Pinned Tweet
After years of manual trading, I'm going full algo.
Why?
β’ I'm tired of emotional decisions
β’ I want systems that trade while I sleep
β’ I believe retail traders deserve institutional-grade tools
Starting today, I'm documenting the entire journey:
β
Every strategy I develop
β
Every backtest (wins AND losses)
β
Every tool I build
β
Every lesson I learn
My background:
π Data analyst (markets, futures, crypto)
π» Python for analysis + automation
π Trading futures (ES, NQ) + crypto since 2020
My goal:
Build a profitable algo trading system from scratch. Share the playbook for free.
Why share everything?
Because I wish someone had done this for me when I started.
Follow if you want:
β’ Real trading strategies (not theory)
β’ Python code (not pseudocode)
β’ Honest results (not fake screenshots)
β’ Free tools (not $2K courses)
First post: Tomorrow.
First strategy: Mean reversion on ES futures.
First lesson: Already learned that "easy backtests" are a myth.
Let's build. π
45
AlphaAutomated retweeted
been working on something dope for quite some time & I'm excited announce the launch of our crypto options analytics platform :
derivativesmonkey.com
No Login or Payment Required. It's completely free to use!
will explain in tweets below what it does below & how it worksπ
AlphaAutomated retweeted
π $2,500,000 GIVEAWAY π
We teamed up with @FundedNext to celebrate their new FLEX Challenge π€
Win 1 of 25x $100K FLEX accounts!!
To enter:
β’ Follow @Tradesyncer & @FundedNext
β’ Like & repost
β’ Comment βDONEβ
β³ Winners announced Friday.
AlphaAutomated retweeted
$1,250,000 TRADEIFY EVALUATION GIVEAWAY π
It's time. @Tradesyncer x @Tradeify
We're giving away 25x $50K Select Plan evaluation accounts. Here's all you gotta do:
β
Follow @Tradesyncer & @Tradeify
β
Like, Repost & Comment "DONE"
P.S. Tag someone for an extra entry.
Good luck! π
Unpopular opinion:
Most "profitable" algo strategies shared online are:
1. Curve-fitted to historical data
2. Cherry-picked best results
3. Ignoring transaction costs
4. Tested on clean data (not real market conditions)
Always verify. Always question. Always test yourself.
#AlgoTrading #TradingReality
12
Best free resources to learn algorithmic trading:
π Books:
β’ "Algorithmic Trading" - Ernest Chan
β’ "Quantitative Trading" - Ernest Chan
π₯ YouTube:
β’ QuantPy
β’ Part Time Larry
β’ Coding Jesus
π» Platforms:
β’ QuantConnect (free tier)
β’ Backtrader (open source)
No paid courses needed to start.
#AlgoTrading #FreeResources
30
Position sizing formula I use:
Position Size = (Account Risk % Γ Account Size) / (Entry Price - Stop Loss)
Example:
β’ Account: $10,000
β’ Risk: 1% ($100)
β’ Entry: $50
β’ Stop: $48
β’ Risk per share: $2
Position Size = $100 / $2 = 50 shares
Never guess. Always calculate.
#RiskManagement #AlgoTrading
8
The difference between backtesting and forward testing:
BACKTEST:
β’ Test strategy on historical data
β’ Shows what WOULD have happened
β’ Fast, cheap, easy
β’ Risk: Overfitting
FORWARD TEST:
β’ Test strategy on NEW data (or live)
β’ Shows what ACTUALLY happens
β’ Slow, real-money risk
β’ Proof: It works now
You need BOTH.
#Backtesting #AlgoTrading
1
10
Free tools every algo trader should know:
π TradingView (charting + backtesting)
π Python + pandas (data analysis)
π QuantConnect (cloud backtesting platform)
πΎ Yahoo Finance API (free historical data)
π Backtrader (Python backtesting library)
All FREE. No excuses.
What am I missing? π
#AlgoTrading #FreeResources
1
29
Common backtesting mistakes that inflate results:
1. Look-ahead bias (using future data)
2. Ignoring slippage/commissions
3. Overfitting to historical data
4. Survivorship bias (only testing "winners")
5. Not testing on out-of-sample data
Which one have you fallen for?
#Backtesting #AlgoTrading
4
Risk management rule I wish I learned earlier:
Never risk more than 1-2% of your account on a single trade.
Why?
Because even a 50% win rate strategy can blow up your account if you risk 10% per trade.
Math breakdown π
#RiskManagement #AlgoTrading
1
5
3/ The Kelly Criterion formula:
f = (bp - q) / b
Where:
β’ f = fraction of capital to risk
β’ b = win/loss ratio
β’ p = win probability
β’ q = loss probability (1 - p)
For most strategies, Kelly suggests 0.5-2% per trade MAX.
1
35
The 5 levels of trading automation:
Level 1: Manual execution (you click buy/sell)
Level 2: Alert-based (system alerts, you execute)
Level 3: Semi-auto (system executes, you approve)
Level 4: Fully automated (system trades, you monitor)
Level 5: Autonomous (system trades AND manages risk)
Where are you? π
#AlgoTrading #TradingAutomation
2
Python tip for algo traders:
Use pandas `.resample()` to convert timeframes instantly.
1-min data β 5-min bars
5-min data β 1-hour bars
Daily data β weekly/monthly
One line of code.
Example π
#Python #AlgoTrading
1
9
import pandas as pd
# 1-minute OHLC data
df_1min = pd.read_csv('price_data.csv', index_col='timestamp', parse_dates=True)
# Convert to 5-minute bars
df_5min = df_1min.resample('5min').agg({
'open': 'first',
'high': 'max',
'low': 'min',
'close': 'last',
'volume': 'sum'
})
# Or convert to 1-hour bars
df_1hour = df_1min.resample('1H').agg({
'open': 'first',
'high': 'max',
'low': 'min',
'close': 'last',
'volume': 'sum'
})
Game changer for multi-timeframe analysis.
No need to download separate datasetsβjust resample.
9
The #1 mistake in algorithmic trading:
Optimizing for win rate instead of risk-adjusted returns.
A strategy with:
β’ 40% win rate
β’ 2:1 reward-risk ratio
β’ Sharpe ratio 1.8
Will outperform:
β’ 70% win rate
β’ 1:1 reward-risk ratio
β’ Sharpe ratio 0.9
Why Sharpe > Win Rate π
#AlgoTrading #RiskManagement
1
7
4/ What to optimize for:
β
Sharpe ratio (>1.5 is solid, >2.0 is excellent)
β
Profit factor (total wins / total losses, aim for >1.5)
β
Max drawdown (<15% for aggressive, <10% for conservative)
β
Recovery factor (net profit / max drawdown)
β Win rate alone
1
3






