No cab, baffle, nothing?

You can't evaluate a speaker like that. Of course you're not getting all the volume... or frequency response or efficiency or anything else. That's like revving up a Boss 302 while it's laying out by itself on the ground and wondering why you can't get any speed out of your Mustang.
Put it in a proper cab and listen to it. If you don't have one/can't afford one/whatever, then put your motor back in your Mustang, run down to Lowe's, and buy yourself a 4'x8' sheet of 3/4" MDF. Cut the proper size hole right in the middle of it, screw the speaker in it, and try that. You still won't know how it might sound in a sealed cab, but this test will give you an idea how it would behave with an open or semi-open back.
Letting a speaker hang out in the breeze is only good for making some sort of avant garde sound effect for recording or building an emergency radio from what's left of your airplane after crash landing in the middle of nowhere. Definitely not for speaker evaluation listening tests.
EDIT: After re-reading this, I thought I should explain a little more so you understand why this isn't good. Speakers produce sound waves according to the electrical input signal they receive. You know this part already. The electrical signal we send them is a complicated one comprised of multiple frequencies. Some speakers are better than others at translating these electrical signals into acoustical sound comprised of the same frequencies. All guitar amp speakers, including the Legend 1518, will be capable of producing a very wide (by guitar standards) range of frequencies at least to the extent that we'll classify what we hear as "music" and not just multiple overlapped tones. The way it does this is by pushing the cone outward (air compression) and retracting it back (air rarefaction). This causes the air pressure on your eardrum to vary from the nominal 14.7psi in a corresponding manner to the cone excursion. As a result, you "hear" the speaker's movement. The problem is that a speaker is a two-way device: It produces it's sound from the rear as well as the front.
Because your Legend produces sound from both the front and rear, you can experience cancellation of like frequencies. A "+" value of anything, when in the presence of a "-" value that's equal in scope and magnitude will eliminate each other. Take an empty 8oz glass. Fill it with a positive 7oz of water. Then fill it with a negative 7oz of water (pour it out). You're left with the same empty glass you started with. As your Legend's cone moves outward, it causes a compression of air (positive). At the exact same time and looking at the speaker now from the rear, the cone is also moving inward causing a rarefaction of air (negative). The compressed air, which you should normally hear, sneaks around to the rarefaction if it can. As the cone moves the other way, the opposite happens. It's as if the cone never moved. Or it would be if the world were perfect, but even in the inefficiency of this situation, it's inefficient - not all of the air can move fast enough, so you still hear something. You'll still have some water droplets in your glass after you pour it out. To really kick inefficiency in the can, we need a baffle.
Now, if we could place the speaker into a hole in a wall inside your house, seal it up, then play it - and close the doors to that room so there's no possibility of air sneaking around from front to back - then you could hear the full capability of the speaker. Or we could just build a sealed box, cut a hole in that and do the same thing (ala Marshall 4x12). Either way, it forms an infinite baffle. We don't always desire the absolute isolation an infinite baffle provides. Sometimes a little cancellation with the right circuitry, cabinet materials, speaker(s), etc, can be magic. In that case, we'll simply cut a hole in a flat piece of wood, fold the edges around so it forms a box with not much back to it, and call it an open cabinet. That allows some isolation, but not completely.
Anyway, a speaker with only itself as a baffle just doesn't work.