@kerranm TradingView prints CME settlement as the daily close on futures. Our C++ rolls up last trade at the 4pm CT session end. Same day, two different "closes." Fix: both rebuild dailies from the same 1m data at the same boundary. Ninja's bar is the referee.
My backtest and TradingView disagreed on the same indicator for the first time.
Build Log: Day 6. The AI that wrote the backtester also found why: each side was using a different daily close.
If your backtest and your chart don't match, check that first. Then decide which one to trust.
I am currently testing on Ninja as well. So far C++ is working better than pinescript at backtesting original pine code with AI.. HMM
Monetizing slowly. Last thing I want is to be a Furu. Being trusted initially is more important to me than money. Money will come as long as trust is there and everyone else is making money. I charge for 1 on 1 classes and only have a couple of students. Let go of most of them as most don't have what it takes.
But alpha must be researched and used on one's own recognizance. Often costing money, and most of all time. It is a bigger price to pay than most realize, and what we are building will help with that. That has been enough for me to not charge for anything that doesn't cost 1 on 1 time for me.
This is why I have coded my indicators on 3 platforms.. Pinescript isn’t even close to the best way to get ideas onto your chart.
This account will soon be handing out alpha for free left and right.
I built a trader with zero edge. A literal coin flip. $3k target, $2k drawdown, 1:1 payoff. Ran him through the eval 200,000 times. Static drawdown: passed 40% of the time. Trailing drawdown: 25%. Passing an eval isn't proof. It's a coin that landed well. n=200,000 simulated, no fees.
*Add $5 of commissions per trade and the trailing version drops to about 19%. The rulebook is the edge, and it isn't yours. Full sim and the code in Sunday's newsletter." Source: our own Monte Carlo, Sep 2026; static result matches gambler's-ruin math (2,000 ÷ 5,000)*
This side project has been the most entertaining thing I have done in a long time.
Maybe after serious live testing, the private discord will go public, but instead of trading Alpha, it will be AI Alpha. Plans change. Shout out to @milesdeutscher. I too don't roll anything out until I know it will help the community. Starting off strong 💪
A green candle doesn’t mean the macro problems disappeared.
A few things worth understanding if you trade futures or crypto:
Higher rates don’t hit the economy overnight. More expensive borrowing takes time to work through new mortgages, business financing and debt refinancing. Expecting an immediate crash misses that delay. Understand the economic backdrop, but don’t confuse a big picture opinion with a timed trade.
Being bearish on the economy doesn’t automatically make the next candle a short. Being bullish long term doesn’t automatically make every dip a buy.
A macro thesis is not an entry signal.
Some Glean Collective Tilt Rules
• 2 losses = stop
• Cut size after loss
• No instant re-entry
• Daily max loss = done
• Trade system, not PnL
• Reset or stay out
Tilt management is the highest value edge in trading.
No one “masters” it.
You manage it… every single day.
That’s why collectives win.
Glean Collective is building
Big things coming
The trading space is broken
No real education
Just noise
But there are real ones
Gleaned this setup from @Tradersreality
Tino is one of the real ones
Quick glimpse:
Liquidity tells the story
I see where this is likely headed
Curious who else does
Glean not chase
Most people don’t understand what “Glean” means.
It means to gather what others miss.
In trading, that’s edge.
Not found in one setup.
Not found in one trade.
Built over time.
Edge evolves.
That’s what we focus on here.
— Glean Collective