Mr. Auslander can answer question one for you. I believe it's his trademark for his initial patent counseling which attempts to determine whether your invention is likely to bring in enough revenue to justify getting a patent in the first place. I honestly don't know how he does it since it requires as much market knowledge as it does legal and technical knowledge, not to mention thorough knowledge of the applicable prior art. As far as I know, he hasn't addressed those concerns here but might do so to a prospective client in private communication.
As for the one in 10,000 number, it's not true. Don't believe it. I've challenged everyone I've heard quote that number to come up with authoritative support for that number and no one has even responded -- even with lame, amateurish reasoning, let alone a real source of authority. It's a myth at best and possibly even a lie. Don't believe it.
The problem I have with the number is that my experience doesn't jive with that number -- most of my clients make their money back. However, you should understand that most of my clients are companies that make and sell their patented products and services. Evaluating the worth of a patent in this instance is quite difficult. For example, I have a client who had a portfolio of patents and threatened infringers with law suits. Most of them stopped infringing activity. What's that worth? It depends on how much inventory those parties could have made and sold at undercutting prices. Since it was stopped, we don't know what the value was.
So, how do you measure the value of a patent that others quietly avoid? You don't see licensing revenue. You may not even know that the competitors are shying away from competing. Yet you see a real benefit in the marketplace.
I think what the 1:10,000 number is supposed to reflect is that it's extremely difficult for an individual to get money from a manufacturer/reseller for little more than an idea -- patented or otherwise. I still haven't seen any authority for the number itself, but you should be very aware of the difficulty of making money in this route. While I think the number is a bit of hyperbole, you should not underestimate the difficulty in taking this route and the number is meant to illustrate that high level of difficulty.
Having said that, I'd still like to see some authority for the number. And citing some newspaper/e-zine article in which someone repeats the number is not authoritative.
As for your hypothetical, I can safely say that adding two numbers together is not patentable (not novel). However, let's assume I've misunderstood that part of the hypothetical and assume that adding A to B is both novel and non-obvious and that you were able to get a patent on it. Independent creation is no defense to patent infringement. Neither is reverse engineering. If your competitor does what your claims describe, she infringes -- however she came to do that.
Lastly, I'll just offer a different perspective on confidentiality agreements (CDA/NDA) signed by attorneys (patent or otherwise). In my 13 years of practicing, I have not signed a single CDA. As drilled into my head by my ethics professor, comingling client and firm funds and violation of client confidence result in "VIRTUALLY AUTOMATIC DISBARMENT." That's a far more stringent punishment than could be meted out for violating a contract (in my opinion).
Is there any harm in signing those agreements? No, of course not. However, I find my work more rewarding when my client fully appreciates the nature and value of what I do. Being asked to sign a CDA is a sign that the client doesn't understand the nature of legal services. It's not bad in and of itself; it just tells me I'm going to have to justify my work, decisions, and perhaps my professional existence. Insisting on a CDA after the above explanation is, in a way, ignoring my legal advice. At the very least, that suggests a problematic legal relationship. One of my fears is the spector of a malpractice suit when a client ignores my advice and bad things happen.
For what it's worth, these concerns just don't exist with my corporate clients. The principles are used to dealing with attorneys and all of this is understood. That lets me focus on doing good work and worrying less about misunderstandings.
Now, does that mean I'll refuse to sign CDAs in the future? No. But be prepared to hear what I wrote above. If I lose potential clients as a result, so be it.
One last qualification: I have no idea how the ethics rules by which attorneys are bound apply to patent agents. Perhaps some of the agents out there can answer that.
Regards.