The Radio-Locator maps are quite accurate, if one understands what they are and aren't supposed to be. If R-L says 12~Thirty puts "x" amount of signal over Gilbert at night (2.5 mV/m, in this case), KOY probably really does put that much signal over Gilbert at night.
But...and this is a huge "but"...the R-L maps don't include several other factors that make a huge difference in one's ability to receive a signal. In particular, they don't show you what else is happening on that frequency, and that makes an enormous difference at night, when the skywave picks up on AM.
During the day, after all, KOY (and pretty much every other AM in the valley) has its frequency entirely to itself. Phoenix has the advantage of being a fairly isolated city, geographically; you have to go hundreds of miles to find any other co-channel AMs to most Phoenix signals. The nearest 1230s, if I'm not mistaken, are in Kingman, Safford, Winslow and Las Vegas. During the day, the poor ground conductivity of Arizona means those signals are effectively gone long before they can ever get to Phoenix, so KOY's own signal can go until it peters out of its own accord.
But at night, when the skywave kicks up, all of the hundreds of stations on 1230 are each contributing their 1000 watts to the ambient noise level on that "graveyard" channel. If KOY went off the air, you wouldn't be able to pick out much of any individual station, but they're all in there at low levels, creating a huge noise floor that KOY has to overcome to be heard.
It's possible to calculate just how much signal KOY (or any other station) has to deliver to your receiver at night to be clearly heard over all the incoming skywave interference. This is called the "nighttime interference-free" (NIF) level, and it's an important tool engineers and station owners use to determine what their real-world coverage is at night.
So how bad is it for Kay-Oh-Why? That red line on the Radio-Locator map is the 2.5 mV/m contour, which looks nice on a map but is actually a rather weaker signal than most engineers believe is needed to be clearly heard even by day over the din of all the electrical noise out there now. If Sr. Gleason chimes in here, he'll tell you that an urban AM signal now needs somewhere between 10-15 mV/m to punch through the noise and get listeners...so imagine that inner red contour reduced by half or so to get the real-world daytime useful signal for KOY.
But at night, with all the skywave noise on its "graveyard" channel, 1230 has an even harder task: it has to deliver something in the neighborhood of 30-40 mV/m to shout over the din. That contour, if R-L showed it, would about encompass downtown Phoenix and might get as far as the airport, but that's about it.
Now 5~Fifty, that's a different story! There are far fewer stations operating on 550 at night, and because the Phoenix allocation on that frequency is so old, most of the other 550s in the west (Bakersfield, Craig CO, Corvallis OR and so on) have to use directional antennas to reduce the amount of skywave they send toward Phoenix.
So KFYI not only puts out more signal on its own at night than KOY, but it also has much less incoming interference to overcome. I don't have KFYI's NIF number immediately at hand, but it shouldn't need much more than 2 or 3 mV/m to overcome what minimal nighttime interference comes in via skywave - so that red line on the 550 Radio-Locator map should be just about right in the real world.
Hope that helps...class dismissed!