I believe trading is far more an art than a science. For one thing, if trading were a science, then we should all be able to enter the same trade at the same time, and exit at the same time, getting identical results. We all know that just isn't so. Yet if trading were scientific, we should be able to get identical results by doing the same thing. I believe such expectation describes the "scientific method." In addition, if trading were scientific, we should be able to come up with a "get rich" formula that would work all the time. We could then all retire and never have to work again. We all know this isn't so either. When we, as traders, make a trading decision, most of the time we do not fully know why we are making that decision. You look at a market, you think about taking a trade in that market, and at some point you pull the trigger. You have thought of dozens of things in the time interval leading up to your entry. If I were to ask you, "Exactly why are you buying what you are, or why are you selling what you are?" you would probably not be able to give an accurate answer. You may be able to give a few reasons, but it will most likely not be the full answer. A lot of your decision to enter is subconscious. You do not really know why you entered, especially if you are day trading. To that extent it is more an art than a science, because you cannot fully demonstrate why you are doing what you are doing. But you could say, "I fully know what I'm doing. I am taking the trade because I am following the signals of my method or system." Wonderful, you have just proved my point. When you are blindly trading signals from a method or system, you truly don't know why you are taking the trade. You are essentially acting like a robot, pre-programmed to follow signals whether or not they make sense. I am not disparaging trading that way. If a method or system produces winning results, then what you are doing is following a statistically proven plan. All methods and all systems are based on statistics. The odds on any single trade are never more than 50% win or lose. However, the probability for a succession of trades is quite another story. If you are trading a method that wins seven out of ten times that you enter, and the method has produced a loser three or four times in a row, then the probability for a successful trade increases each time you enter the market. Sooner or later, over a series of trades, you are going to have the result of seven winners against 3 losers. That is statistically valid; however, it is not exactly rocket science. You will have proven that trading is an art -- the art of following a statistically valid plan.
All sciences are not exact. There are strategies that have positive expectations which in my experience are found by hard work and luck. I'll leave the art part to someone else to discuss.
I know exactly why I'm buying and selling and have a specific reason. Sometimes I take shots based on feel, and when I do that, I know why as well, and would say "ehh it just feels like it's down too much and slowing and it held .14 three times so it should probably test .25."