On using the toaster oven to heat treat the blade. Use a pan or tray of sand to put the blade in the buffer the effects of the elements cycling on and off.
Another thing that could be happening is your soak time, and I admit that I'm talking theory here, is that you are not giving the O1 time enough to put the carbon into solution. I know that you say the you're holding it at 1500 degrees for three minutes but if you are using a forge, I wouldn't bet on it. You could be going higher or lower. Too low and three minutes will not put enough carbon into solution and too high you may get the carbides to break down and release the carbon into solution but you risk growing the grain. You could be getting way too much carbon into solution by getting the steel up to 1550°-1600°, which would be very easy to do in a forge, and, even for a short period, you get so much carbon into solution that you form excess retained austinite formation and you can't achieve the hardness that you want either. Personally, I think that the best thing for steels with much over 84 points of carbon is to austinize in a regulated high temperature oven at under 1485° and stretch the soak out to about 10 minutes.
The vast majority of hobbiest or small time knife makers are at a disadvantage in the we have to operate on a bit of theory and a lot of by guess and by golly. A commercial outfit can make a test run, taking careful notes and send the blade out to be sectioned, polished, and examined microscopically and even a look with an scanning electron microscope to check the crystal formation in the steel. Kevin Cashen is the only person that I know of who can do such analysys and I doubt that even he has a scanning electron microscope in his basement. All that we, the average guys, can do is follow the cook book procedures and maybe throw in a third tempering cycle and hope for the best. It's also a good idea to match the steel that we are using to the equipment that we have at hand. For a gas forge and a toaster oven I think that it indicates a hypoeutectic to mildly hyhpereutectic simple steel that is more forgiving of an austinizing temperature that cannot be held within range. I know that we want to use the super duper higher alloy steels, believe me, I'm the same. And yes, we can produce a good knife with the likes of O1 or 52100 but to do it reliably and consistently we need something with more heat control than a gas forge. All it takes is a little brigher shade of orange for just a tad too long and we have problems and no way to see it.
Doug