When you say “all other factors being constant” in this case, you are putting yourself in “assume a spherical cow” and “ignore gravity” type of situation as usually those all other things are nowhere near constant. In fact, it’s those other things that usually determine the frequency being used.
Increasing the wavelength (decreasing the frequency) decreases your resolution and significantly increases the amount of clutter you get in your returns. Usually you end up with so much clutter that the radar is practically useless. One notable exception to this is over the horizon radar, and the main reason this works really well is that the frequencies they use happen to reflect really well off of the ionosphere (one of those other things that isn’t at all equal), which is what gives them their over the horizon capability. Sure, the resolution of over the horizon radar is crap, but hey, it works over the horizon, giving you a much greater early warning capability (and a bunch of false positives that you have to sort out, but hey, you take the good with the bad).
The absorption or reflection of water in the atmosphere also determines which frequencies are more desirable to use. If you are making weather radar, you want to pick frequencies that tend to get reflected really well by clouds and precipitation. If you are making radar to detect enemy planes, you want something that will pass through clouds but reflect off of planes. In either case, you don’t want a frequency that gets absorbed easily by water as your atmosphere attenuation due to water vapor in the air will be horrible. On the other hand, if you are making a very short range radar detection device of some sort (like say a traffic light detector, or an alarm system) you may want to pick one of those frequencies where the atmospheric absorption is high, since you are going such a short distance that the absorption doesn’t matter, and you aren’t likely to get much interference from other radio devices because the frequencies propagate through the air so poorly.
Radio frequencies are also really cluttered with, well, radio. At lower frequencies you’ve got AM and FM radio signals, TV signals, and all sorts of stuff. As people invent new stuff, they generally try to fit it into whatever lowest available frequency they can get, so your lower radio frequencies are extremely cluttered with man-made signals. In fact, the radio spectrum is so heavily cluttered that, on the rare occasion when the FCC decides to auction off frequency bands, the sale price usually ends up being an extremely large sum of money.
There’s also the consideration of size. For the same type of antenna, the antenna scales up with the wavelength. So if you double the wavelength, the same type of antenna design will be twice as large. A half wave dipole antenna, for example, is half the wavelength in size, so obviously if you double the wavelength you double the length of the antenna. If you are making an airborne radar, the size of an antenna that will fit onto the back of an AWACS plane is a lot larger than the size of an antenna that will fit into a small fighter jet (in fact, slight point of trivia, in the plant that I worked in, airborne radar was produced in one building and ground-based radars were produced in another building - the AWACS antenna was built in the ground-based building because of its sheer size).
