Not long ago, I was talking to an individual who came out with a really good algorithm to solve a certain problem. He knows that he has to get a patent for the algorithm, but he is worried that once he puts out the algorithm that others will merely steal it, and not pay him the royalties or license that algorithm from him. In other words he won't get paid for all the years of research in creating it, refining it, and making sure that it works. That's too bad, but he reiterated to me that he has a strategy in mind. He wants to file several patents, some of them bogus, as to fool the individuals who might try to steal his algorithmic strategy.He figures it will cost them a good amount of money and take them years to figure out which one he is actually using, and by then he will have modified it anyway. These mathematical strategies have to do with stock and commodity trading, amongst other things such as sports betting, or trend tracking graphical data. Indeed, I suppose it could be used for anything as long as you had the empirical factual data available in real-time. Perhaps you could even use it to track the weather, although I do realize that today's artificial intelligent supercomputer weather prediction systems are quite robust.Is it possible for him to file bogus algorithmic patents to sideline anyone who is trying to steal his information? Well, I got to thinking about this and I wonder if the answer is potentially no. Let me tell you why. You see, someone who goes through the old data in the stock market, using each of his algorithms could figure out which one he is using based on when he traded in or out of a particular stock. You can use these algorithms with historical data as to put forth "what if" scenarios to determine if you would have made money or not made money based on past stock performances and on certain charts at certain time intervals.Someone trying to reverse engineer a trading algorithm might very well be able to come up with a similar mathematical model with just a few clues. If someone were to file five or six patents with bogus algorithms, you could easily dismiss them one by one by retracing old data to see if they would work or not.Of course if you try to do it the hard way, you might you lose quite a bit of money, or it could take you six months to a year watching each of the filed algorithmic patents to figure out which one they were using. And like I said, he plans on continually upgrading his math and continually refine it to take into account a huge amount of data with unlimited filters. Please consider all this and think on it.