The bands are custom. I use another indicator for entry, exit depends on a lot of things. Can be 2R, 3R, or the opposite band. Stops are compulsory but very rarely hit. My mistake, my wording was incomplete. Entry is only after price goes outside a band and comes back inside. Looking at the chart its obvious that trades have to be taken after price returns into the bands.
Lets see an after the fact example where high and low time frame line up, i.e. the 1 minute outside and return happened after the 30 or 60 minute (forget which you use?) went out and came back in its own band.
Lol you can create your own indicator by running 1std bands around price. The hypothesis around that is that stocks tend to have a peskier distribution of returns, and so you can fade the extremes. In real life, the performance of those strategies are not great. There are a couple of websites that have backtested both intraday momentum and mean reversion (if you’re serious, look them up). While it’s true that the level of price should mean revert in between informational events, in HFT world the trade is the event. So the sort of inertia you’re anticipating is really happening in the sub300ms realm. In lower frequency assets like bonds, you could draw some sort of expected path because you may not get a trade for minutes. But because equities have such greater frequency, your sub300ms strategies don’t scale up to larger time horizons and vice versa. Also, remember that what you’re seeing on a chart is aggregated time and sales data, not b/a data (the leading indicator of price).
I never specified which combination of timeframes I use. The charts I posted were to prove the laggers/repainters wrong. My entry point is determined by another indicator which I've avoided showing in the charts. Not interested in exposing it by posting entries on 2 TFs to give the reverse engineers ammunition. Anyone who thinks it doesn't work, feel free to think the charts were made in Photoshop and my posts are bogus. Nothing is for sale. I have no service or interest in starting one. At some point in the future I may use a trade copier. If you're so interested send me a PM.
Thanks for your reply, but I've heard that reverse engineer excuse for 25+ years. Maybe start saying AI-reverse engineer to spice it up even more.
Trading is an art. Regardless how hard you trade to go fully systematic, the edges come and go very frequently, not enough to make huge profits to get rich from, if you found one (edge), which is also very hard. The correlation and cointegration thingy is hyped too. I tried it hard, it is not working longterm. Cointegration comes and go. And mean-reverting is hard to trade, you need to have a stoploss and then you RRR ratio per trade is not good so you need high win ratio. Trading is an art because the success or any edge is depending highly on discretionary elements, things you cannot backtest properly at all. My two cents (after spending 15 years fully systematic). And one thing, this hype here for this "magic" bollinger band looking custom indi, forget it. There is simply no magic indicator or formula at all in trading.
Generally yes I traded the same instrument over 4 years just VIX options due to the mean reverting nature of volatility it actually has to mean revert unlike the co-integration as you also discovered.
I agree too I didn't no back testing because it would have been counterproductive I did look at past charts though to calibrate the volatility model in my head but I decided that approach is a bit too loose and I'm not trading again until my software is done that can help me enter and manage these positions