My thought has been that draft really doesn't play as big a part in this issue as it gets credit for.
Think of it this way...If/when the rapid production of (excessive) potentionally flammable gases which are UNable to be ignited for lack of oxygen and/or ignition source ...and these same gases continue being drawn through double pass internals of boiler to next be expelled out of the breech into stove pipe then next passing by the baro damper and finally heading upward through the chimney to the 'out of doors'.
Now... at end of firing cycle when flapper opens you now have the oxygen...then if/when fuel to air ratio gets satisfied it only then depends on how long of the delay that an Ignition Source might happen,
Question is...What quanity of those gases and what distance did those gases travel before it ignited into explosion?
This senerio is not slowed or stopped by increasing the draft but it will sure add a good amount of the defient oxygen for an explosion. And it is a way more violant of an explosion if gases ignite all the way from the boiler and into the chimney at once .