Of only really having a year and halfs experience of following wild wild dogs, albeit an intense bunch of sightings of which I recorded much info, and subsequently got reports since, giving about 3 years of info for the pack that frequents the area around Skukuza northwards I would make the following conclusions as to what limited this pack's size...
Firstly I think the wet/dry cycle plays the biggest role as it obviously determines weakness of game (thereby easiness with which to catch) and then ofcourse thicker bush would allow for ambush by lions, of which 3 deaths due to this was officially recorded for this pack.
In other words in drier years the pack would get bigger as food wouldnt be limiting and predator avoidance would be maximised.
In South Africa, historically packs have been upto even a hundred or more...in the Sabi Sands a pack was about 60 animals in the mid 20th century. It was in the 19th century, that the early hunters recorded the even larger packs. I am sure that food availablility has a major role in determining pack size and would think that the great SPringbok herds as well as larger herds of other species would have been one of the reasons why packs could get so large.
I have no scientific evidence, but from what I have seen with regard to the amount of meat this particular pack that I viewed, consumed, I would say that 'bushbuck' was there main food source (killed about 50 in 2 months over the denning period along the sand river), and that this prey specie (bushbuck) and adopted hunting strategy only allowed for an optimum size of about 12 dogs per pack. Just to repeat, this is not for an official PHD etc it was a small experiment based on many observations and simple statistics.
Ofcourse, territory size will be important and one wonders if although as large as KNP is, if perhaps the dogs home ranges are still negatively being affected.
Egilio seems to have a lot of interesting info and from what I have read of his and the little I know about the Luangwa it is/would be of no surprise if the packs were larger in that part of Africa...availability of food-perhaps more open savanna these days ETC??
I am also sure that the bush encroachment has negatively impacted this species in the last 15 years. It is scary how quickly many partially open areas have been/are getting clogged up with Combretums and Terminalias in the Park...I am sure a big factor leading to lower carrying capacities of this animals.
Anyway further thoughts of mine...
What defines a wild dog hunt??
In my experience of 30 wild dog hunts upwards for a single pack that I got to know well, taking each individual animal singled out and chased vs success of catching them I would say that this species doesnt have as much of a success rate as is claimed. In fact I would put it down less than 50% of the time for this pack. In the average hunt this pack would come across a few herds impala, some kudu, a couple of duikers and a steenbok or two. You could follow them for 2 hours where they would go about chasing these animals madly making many unsuccessful attempts...then suddenly they would hit the thick reeds of the river bed and pull down 2 bushbuck in seconds. So although they had been running madly on one hunt session, the number of attempts was large, and probably almost as 'bad' as that of lions.
Then the next day the pack would bring down 3 impala rams in an hour out of far fewer attempts. I just think that its hard to monitor hunts properly in the thick savanna.