This is a pretty good method, but sometimes it won't work. I've cracked several master locks in my life, and I found one for which the above method did not work--I kept getting the wrong 3rd digit, no matter how carefully I looked for the .5's. (Trying several hundred combinations in the process.) I recall another occasion when I gave up after finding two wrong 3rd digits (and trying 200 combinations), and another when my friend told me that the 3rd digit I found was wrong. With Google, I found this method, and it seems to work perfectly:
http://www.angelfire.com/ma4/masterlockcrack/3rd-2.htmlIt gives the same answers as the first method most of the time. Think about it: if the centers of the ranges of the lock are in fact arranged in a group of four xx.5's (e.g. 2.5, 12.5, 22.5, 32.5), a group of four xx.0's, and a group of three xx.5's and an xx.0, which is what the first method predicts, then the second method should give the same 3rd digit.
I know nothing about the internal workings of the master lock, so I can't argue for either method on its own merits, but I do know this:
1. Using the first method, I've failed at least three times, and I don't think I suck at implementing it. Using the second method, I have not failed yet.
2. On that lock for which I, using the first method, failed to find the correct 3rd digit after about five tries, I found the correct 3rd digit with the second method on my first try. This digit was, in fact, one of the digits that I was quite sure was not correct according to the first method.
3. The second method is easier to do and to feel confident about. Instead of seeing what looks like a .25 and trying to decide whether I should call it a 0 or a .5, I just need to see whether this range is the same as that range and these other two ranges.