if it was inside a building, it could have been picked up by the wiring and coupled into the TV by the rather large capacitances of the bridge rectifier. there might be enough energy there to cause some "soft" failures, such as degrading the transistor junctions in any microprocessors, and other chips in the TV, maybe even enough to take out the micro, since it's running off the standby supply, and any soft failures there may turn into hard failures after a few minutes to an hour of operating in a degraded condition, or a lockup condition (which is thermal disaster for small low voltage micros). most oven magnetrons by themselves have an isotropic antenna sticking out. if they're still mounted in their wave guide assembly, there's very little loss until the waves exit to free space, after that the inverse square law takes over. same with directional antennas, all of the gain is in the directional assembly. once the wave get's into free space it spreads out in the same manner as any other wave. in the near field of the antenna, the waves that would have been radiated in all directions are reflected in phase with the direct wave. the antenna has gain in this direction, and all of the direct and reflected energy reinforce each other, so for instance instead of a 100kV/meter measurement 1 wavelength from the antenna, it's now 400kV or 800kV, but once in free space, it's still going to diminish according to the inverse square of the distance, just it's travelling generally in one direction rather than all directions. it's still spreading out. even a laser does the same thing. this is why directional antennas and even lasers are specified according to their beamwidth in degrees or radians a 0.1degree beamwidth of a laser still means that if you measure the lumens per square centimeter of a laser at say, one meter, when you measure again at 2 meters, it will be one quarter what it was at 1 meter and 1/9th at 3 meters. all of the energy of the beam is still there, just spread out more. with a gas laser for instance, the beam width is roughly determined by the ratio of the mirror diameter and the length of the "cavity". even collimation of the beam has it's limits. you will often see fresnel rings on a microwve antenna. the purpose is to collimate the beam. there will still be spreading of the beam, since the beam still contains off-axis components