I've been trading for a little over 1 year now. For the first 6 months my experience was pretty miserable, but I've been doing well lately after implementing a few strategies that I'm constantly trying to optimize. I'm bored and I like to share information so I'll talk about one of my strategies, but I won't mention any specific stocks to avoid the crowding effect.
Basically, I discovered that it is SUPER easy to manipulate algorithms in stocks (priced around $5.00) with tight spreads and low volume. By "tight spreads," I mean stocks with a bid/ask spread of no more than three cents on average. And by "low volume," I mean volume from approximately 250,000 to 1,000,000. This type of stock is run almost exclusively by algorithms, so if you understand the criteria that they use to set and cancel limit orders, then you can manipulate them pretty easily.
First of all, whichever index is associated with the stock is probably the most important criterion that they use. If the stock we're dealing with is associated with the XBI, for example, then the stock will tend to head in whichever direction the XBI is heading. It's biased. This is measured through technical analysis. Although I really hate technical analysis, it is abundantly clear that algorithms use it because it's the only thing that they CAN use to gauge their surroundings. So, what I'll do is apply "pitchforks" and draw basic lines of resistance/support (a brain dead nine-year-old can do this). If the XBI breaks out of resistance and is performing well, then I'll go long. If it breaks below support, then I'll go short because you don't want to fight the algorithms. If they are associated with the movement of an index, then do not fight their association. They want to "go with the flow" most of the time.
Notice I said "flow." The algorithms want to go with the "FLOW," not a single, massive buy order. If you set one big buy order, the algorithms will TRAP you and bring the price down to put you out of the money. If you break your buy order into several components over the course of 30 seconds, the algorithms will go with the flow and they probably will not trap you, as long as you're buying at the right time (based on the index). By the way, most of the time the algorithms will not initiate the "flow," that is something that YOU do by going long or short. They don't automatically adjust the stock to the index. Most of the time, they adjust the price of a stock after humans interact with it, so if a human buys and while the index is heading upwards, the algorithms will go with the flow of the index—the flow that YOU initiated.
So, with this in mind, if you predict that the index will head upwards throughout the day, you might think it is a good idea to take advantage of the three-cent spread by placing a limit order to sell just beneath the next best offer, right? Because, then, you could buy even lower than the next best offer before you push the price up. It's not that easy. You should do the opposite because the algorithms aren't stupid and they know what you're trying to do and they won't let you push the price up. In fact, they'll push it down if you try to do them dirty like that.
Anyway, we talked about 1. when to push the price up/down and 2. how to purchase shares before the time comes that you want to push the price up/down. But HOW do we push the price up/down? That's where it gets really fun and it's the only reason why this isn't boring (besides making tons of money). Basically, as I mentioned earlier, you want to break your buy order into several components. It's a good idea to know your risk tolerance. It's not a good idea to put down hundreds of thousands of dollars because you have to UNLOAD all of that. You need to put down at least 60 thousand dollars to have a reliable impact on the price (what I do is, on a $6.00 stock, purchase between 1500 and 2500 shares 4 times in a row—if you don't have the money then just work for a few months and play with a few pumps/dumps). The numbers 1500 and 2500 work very well most of the time regardless of the stock's price. Sometimes I get pissed that the stock isn't moving and purchase even more than that. But don't get too pissed or else you're gonna have tens of thousands of shares that you need to unload at a price that never moved. That's really bad because YOU DIDN'T EVEN MOVE THE PRICE, BUT YOU NOW OWN 30,000 SHARES THAT YOU HAVE TO UNLOAD SOMEHOW! That happened to me earlier and cost me several thousand dollars, but I was able to use those shares to manipulate it to the downside until it was oversold. If you're ever in that situation, just keep adding volume to the system until something works. So, basically, to push the price up/down, you do it while purchasing "shares before the time comes that you want to push the price up/down," as I mentioned at the beginning of this paragraph. They occur simultaneously.
If the stock does not move even when you try to move it in EITHER direction, then that's not good. You want to add chaos to the system by buying and selling thousands of shares simultaneously. This often results in a small loss, but it is worth it because it creates opportunities and puts the algorithm in a different "mode." This is similar to how Julius Caesar would win a battle when there are few options available: he would create more options.
Now, how do you unload your shares to realize profit? Basically, this entire process should not take more than 30 seconds. But you can replicate it dozens of times throughout the day on different stocks. But to unload shares, you want to do everything I mentioned earlier but in reverse. Set counter-productive limit orders to make it seem as though you're buying, then sell. The algorithms are sticky in whichever direction they're biased, so it's easy to unload without causing the price to fall much. You should also set dozens of 100-share limit orders on both both sides of the spread. This is so that you don't have to manually input limit orders when you need them without delay. I execute this entire process on the Thinkorswim platform.
I'm constantly trying to optimize but it seems to work out really well. Profits are not guaranteed on every trade but it's pretty much guaranteed on average. There are tons of variables. Every algorithm is different, sometimes algorithms are trying to accumulate shares and are biased regardless of the XBI or IWM, sometimes they have different objectives in mind, but it's a lot of fun overall. It's a good way to waste time while doing other things on your computer. It would help if you have two monitors. I've been using this strategy a lot because this market is overvalued trash and not worth investing 1 dime in. Day trading is supreme, don't listen to all the failed day traders who rail on it all the time.