Day Trading vs. Swing Trading vs. Position Trading 86 replies
Disliked"claude make me rich" why spend time to make a bot that cant learn from you? we are not there yet and hopefully i assumed correct that the people that are building these things are profitable... but i feel like trying to vibecode a bot should be just a hobby, and a lot of people i see here take it serious, says a lotIgnored
Disliked{image} I have been busy building myself a bespoke analysis platform. Mainly so that I can turn the time and sales data into something useful for my eyes to watch. On screen we have 1-min candles, Volume histogram, Volume Profile, Cumulative Delta and Volume Delta. There are levels for VAH, VAL, POC from the previous day and in the developing day. The GEX Levels are also present, so I can tell whether the market makers are -ve or +ve gamma. This isn't an execution software project; it is purely for trying to work out in finer detail who is in control...Ignored
Disliked{quote} This is the kind of stuff I wish TradingView did natively.how are you calculating the GEX levels?Ignored
Disliked{image} For those following along, I just updated the strategies site. Tested 112 different ideas. Most failed. https://www.rfastrat.com/module1 PS. This is a free resource, not some sort of sales funnel.Ignored
Disliked{quote} Shoutout for building that site for everyone! I’ve been working on something similar myself. I pulled hundreds of TradingView strategies using Codex, then ran them all through FTMO’s $100k account rule set to filter out the bad ones. Quick heads-up on FTMO’s rules: their $100k funded accounts used to let you hold trades overnight, but that’s no longer allowed now. I built a full automated screening pipeline around these compliance rules. I ran 400 distinct strategies through this auto-check system, and only one ORB strategy passed all the...Ignored
I turned every single one of these rules into weighted scoring metrics, then ran a 5-year backtest on gold. Shockingly, the strategy stayed profitable. Stretch it out past five years, though, and it falls apart—10-year returns are barely anything. But its performance over the most recent five years is really solid.
My main point is this: my original design didn’t use weighted scoring at all. Now that I’ve rebuilt it around a weighted scoring system, the entire dynamic of how the strategy trades is totally different.
DislikedI came up with a setup that feels kinda ridiculous, especially coming from someone who’s traded ICT and SMC full-time before. I’m not bringing LLMs into this at all for now. Here’s the breakdown of my criteria: Close price relative to EMA20, EMA50 and EMA200 The alignment/relationship between the three EMAs themselves MACD histogram above or below zero RSI score reading Correlation between H4 closing prices and the current M15 price action Bollinger Bands conditions I turned every single one of these rules into weighted scoring metrics, then ran...Ignored
Disliked{quote} Have you considered the possibility that you generated a strategy that fitted a curve, and even though it lasted 5-years, it didn't hold up robustly? You have a lot of variables, so I would be interested to know why you chose those specifically.Ignored
Disliked{quote} I’m swamped with house renovations right now, but once I wrap all this up in a few days I’ll generate a prompt with Codex and send it over so you can see exactly what I’m working on. Fr fr, I don’t know jack squat about coding at all—all I do is chat back and forth with Codex to build this stuff. Here’s the big question I’ve been stuck on: everything boils down to model weights anyway. My original setup uses an LLM to pull feature data based on a design I mapped out earlier. But the post I mentioned earlier was about ditching LLMs entirely...Ignored