Get yourself a piece of cheap cord/string and tie a super eight in it, next cut one of the loops, then pull on the loop as if it were the master-point. You will notice that both loops fail. The BHK won't do this. Another knot that won't fail this way, uses less rope, and is easier to untie after loading than almost any other knot is the Bowline on a bight. >
You've received some good feedback mixed with some quintessential hilarious MP offshoots. To add my own perspective, I'll start with a reminder that a key feature of climbing is not taking risk, but managing it.
You can climb on this anchor many times and probably not die, especially if it's not rubbing over any edges. If you take home one message from all of this forum, take home the BHK - know it's out there and know it would make your system a little bit better (a LOT better if the master point was around anything sharp, and a good thing to do as a default).
You are wise to practice this in your apartment, and I'm sure you are reading up a bit, but remember that it would be especially wise to connect with someone who has a good deal of experience and shares your level of acceptable risk (a hard thing to judge - how is an aspiring outdoors climber to judge the experience, level-headedness, and quality of instruction of an experienced climber... but that's a topic for another MP discussion). Fiendishly devour the classic textbooks - "Freedom of the Hills", "The Mountaineering Handbook", and "Rock Climbing Anchors" to get your self-education, but remember there is no substitute for climbing with an experienced sensei when it comes to becoming fluent with these skills and thinking about all the contingencies as you're building an anchor on the cliff. MP has a lot of wise people who can give you a lot of good feedback, but it also has a lot of people who may not think like mere mortals and would tell you they'd take a whipper on a girth hitch around a dead log, and so I wouldn't recommend a dear friend try crowdsourcing something as high-stakes as anchor-building skills. One big, major step towards managing risk in climbing is finding a sensei you can trust, and climbing with them again, and again, and again.
Otherwise, your anchor looks good. I would take a whipper on that ;-D.
I did just as MTKirk suggested. I tied a piece of twine into a super eight, cut one of the loops, and weighted the other (with my finger). One of the cut strands pulled through the knot and the remaining loop failed.
Edited for tone.
So when you cut one loop of the two "bunny ears", looks like the remaining loop, and how it's threaded through the rest of the 8 is similar in form to the finish of a bowline. Which is secure, generally, if you remember to finish it. Pull tests showing evidence of failure? Personal experience?