I wish is was as simple as it appears here. Alas, it is not.
While the frequency of operation can be important, so is the size of the coil, and the amount of power applied to it. And obviously, the receiver is important too. All of these, in combination, determine how deep, and how sensitive any detector is to any given metallic mass. Change any one, and you effect the other parameters too, especially battery life.
The actual makeup of the the metallic mass (ferrous, non-ferrous), its shape, its size, all have an effect as well. A good example is a ring laying parallel to the detector's coil. Since it's a closed circuit, the detection depth is much greater than it would be if it had a gap in its circular shape. Further, if you rotate the ring perpendicular to the coil, the detection depth is even less. If you understand how electromagnetic fields interact with one another, this is an obvious scenario.
The question is then, is frequency all that important, all else being equal? Think about this. The old BFO machines typically operated between 200 kHz, and 500 kHz. Yet, under nominal conditions, they'd detect a quarter 6 to 8 inches deep using an 8 inch coil. This isn't much different than today's TR machines. Before you trash this thought, it's important to remember the other parameters involved.
I'm often amused with the mythical hype given to some specific designs as if they're the panacean answer to the holy grail of metal detecting. All too often as a result of the placebo effect, the mythical belief turns into pseudo fact. This scenario isn't limited to metal detecting either. They're just as much of it, if not more, when it comes to amateur radio antennas; a very similar product from a technical aspect.
The bottom line is, there isn't any free lunch, and there isn't just one, all-encompassing, universal parameter.
Alan Applegate, K