EventTrader
Financial AI Models
PAPER
Menu
Dark Mode
Plain English Mode
PAPER TRADING MODE — Enable real trading on your Account page
Back
Backtest Guide

Test Your Bot
Before You Trade

Clone a bot, equip algorithms and skills, run backtests on historical data, and compete for daily rewards.

Start the tour ↓
Step 01

Clone a Bot

Browse the AI agent roster and clone any bot to create your own custom version. Your clone inherits the base strategy but is fully customizable.

DT
DELTA
Momentum Agent
Multi-timeframe momentum with trend-following bias
➡
YOUR CLONE
MC
My Clone
Custom Agent
Your own strategy based on DELTA
Step 02

Equip Algorithms, Skills & Tools

Customize your clone by equipping different algorithms for decision-making, skills for specialized analysis, and data tools for market intelligence.

Algorithms
📊
MACD Crossover
Moving average convergence-divergence signal detection
🧠
Headline Sentiment
NLP-powered market sentiment from news headlines
⚡
Momentum Rider
Trend-following momentum signal generator
Skills
🔍
Historical Pattern Matching
Detects recurring chart patterns from historical data
🛡
Volatility Filter
Filters trades based on volatility regime detection
Data Tools
📰
Binance WebSocket
Real-time price and order book data feed
💹
DeFiLlama Oracle
Cross-chain DeFi price and TVL data
⬇
MC
My Clone
Fully Equipped
MACD Crossover Pattern Matching Binance WS
Step 03

Run Your Backtest

Test your configured bot against real historical market data. Choose your asset, timeframe, and hit run — or use "Run Random" for an instant surprise backtest.

How the data is used

  • Bars: crypto uses Binance 5-minute bars; stocks use Yahoo 1-hour bars for windows of 30 days or less and daily bars beyond. If the live feed has nothing for a catalog symbol, stored history is used (daily equity OHLCV; crypto price points as flat bars).
  • Time: every timestamp is UTC; the window is now minus your period, and the result shows the dates actually covered.
  • Missing data: nothing is filled in or interpolated. The engine runs the bars that exist; a signal on one bar fills at the open of the next bar present. No bars at all means the run fails and your run credit is refunded.
  • Universe: the catalog is today's catalog, so delisted instruments are absent (survivorship bias). Each pick carries an asset_id such as crypto:BTC or stock:AAPL so you know exactly which series you tested.
MC
My Clone
Backtest Configuration
Asset BTC/USDT
Period 30 days
Trades ~12 expected
Running backtest... 100%
+0%
Total Return
0%
Win Rate
0
Total Trades
0
Sharpe Ratio
Step 04

Save to the Leaderboard

Happy with your results? Save your backtest to the public leaderboard and compete with other users for the top spot.

✅ Saved to Public Leaderboard
Illustrative example only. These names and returns are demo data, not reproducible runs; the live board at /backtest-results ranks standardised runs by risk-adjusted return. Past performance does not predict future results.
1
🏆 you WINNER
My Clone · BTC/USDT · 30d
+142.5%
2
alice
DELTA Clone · ETH/USDT · 14d
+98.3%
3
bob
STAT Clone · BTC/USDT · 30d
+76.1%
4
carol
CORR Clone · SOL/USDT · 7d
+52.8%
Step 05

Win Daily Rewards

Every day at 00:05 UTC, the top-performing backtest saved that day wins a reward. The reward is credited directly to your internal balance.

ET10
Leaderboard Rewards
📈
Highest Return Wins
Top total_return_pct on the day takes the prize
💰
Depositors Only
Must be a registered user with at least one deposit
📊
Min 3 Trades
Backtest must have at least 3 trades over 7+ days
🛡
Anti-Gaming Cap
Returns capped at 10,000% to prevent exploits
ℹ
Eligibility: You must be a registered user with at least one completed deposit to qualify for daily rewards. Backtests must be saved as public on the leaderboard.
🏆
Yesterday's Winner
Loading...

Ready to Compete?

Clone a bot, configure your strategy, run a backtest, and claim your spot on the leaderboard.

↑