Automated Systems and Strategies Developer | Founder Tradingkernel.com | Time Based Trader
United States
- Tweets 675
- Following 4
- Followers 81
- Likes 47
Imagine a Neuralink implant feeding a full 360° camera view straight into your brain. This could let you see everything around you - front, sides, behind, without turning your head. Attention already works that way (think rear-view mirrors). The brain just needs the extra data.
6
This might be my 2026 Portfolio, along with several other strategies currently running, I’m deciding to stick to just 4 strategies for now, depending on what I find before the year ends.
myfxbook.com/members/Trading…
26
I haven’t really had any consistent trading account in 2026, this is because I decided to put in all my effort into prop firms around late last year, where I got funded but got denied payouts from several prop firms. Started something 2 months ago.
37
Bro, Algo Trading is Hard!!!
The amount of research you have to do weekly, especially over the weekends, and you don’t even get paid for it.
Shout out to every Algo Trader doing the research and putting real gems out here.
I respect you!
1
24
I ran the same condition a few different ways on recent data and got the opposite edge.
When Friday’s high (both 24h and RTH versions) was still unbroken into the Monday session, the short lost money in every window I tested.
The long side on those same Mondays was positive
This strategy shorts NQ futures every Monday when Friday's high was never broken.
Backtested on real tick data from February 2022 to May 2026 👇
Return: +573.30%
Profit Factor: 2.96
Win Rate: 22.6% (yes, wins less than 1 in 4 trades)
Avg R:R: 10.13
Max Drawdown: -15.2%
Total Trades: 62
Low win rate. High reward.
Every loss is small. Every win is massive.
See first comment to get the strategy 👇
46
I wrote a research report on Trading the Nasdaq-100 Closing Imbalance.
It looks at how the index behaves during the final auction, what patterns show up in the data, and whether there’s any consistent edge.
You can read the full report here:
tradingkernel.com/trading-th…
1
3
42
Did you know that 77% of Gold’s entire 2023 - 2026 move happened inside a 4-hour window? That is 14.3% of clock time. In-window 2.10bp/h; rest of day 0.05bp/h.
1
42
There is no way Shaw, Martinez and Mount, plays twice a week, for 2 months without getting injured, we are not asking for squad, we’re asking for squad depth!!
Carrick has already said the squad is balanced, so it is now expected of him to sort out the issues within the team and make sure we are winning games.
If the left-back position was not strengthened, then he has to find a solution and ensure that weakness does not cost Manchester United points during the season.
He has said the squad is balanced, so there should be no excuses later when injuries or tactical problems expose that position.
The responsibility is now on Carrick to prove that his assessment of the squad is correct.
51
Starting to think the Premier League has a secret agenda against Manchester United winning the league or Champions League. They don’t want another La Liga or Ligue 1 situation. How else do you explain City dropping £84m in January and £350m in August while United can’t get deals?
Jose Mourinho finished 2nd, rewarded him with a shit transfer window
Michael Carruck finished 3rd, rewarded him with a shit transfer window
This is why our banter era will never end, we aren’t a serious club
201
A system (bot, algo, or script)
that follows rules exactly as written,
with zero fear, greed, hesitation, or emotional interference,
will outperform a large percentage of manual traders,
simply because
most human traders struggle with consistency and emotional discipline.
1
28
If you’re running strategies on Gold or Indices, always filter them by monthly seasonality.
Some setups that look solid the rest of the year quietly die in December/January.
Don’t skip this filter.
1
29
You don’t even need to export your CSV, you can just ask it to go to your MT5 Instance. Or even better check your trade history directly on the chart.
Can confirm, this is super helpful. I haven't posted about my Trade Lab yet but it's pretty cool and also open source
Spoiler alert: the conclusion will be to trade less size, cut losers sooner, and stop trading after 2-3 losses in a row
30
Even Myfxbook and Tradezella can be faked.
The only thing that matters and is most important is INFORMATIONAL EDGE.
Give that to anyone and they make money on the long run.
This is why Algo Trading is the Future.
Manual trading would be obsolete.
You can fake a broker statement.
You can fake a prop firm withdrawal certificate.
You can fake a trade screenshot.
You can even fake a deposit.
All of that can be fabricated.
What you can’t easily fake is an automatically synced trading account connected to platforms like @TradeZella or Myfxbook through a reputable broker.
That’s where you can actually verify the track record.
1
1
137
When you develop a strategy, sometimes it might not look as good as it should be, usually because the informational edge you have is incomplete. and there exists a condition, where if added improves your performance significantly.
Here are some tips most Algo traders avoid :
Before throwing the whole thing on Claude/codex, try to reform the hypothesis of the strategy and turn it into a mathematical model, that truly can be tested and validated since it is NUMBERS, not plain English.
After that, try to understand and investigate the strategy deeply ( i usually take 10 days through this ), try to identify its strengths and weaknesses, its unfavorable regimes….etc
Before coding:
Manually, visualize the last 20 taken trades, code the strategy, recheck the code structure with another AI model ( a small difference in calculation can ruin the whole logic )
After coding and running the backtest, compare the last 20 trades to the 20 trades you visualized manually, to make sure AI coded the right strategy…. Most of the time, you will find AI playing the hamster game with you.
It can take you up to 20 days to get all these done, nothing to rush to otherwise.
1
2
190
Good Example Here, you can see the same strategy performing well across multiple intruments, with a similar Sharpe.
also
if it performs very similarly on related instruments, then it's likely you strategy is based on a solid informational edge.
A strategy can pass every robustness test on NASDAQ and still be unreliable.
That’s why we run one final check before going live:
Cross-market validation.
We take the exact same model (same rules, same parameters, zero changes) and apply it to markets it has never seen.
In our NASDAQ foundational model, we tested it on:
> E-mini Dow (YM)
> E-mini S&P 500 (ES)
> E-mini S&P 400 (EMD)
> E-mini Russell 2000 (RTY)
> Nikkei 225 (NKD)
Result:
All 6 markets profitable across all time periods
(including the last 1 year and last 2 years)
Even Russell 2000 (notoriously difficult) and Nikkei 225 (completely different country and hours) came out positive.
That’s the difference between a model that only works on one instrument and a model that captures real market behavior.
This is the final and hardest test.
If it fails here, we go back to the drawing board.
3
1
18
2,314
Chimelem Akani retweeted
It’s very rare to find a candle-based pattern that actually holds up in testing. This one seems to be different.
Very interesting results from the initial tests across multiple assets. The performance looks promising so far, but obviously needs a lot more testing before drawing any conclusions.
I’ll be testing this further and looking at how robust the pattern really is across different market conditions.
Three candles closing lower isn't a trend. It's delivery.
CISD is the moment delivery changes hands.
Price prints lower closes, then one candle closes back above the open of that leg. Sellers were in control right up until that close — now they aren't. That's the shift, and it's mechanical.
You don't have to predict the turn. You have to notice it.
2
18
3,273
The best version of your strategy is often the simplest form.
Every time I try to be a “smart guy” and add some complex quantitative concept to my strategy development process, I eventually end up ditching it and going back to the simple stuff.
At this point, I’ve fully accepted it: simplicity always wins.
2
167
Another misleading post. These companies combined hold over $900 billion in Japanese Bonds. and they get the full face value at maturity.
🚨 JAPAN’S BOND STRESS IS GETTING SERIOUS 🇯🇵
Japan’s four largest life insurers are reportedly sitting on a staggering $96 BILLION in paper losses on bonds.
Higher yields have consequences. 😳
1
93