Trades 🌱

The polygon.predictions.trades table provides comprehensive trade data from prediction market platforms on Polygon. Currently featuring Polymarket data, this schema is designed to seamlessly integrate additional prediction market protocols as they become available. Use this table to analyze trading patterns, market volumes, participant behavior, and event outcomes across prediction markets on Polygon.

Table Details

Property	Value
Table Name	`polygon.predictions.trades`
Table Status	Beta 🌱
Unique Key	`block_timestamp`, `transaction_hash`, `log_index`
Clustering Key(s)	`block_timestamp::date`
Search Optimization	`transaction_hash`

side and side_name are the maker’s resting order, not whether the fill was a sale. BUY means the maker posted a bid. SELL means the maker posted an ask. Each fill has two parties; taker is the other party. These fields are null on FPMM rows, which have no maker.

contract_version = 'v3' is Polymarket Combos (ComboExchange parlays). condition_id on those rows is the combo id, not a markets.condition_id. asset_id is a PositionManager id, not markets.token_id. Each parlay leg’s market token is on combo_legs.

Table Columns

Data Precision Notes:

_raw fields contain unnormalized amounts for maximum precision
_str columns are numeric fields cast to varchar to retain precision
USD amounts reflect value at time of trade

Column Name	Data Type	Description
event_name	VARCHAR(16777216)	Name of the prediction market event.
project	VARCHAR(16777216)	Name of the project (e.g., Polymarket).
protocol	VARCHAR(16777216)	Protocol name (e.g., polymarket).
exchange_type	VARCHAR(16777216)	Type of exchange (e.g., Polymarket Fixed Product Market Maker, Orderbook, ComboExchange).
contract_version	VARCHAR(16777216)	`v1`, `v2`, or `v3`. Null for FPMM. `v3` is Combos (ComboExchange).
market_address	VARCHAR(42)	Address of the prediction market contract.
order_hash	VARCHAR(16777216)	Unique hash of the order.
maker	VARCHAR(42)	Address of the maker in the trade.
taker	VARCHAR(42)	Address of the taker in the trade.
side	NUMBER(38,0)	Maker order side. 0 = BUY (bid), 1 = SELL (ask). Null for FPMM.
side_name	VARCHAR(16777216)	BUY if the maker posted a bid, SELL if the maker posted an ask. Null for FPMM.
builder	VARCHAR(16777216)	V2-only attribution field (bytes32). Null for V1 + FPMM.
metadata	VARCHAR(16777216)	V2-only arbitrary per-order data (bytes32). Null for V1 + FPMM.
condition_id	VARCHAR(16777216)	Market condition id on v1/v2. Combo id on v3 Combos rows. That combo id is not a `markets.condition_id`.
asset_id	VARCHAR(16777216)	Outcome token id on v1/v2. PositionManager id on v3 Combos rows; it does not match `markets.token_id`.
collateral_amount_raw	VARCHAR(16777216)	Raw collateral amount (unnormalized).
token_amount_raw	VARCHAR(16777216)	Raw share amount (unnormalized).
price_str	VARCHAR(16777216)	Share price as a string.
price	FLOAT	Share price as a float.
maker_amount_raw	VARCHAR(16777216)	Raw amount provided by the maker.
taker_amount_raw	VARCHAR(16777216)	Raw amount provided by the taker.
fee_charged_raw	VARCHAR(16777216)	Per-row gross fee, allocated from the per-tx FeeCharged sum across kept maker-side trade rows weighted by collateral_amount. SUM(fee_charged_raw) over rows in a tx equals the per-tx total exactly. Sources: V1 std + V1 NegRisk + V2 std + V2 NegRisk FeeCharged decoded models.
fee_rebate_raw	VARCHAR(16777216)	Per-row maker rebate, allocated from the per-tx FeeRefunded sum across kept maker-side trade rows weighted by collateral_amount. 100% of rebates go to the order’s maker (verified on 15.4M events). V2 has no rebate — 0 for V2-only txs. Sources: V1 Module 1 legacy + V1 Module 2 std + V1 Module 2 NegRisk FeeRefunded decoded models.
fee_raw	VARCHAR(16777216)	Raw fee amount.
fee_charged_event_count	NUMBER(38,0)	Count of FeeCharged events in the transaction. Same value across all rows in the tx.
fee_rebate_event_count	NUMBER(38,0)	Count of FeeRefunded events in the transaction. Same value across all rows in the tx. 0 for V2-only txs.
collateral_token_address	VARCHAR(42)	Address of the collateral token.
collateral_token_name	VARCHAR(16777216)	Name of the collateral token.
collateral_token_symbol	VARCHAR(16777216)	Symbol of the collateral token.
collateral_token_decimals	NUMBER(38,0)	Decimals of the collateral token.
collateral_amount	VARCHAR(16777216)	Collateral amount (normalized).
fee_charged_amount	VARCHAR(16777216)	Normalized per-row gross fee (divided by collateral decimals).
fee_rebate_amount	VARCHAR(16777216)	Normalized per-row maker rebate (divided by collateral decimals).
fee_amount	VARCHAR(16777216)	Fee amount (normalized).
usd_exchange_rate	FLOAT	USD exchange rate of the collateral token at the time of trade.
usd_collateral_amount	FLOAT	Collateral amount in USD.
usd_fee_charged_amount	FLOAT	USD value of the per-row gross fee.
usd_fee_rebate_amount	FLOAT	USD value of the per-row maker rebate.
usd_fee_amount	FLOAT	Fee amount in USD.
transaction_index	NUMBER(38,0)	Index of the transaction in the block.
transaction_hash	VARCHAR(66)	Hash of the transaction.
log_index	NUMBER(38,0)	Log index of the trade event.
block_timestamp	TIMESTAMP_NTZ(9)	Timestamp of the block containing the trade.
block_number	NUMBER(38,0)	Block number of the trade.
block_hash	VARCHAR(66)	Hash of the block containing the trade.
unique_id	VARCHAR(16777216)	Concatenated identifier for the trade.
_created_at	TIMESTAMP_NTZ(9)	The timestamp of the creation of the trade.
_updated_at	TIMESTAMP_NTZ(9)	The timestamp of the last update of the trade.

Sample Queries

Recent Trades
Daily Volume
Combos v3

Get recent prediction market trades with essential details:

select
  event_name,
  project,
  price,
  usd_collateral_amount,
  maker,
  taker,
  block_timestamp
from polygon.predictions.trades
where block_timestamp >= current_timestamp - interval '3 days'
order by block_timestamp desc

Calculate daily trading volume and trading activity for the past month:

select
  date(block_timestamp) as day,
  count(*) as total_trades,
  sum(usd_collateral_amount) as daily_volume_usd
from polygon.predictions.trades
where block_timestamp >= current_timestamp - interval '30 days'
group by day
order by day desc

Combos fills (contract_version = 'v3'):

select
  contract_version,
  exchange_type,
  condition_id,
  asset_id,
  price,
  usd_collateral_amount,
  block_timestamp
from polygon.predictions.trades
where contract_version = 'v3'
  and block_timestamp >= current_timestamp - interval '7 days'
order by block_timestamp desc

Protocol Coverage

Protocol	Status	Exchange Types Supported
Polymarket	Active	Fixed Product Market Maker, Orderbook, ComboExchange

More protocols will be added as they launch or integrate with Polygon.

Overview

Data Verticals

Solana

Hyperliquid

Lighter

Bitcoin Ecosystem

EVM

Cosmos Ecosystem

Move Ecosystem

Other Ecosystems

Table Details

Table Columns

Sample Queries

Protocol Coverage

​Table Details

​Table Columns

​Sample Queries

​Protocol Coverage

Table Details

Table Columns

Sample Queries

Protocol Coverage