My (utterly unsubstantiated) pet theory on the development of the torpedo stern is as follows: As the nascent inboard gasoline engines grew more powerful, the cruiser-stern (double-ended) small planing & semi-planing boat designs of the day had a tendancy to squat as the boat exceeded the 1.34 (approx) speed-to-length (S/L) ratio threshold. Design theory of the time held (accurately) that S/L ratios higher than 1.34 were achievable with high length-to-beam (L/B) hullforms, and that a flattish bottom with hard bilges and straight buttocks allowed even higher speeds. It would be reasonable to think that a redistribution of weight at the stern would alleviate the squatting, so the cruiser stern was drawn reversed, making a torpedo stern. This has the effect of maintaining a high L/B ratio, provided adequate coaming height to keep water out and people in, and moved the weight of the stern structure forward. Nice plan, too bad it didn't work much over a S/L ratio of about 2.5. Later developments added the "sponsons" at the stern, as shown in the Stancraft boat in Donwest's photo, to provide reserve buoyancy underway. These quickly became incorporated into the hull bottom as an extension of the bottom planking, and we were well on our way to developing the square stern. Developments in MTB's chronicled by Lindsay Lord during WWII and later break-through hullforms by Ray Hunt sealed the fate of torpedo sterns on fast boats. But they live on in some quarters 'cause they're so darned pretty. That's my story and I'm sticking to it (at least until it fails utterly under withering assault)!
Hope for the best, but plan for the worst.