Polymarket Copytrading Bot

Copies trades from one or more leader addresses on Polymarket. Config in trade.toml, secrets in .env.

Market Types

• Polymarket includes Politics, Sports, Crypto, Economic, Geopolitical, Entertainment, and Experimental markets.
• Politics & Macro markets are longer-term and news-driven, suitable for medium-term copy strategies.
• Sports & Crypto markets are fast-moving and require quick execution (websocket mode recommended).
• Entertainment markets tend to be slower and lower volatility.
• Experimental / low-liquidity markets carry higher slippage risk and should use size limits.

Adjust filters like entry_trade_sec, trade_sec_from_resolve, take_profit, and buy_amount_limit_in_usd based on the market’s volatility and duration.

NOTE: Based on experience, crypto prediction markets can be highly volatile and risky, especially due to rapid price swings and fast position flipping by large traders.

For this reason, this bot is designed to copy trades across all market categories (Politics, Sports, Macro, Entertainment, etc.), rather than focusing only on crypto markets. Diversifying across different market types helps reduce concentration risk and smooth overall performance.

Setup

cp .env.example .env
# Edit .env: WALLET_PRIVATE_KEY, PROXY_WALLET_ADDRESS (if Magic), SIGNATURE_TYPE
npm install

Config: trade.toml

Key	Description
clob_host	CLOB API base URL (default mainnet)
chain_id	137 = Polygon mainnet
simulation	true = log only, no orders
[copy]
target_address	One or more leader addresses (single = websocket; multiple = polling)
revert_trade	true = copy BUY and SELL; false = BUY only
size_multiplier	Multiply copied size (1.0 = same)
poll_interval_sec	Polling interval when using multiple targets
[exit]
take_profit	Sell at this % gain (0 = off)
stop_loss	Sell at this % loss (0 = off)
trailing_stop	Sell if price drops this % from high (0 = off)
[filter]
buy_amount_limit_in_usd	Max USD per copied trade (0 = no limit)
entry_trade_sec	Only copy if leader traded within last N sec (0 = off)
trade_sec_from_resolve	Skip if market resolves within N sec (0 = off)

Run

npm run dev    # tsx src/index.ts

Project layout

src/
  index.ts       # Entry: load config, start copy + optional exit loop
  config/        # Load trade.toml + .env → AppConfig
  client/        # CLOB client (ethers + API key)
  trading/       # copyTrade(): place order from LeaderTrade
  polling/       # Multi-target: poll positions API, diff → trades
  realtime/     # Single target: websocket activity → trades
  filter/       # shouldCopyTrade(): revert_trade, entry_trade_sec, trade_sec_from_resolve
  exit/          # recordEntry(), runExitLoop(): take_profit, stop_loss, trailing_stop
  types/         # AppConfig, LeaderTrade, etc.

Env

• WALLET_PRIVATE_KEY – EOA or Magic export
• PROXY_WALLET_ADDRESS – Polymarket profile (required for Magic; optional EOA)
• SIGNATURE_TYPE – 0 = EOA, 1 = Magic/proxy, 2 = Gnosis Safe

Journey of Builder - xstacks

While many traders focus heavily on crypto prediction markets, I noticed that they can be extremely volatile and unpredictable in short timeframes. Rapid price swings, aggressive position flipping, and sudden liquidity shifts make consistent automation difficult.

Because of this, I shifted my focus toward non-crypto markets — such as politics, macroeconomic events, sports, and major world events — where price movements tend to be more structured and event-driven rather than purely momentum-based.

I began developing this strategy in early February 2026, spending about two weeks testing and refining it. Instead of chasing high volatility, the goal was consistency and risk control.

After running simulations on historical market data and testing live with a small balance, the strategy showed steady results — averaging around $50–120 per day from a $200 balance, depending on market conditions.

What I found is that non-crypto markets often:

• Move more gradually
• React to scheduled events
• Offer clearer sentiment shifts
• Have less algorithmic noise compared to short-term crypto markets

Rather than targeting a single category, the bot is designed to operate across multiple non-crypto markets, allowing diversification and reducing dependence on one asset class.

The focus is not explosive gains, but stable performance and controlled risk. The system is still evolving, and I continue to refine filters, timing logic, and position sizing to improve long-term consistency.

This is a long-term strategy — built around discipline, diversification, and steady growth rather than hype.