Yesterday, I was checking out Steroid-Rx.com, and the idea of Anabolic and Androgenic scores came up in this thread. And frankly, all over the Internet, I see people posting tables and lists with various anabolic and androgenic scores (presumably compiled from the profiles I wrote for Steroid.com). These scores are comforting, because they assign a numerical value to the differing properties of various anabolic steroids - but they are almost totally misleading, to be honest. I put them in my first book because I thought that they were important, I put them in my second book because I thought people wanted to see them, and the current profiles I'm writing (for a site-yet-to-be-named) don't have them at all - and for good reason.
Here's why (this is a copy/paste from my answer on Steroid-Rx.com):
In 1950, two scientists named Eisenberg and Gordan, proposed the use of the rat levator ani muscle as an indicator of the anabolic activity of various anabolic steroids. This is the system we still use today.
This muscle is part of the perineal muscle complex that envelops the rectum. It was chosen because it had been found that testosterone propionate stimulated the growth of the perineal complex in infantile rats (and because it was easy for researchers to separate from the rest of the muscles). Simply stated, the levator ani responds well to anabolic steroids, so that's the muscle we look at to gauge their anabolic effects.
In 1953, another scientist came up with the method we now use for the androgenic portion of these ratios, namely weighing the ventral part of the prostate.
Now, this is how we arrive at our anabolic/androgenic ratios...it's all based on the weight of two semi-randomly (and badly) chosen muscles in rodents.
Like the index value, the anabolic or androgenic effects are usually expressed as ratios to other reference steroids, for example, we compare many orals to methyltestosterone. Methyltestosterone has an anabolic rating of 115-150 compared to testosterone, and Anadrol 50's anabolic rating is 320 compared to Methyltestosterone. Halotestin has an anabolic rating of 1900 compared to Methyltestosterone - and as any experienced steroid user knows, you have no chance of gaining much muscle (at all) on Halotestin, and certainly not 5-6x as much as you would gain with an equal amount of Anadrol 50.
So as you can see, although they are comforting to look at, anabolic and androgenic scores tell us precious little about the actual muscle-building properties of various anabolic steroids. This is also why we see do-nothing-prohormones being advertised as a billion times more anabolic than testosterone, when in reality, it provides negligible muscle gains at best (and certainly nothing compared to testosterone).