Well, fancy that. Going long overnight into a 20% correction can be profitable. Why do I have a hard time believing this. My results show that going long overnight in a bullmarket can cause a loss. So what is the skinny? Run it real time, with manual entries, and then get back to us with results.
I don't have time for that. Nonetheless, the numbers don't lie. If after 1000 trades with staggered tests done in different time periods the results maintain 58%, then you have an edge. I've run my own automated strategies real-time on baskets of commodities and have had some success in the past. However, my own brain seems to net me a bit more, with more consistency, so I stick with that. Good luck.
Was just reading that corporate buybacks to the tune of $227,000,000,000 In the first quarter of 2019 compared to $143,000,000,000 in the first quarter of 2018.... That's the reason for the risk free gains...the continuous buying of their company stock is putting an artificial rise to the market. Again another way of propping up markets day in and day out.... Trillions of dollars of buybacks to keep the bull market afloat.....
Unfortunately, our only decision is to buy or sell. We can't choose if they buy back or not. We can't choose if they print free money to give to companies who buy back. Nothing we can do here. Only buy or sell. So it is a rigged game. Either don't play or play wisely.
Thank you for taking the time to do this. Ive known it was profitable for years and have mentioned it here countless times. Sometimes I use QQQ because in the wee hours of the morning it can't get stopped out. I do this when volatility is high.
As a filter, try this: Buying the close and selling the Open after 1) A negative close 2) A close below the open You should get above 60% with this. There are also weekday biases which can be utilized. Better than simply selling the Open might be to utilize a profit target/order as there are many times the market rise during ETH, but drop by the Open.
No one that I know who uses strategies trades manually, otherwise the results are skewed as you demonstrated earlier in this thread by missed orders and non-closes and all the idiot mistakes that we make as humans to interfere with systems trades. This is not rocket science here - the system is designed to submit a buy order automatically at exactly 4:00pm and a sell at 9:30am. Having run live strategies of my own, on a basket of commodities, my real-time results actually exceeded my backtests slightly because I kept getting positive slippage. That being said, over thousands of trades, with staggered backtests and walk-forwards, the numbers can be relied upon if the parameters are simple, which in this case, they are. It is my un-expert opinion that people often expect too much from systems, too fast, they discount huge factors like commissions and drawdowns, and they don't have the staying power to see anything through (ie. get a large enough sample from which to draw results) in real-time. Ultimately, this is glorified gambling and the law of large numbers comes into play eventually providing that the system does have a positive expectancy. But that takes time. And paradoxically, you still have to be a trader if the system starts exceeding worse-case backtest results and shut it off. Is there an edge to the Rickshaw strat? Definitely. Roughly 8% before commissions if it were run vanilla with no filters.
The US stock market capitalization was 34T in 2018. There is no way YOUR numbers are correct. Notwithstanding the error, buybacks are one of several reasons for the extended bull. But still, only one of several reasons, and with no where near the impact you've ascribed.