"Effectively inside the horizon" isn't a good way to think about GR. Observers have reference frames, and can only observe according to their own clocks and rulers.
If you (a distant observer) watched your blinking probe approach a black hole, you would see its blinks slow and red shift as the probe approaches the horizon. You cannot see it enter the horizon, as light cannot escape from inside the horizon. There would be a last blink, as the red shift approaches infinity.
If the event horizon were to increase in radius, the effect would be to further slow and red shift the blinks.
Can't you claim, that if probe reached R+0.5m (which you could potentially see), and then subsequently BH grew to R+1m, that probe's last blink happened in the area, that later became BH?
If you (a distant observer) watched your blinking probe approach a black hole, you would see its blinks slow and red shift as the probe approaches the horizon. You cannot see it enter the horizon, as light cannot escape from inside the horizon. There would be a last blink, as the red shift approaches infinity.
If the event horizon were to increase in radius, the effect would be to further slow and red shift the blinks.