Most financial articles, financial writers, and stock theorists will include some type of phrase that past performance does not equate to future performance. Basically, just because we did well or the stock did well (or poorly) in the past that doesn't mean that the same will be true in the future.
This is great advice even if it is hogwash. It reminds me of noise that was written by lawyers. Yes, I have similar phrasing on my site and in my book. Let's be truly honest though - the past is the best way for us to predict the future. In a few weeks, I will be publishing some historic back testing of my investing strategy and I have already published an example of IBM. There is simply no better way to understand how a strategy worked in the past as to how it will do in the future. It won't guarantee success but what test is more reliable?
Let's talk about a few examples that you probably can relate to that have nothing to do with stocks.
If you are hiring a new person at your company then you will inquire about that person's past. There are many questions that you will ask to ascertain if that person has the traits to be successful in the future. Most of those questions are about what the person did in the past that would give you some inkling as to what they will do in the future at your company. You might ask, "How did they handle this type of situation?" or "Give an example when this happened." These are valid interview questions that provide some idea of past performance and how it would result in future performance.
In the US, we just finished a Presidential election. While US politics is pretty ugly and vicious, part of the conversation was what the candidates did in the past which gives some voters some idea of what they will do in the future. Some voters don't use this information to make an informed decision but some will use it. It is valid to say that a candidate supported a certain law or effort which means they are likely to act a certain way on a certain type of legislation.
If a stock is trading in a range between $7 and $10, you would not expect it to go to $20 in the next 90 days without some instigator to change its behavior. Similarly, if a stock has been growing at 20% per year, you would not expect it to drop 50% in the next 4 weeks unless there was a particularly bad piece of news affecting the company's health.
Any good analysis of a stock needs to have some historical perspective, For this site (and my book) I look at 10 years worth of data. I weight the most recent years more heavily than years 8, 9 and 10. Sometimes that means that I ignore the hot company that went public last month but that is okay. I want to be confident that the company I invest in has a track record of success. I look for good past performance because anything else is just a guess. I don't guess with my money and I don't think you should guess with yours.