I meant to short PG at what is currently the exact HoD. However, there is a 5 second delay on trades from entry and exit. I suppose it makes sense in that people can't game it via data lag/delays but it was a few cents slippage because of it. It's still good though. My strategy doesn't care about it a lot of the time but here and there, some trades react almost instantly and become a factor.
Prediction Based Trading
- Thread starter stockpredictor
- Start date
Wiped most of the profit from the TWTR trade on this PG short. System has 1 day to turn a profit/breakeven... I suspect the r:r would be improved by cutting trades because most of them have low drawdowns. Price can run in a day though. I don't want to cut a trade just because another trade was profitable. There is a better way to manage the trades but I do not know them. There were 3 correlated assets around the time I shorted PG: PG, BMY and JNJ. PG was the worst one as it kept going up, JNJ only small drawdown, and BMY profitable. I think, anecdotally speaking, it would be best to pick 2 and split the usual position size among them. Regardless, it still isn't too bad of a drawdown. There was a trade some days ago I had not called. Long JD on the gray level into close, then it returned to B/E the next day, but with 6.8% drawdown.
Attachments
Yeah, today is probably not the best day to short P&GWiped most of the profit from the TWTR trade on this PG short. System has 1 day to turn a profit/breakeven... I suspect the r:r would be improved by cutting trades because most of them have low drawdowns. Price can run in a day though. I don't want to cut a trade just because another trade was profitable. There is a better way to manage the trades but I do not know them. There were 3 correlated assets around the time I shorted PG: PG, BMY and JNJ. PG was the worst one as it kept going up, JNJ only small drawdown, and BMY profitable. I think, anecdotally speaking, it would be best to pick 2 and split the usual position size among them. Regardless, it still isn't too bad of a drawdown. There was a trade some days ago I had not called. Long JD on the gray level into close, then it returned to B/E the next day, but with 6.8% drawdown.
I mean the Dow was 600 with S&P down 1.5%
Granny stocks like P&G are safe havens in times of uncertainty.
They go up.
You have to make sure you set the delay. It defaults to no delay. You should have set it when you created your strategy, if you haven't already.
Lol, good reframe. Have you considered other markets like futures and/or crypto?
Thank you.
It's easier for me to mentally gauge better position size on stocks rather than futures. Trades can take weeks to occur, thus a fut trade such as gold or oil may only occur occasionally. The strategy does seem to do well on bitcoin; I do not know about other coins.
Dang, the price went beyond breakeven for a bit. Could have gotten a win for 100% winrate but smashed market buy instead of limit because market was on its final minute and it was at a tiny loss. So its 50% winrate instead. This thing is making me a lot more conscious about numbers. plus, if I had used a limit order for the entry, the breakeven price would have been higher, although still right at the end. Oh well.
Ya know for real, as a whole, if you eliminate holding overnight, earnings/news events, and once the 9:30AM EST rush (dust has settled), most stocks move in concert with their respective sector.
If your system really works, you'd do well to tweek it to SPY, and use your signals to trade ATM near expiration options.
If it works that is.
If you're for real, you'll be ungodly wealthy in a few years time.
Yes, I avoid holding overnight when possible. There are occasional trades where next day direction/gap can be estimated with good probability (as long as its not earnings). I do trade the early morning volatility at times as well. It isn't quite as precise as usual but still has positive reward expectancy. But you are right; other than those things, it does well.
I used to use ATM weeklys but it actually doesn't work that well. Sometimes there are runs when it does, but the system itself requires up to a day to ensure probabilities of certain technical events. So I could lose a chunk if price stays flat for a day. If you look at the spreadsheet, the most recent GME loss was because of this. My system would have gotten out at breakeven but I held because they were weeklys and at a loss and I held it so it could be gotten out at a profit which didn't happen. The system can assess failures into breakeven even when entering (like the JD trade image I showed) but on stock, not options. Today's PG trade was basically a breakeven trade; if it were Friday expiration, it would be a 100% loss. Even though it is Monday, theta would take a chunk. Also, I think mentality is important; even if trading the system in a suboptimal fashion, if I tailor it to my personality, it improves moreso than taking the 'correct' approach due to human error. I think this way gives me a larger margin of error. That said, I did use biweeklys with one third of account size per trade for a month. Account went up well over 100% and I followed the system strictly and did not miss too many opportunities.
The system will not work if I traded it for just SPY. It can take weeks for a signal on any given stock to occur; I basically have lots of vehicles in rotation to trade each one occasionally.
The 5 second wait from entry isn't there anymore. Scored another half percent. Need to hold these shorts sometime; market keeps going down and no drawdown on these trades either