Algorithmic Trading Bots for Prediction Markets

Sign each request with a fresh HMAC signature. The cURL and HTTP header values below are placeholders, not reusable credentials. For runnable signing code, use request signing examples. Python examples use the signed session from that page; Node.js examples use its bravadoFetch helper.

The problem

Polymarket order books are thin. Common markets have

2,000 to

10,000 of depth at top-of-book. A serious algorithmic trader running signal-driven strategies needs execution primitives beyond LIMIT and MARKET, or they eat the entire cost of their edge in slippage.

What Bravado provides

The Trade API v2 exposes six order types beyond the exchange native LIMIT and MARKET:

TWAP to schedule fills over time.
ICEBERG to hide displayed size while filling continuously.
PEGGED to track top-of-book without manual repricing.
STOP_LOSS, TAKE_PROFIT, TRAILING_STOP for server-side conditional exits.

All six execute server-side on Bravado’s engine on top of the Polymarket CLOB. That means you don’t need to run your own execution loop, hold state during retries, or reason about partial fills across a schedule.

APIs used

API	What a bot uses it for
Trade API	Eight order types including TWAP, Iceberg, Pegged, and the three stop variants, plus batch submit and cancel
Data API	Historical fills and PnL series for signal generation and post-trade analysis
Combos API	Multi-leg positions when a strategy spans several correlated markets
UMA API	Resolution timing, which determines when capital is released from a position

Worked example

Signal fires, size the position, execute with TWAP + ICEBERG, and set a trailing stop for the exit. All in one flow:

from bravado_auth import session as http, BravadoAuth
import os, uuid, requests

API = "https://partner-api.bravadotrade.com"
KEY = os.environ["BRAVADO_API_KEY"]
H = {"Content-Type": "application/json"}

def new_key():
    return {"Idempotency-Key": str(uuid.uuid4())}

symbol = "71321045679252212594626385532706912750332728571942532289631379312455583992646"
target_size = "20000"

# 1. Enter with TWAP + ICEBERG.
entry = http.post(f"{API}/v2/trade/order", headers={**H, **new_key()}, json={
    "symbol": symbol,
    "side": "buy",
    "type": "TWAP",
    "size": target_size,
    "limit_price": "0.65",
    "twap": {"duration_sec": 1800, "interval_sec": 60, "randomize": True},
    "iceberg": {"display_size": "150"},
}).json()

parent_id = entry["order_id"]
print(f"Entry: {parent_id}")

# 2. Attach a trailing stop for the exit.
exit_stop = http.post(f"{API}/v2/trade/order", headers={**H, **new_key()}, json={
    "symbol": symbol,
    "side": "sell",
    "type": "TRAILING_STOP",
    "size": target_size,
    "trailing": {"delta": "0.03"},
    "trigger_after_order_id": parent_id,
}).json()

print(f"Trailing stop: {exit_stop['order_id']}")

Two API calls: enter and exit. Bravado handles the child orders, repost logic, and trigger conditions.

Why this is different on Polymarket

Two constraints matter specifically for prediction markets:

Books are shallow. Splitting size across time and hiding the visible slice is more impactful than in equities.
Prices are bounded 0 to 1. A trailing stop of $0.03 on a$ 0.62 share is a 5 percent move, which is normal daily volatility on many markets. Set your stops accordingly.

Get Started

Supported Prediction Markets

Market Data

Trader Data

Execution

Automation

Resolution

Tools

Use Cases

SDKs

Algorithmic Trading Bots for Prediction Markets

The problem

What Bravado provides

APIs used

Worked example

Why this is different on Polymarket

​The problem

​What Bravado provides

​APIs used

​Worked example

​Why this is different on Polymarket

​Related

The problem

What Bravado provides

APIs used

Worked example

Why this is different on Polymarket

Related