The overnights killed me in one day. The study is fatally flawed. Why? Because anyone using REAL money to do it cannot afford to continue doing it, when it fails.
"...A first pass look at the results are pretty amazing. Both tests buy 2,000 shares ($120,000 worth) in 1993. A buy and hold yielded $268,000 net profit. The second test which only bought on Friday, Monday or Tuesday, and only when the stock was either up 1.5% or down -1% over the past week, yielded $1,035,000. A massive quadrupling of the returns..." Then do it live with your own money. Let us know the results. I suspect it will not go as well as you think your sim model shows.
What on earth are you talking about, lol. How appropriate that Cramer is your avatar, you sound just like him.
Ok, since you asked so nicely. Here is an edge that will never go away. In the chart posted you can see my entire trade plan. I only buy above the blue 50 period SMA. I only sell below it. I use a 1 period SMA to help me easily see tops and bottoms. You only want to use closes for this not wicks. The first trade today was short right off the open. I marked it with a rectangle and a red down arrow. Note the red horizontal line. This line marks the low formed on the close of the 6am bar. That low was not tested until the market opened. This is the setup and can be traded with limit orders. Later in the day you can see a buy that was, by coincidence at the same price after the high was formed during the sell set up. Buy trade is also marked with an arrow and outlined. These are very common set ups and can be traded with limit orders and never have slippage due to being counter to the short term move into the levels. I usually do enter one tick below or above the levels in order to help insure a fill as these levels are competitive and you often have price hit them but not fill at the exact price. I also posted my weeks trade performance. This is a real money account. This is what is possible with simple robust set ups.
I know the question is coming so here is the proof its a real account. I marked out the personal info but you can clearly see its a real account as the sim is not checked, only the real account.
I did not expect you to actually post anything. You did, and it even seems like it might be legit, so can I admit I was wrong. Cheers for that I think I will try to backtest something based on this once I get some futures data (currently I only trade equities and I don't think it would work as well there) To summarize the way I understood what you wrote: -futures, 15 minute bars. -trade only with the 50 period sma trend (short below, long above). -use local minima/maxima in the 1-period sma to identify "key levels" of support/resistance. -when approaching a key level from below (above), go short (long) 1 tick below (above) the key level. If they are "competitive", doesn't that imply that the edge has limited capacity - hence it would stop working if enough people traded it?
It works on any instrument on any time frame. That is what makes it robust. You can trade bonds, oil, gold, e minis, softs, basically anything that trades. This is pure support and resistance. It is competitive because it is a breakout retest trade and they are often the best entries with the best risk reward. Market makers and institutions will spend massive capital to "get: to these prices. Read Trading and Exchanges by Larry Harris to get more insight into why this is true. There are PDF copies on the net. I pulled up a weekly chart to illustrate. Look at the low from Friday. Scary good for a swing trader.
Here is the trade on a one hour chart. Time frame is not important. I normally use five minute or fifteen for entries. Basically wherever I can see the best set up.