears). This is using firm .020-.025" thick commercial or home-made cards at least .015" larger than bore diameter; with thinner materials like milk cartons, it is 4 minimum and 5 preferred. Softer materials like cushion wads, packed tow or bark fiber, etc. appear to need a minimum thickness of something like 1.5x the bore diameter to seal, depending on the material. Materials with intermediate hardness, like hard felt, appear to have thickness requirements in between the two, as one would expect. I haven't seen anyone work out what the minimums are when using combinations of materials to seal over the powder, e.g 1 O/S card + 1/2 of a cushion wad.
Regads,
Joel