I spent almost 2 years building a quantitative swing trading strategy - with a winrate of upto 80%
I have been working for close to 2 years refining my swing trading strategy, Python scripts, quantitative tools and statistical models that are used in the strategy.
**Strategy Overview:**
My strategy is largely based on quantitative and statistical modelling of underlying securities.
Basically, I take a security, spread its historical price action over several years, and study what happened when the security was trading at similar statistical levels in the past.
The objective is to identify historically rare conditions where the underlying security has demonstrated favourable forward returns.
**How it works:**
**My Stock Universe:** I take Quality names from the S&P 500 and QQQ only - I update this list every night with my own criteria on top.
**Quantitative Models:** Then I built a tool that analyze historical price distributions, setup frequency and subsequent performance across multiple time horizons for each security from my stock universe.
**Intraday Scanner Job:** Runs these models on every stock in the universe 4 times during trading hours.
**Alerts:** Then a built another python script that sends me the list of the stocks that are revisiting their rare levels and have the potential of higher forward return
**Transparency:** For transparency, I saved all the alerts with their alerts price and market cap at alert and the setup historical frequency they were alerted at. Then I build another statistical model of these said alerts - the winrate of the stocks with market cap > $500 billion was more than 80%. However, these opportunities are very rare. For Example, the system alerted NVDA only once in the last year and it went up 30% in 20D after the alert.
I am personally currently using this strategy. I would call it a semi-passive strategy. This strategy has also helped me get out of my daytrading/gambling addiction lol.
I rarely see charts or rely on charts. Stock News is absolutely muted on my part using this strategy.
I already had the python scripts and my jobs were running on a scheduler. I put everything on a simple website if anyone wants to try it live and look up the stats themselves.
The website is [3horses1dog.com](http://3horses1dog.com) (mods please let me know if I am not supposed to add this link, but this is not a paid service. Its basically a stats tool)
Thank you so much if you have read this so far. I wish you all the best.
p.s. this is strictly not a trading or financial advice - this is just a quantitative research and educational tool