Crypto Price Data Guide

Crypto Price Data Guide

Where do crypto prices actually come from? This guide breaks down the entire data pipeline — from raw exchange data to the numbers you see in apps.

The Price Data Pipeline

┌──────────────────────────────────────────────────────────┐
│  LAYER 1: Primary Sources (where prices originate)       │
│                                                          │
│  CEX Order Books          DEX AMM Pools                  │
│  ┌──────────────┐        ┌──────────────┐                │
│  │ Binance      │        │ Uniswap V3   │                │
│  │ Coinbase     │        │ Curve        │                │
│  │ Kraken       │        │ Camelot      │                │
│  │ OKX, Bybit   │        │ Balancer     │                │
│  └──────┬───────┘        └──────┬───────┘                │
│         │                       │                        │
├─────────┼───────────────────────┼────────────────────────┤
│  LAYER 2: Oracle Networks (aggregate + post on-chain)    │
│         │                       │                        │
│         ▼                       ▼                        │
│  ┌──────────────────────────────────────┐                │
│  │ Chainlink     Pyth     Redstone      │                │
│  │ (median of multiple data sources)    │                │
│  └──────────────────┬───────────────────┘                │
│                     │                                    │
├─────────────────────┼────────────────────────────────────┤
│  LAYER 3: Aggregators (combine everything)               │
│                     ▼                                    │
│  ┌──────────────────────────────────────┐                │
│  │ CoinGecko  DeFi Llama  CoinMarketCap │                │
│  │ DexScreener  CoinCap   CoinPaprika   │                │
│  └──────────────────┬───────────────────┘                │
│                     │                                    │
├─────────────────────┼────────────────────────────────────┤
│  LAYER 4: Consumer Apps                                  │
│                     ▼                                    │
│  Portfolio trackers, trading bots, DeFi protocols,       │
│  AI agents (like SperaxOS), wallets                      │
└──────────────────────────────────────────────────────────┘

Layer 1: Primary Price Sources

Centralized Exchange (CEX) APIs

CEXes are the dominant price source for high-cap tokens. Their order books determine the "true" price through supply/demand matching.

All of these are free, no API key required for public market data.

Example: Binance Spot Price

GET https://api.binance.com/api/v3/ticker/price?symbol=BTCUSDT

Response: { "symbol": "BTCUSDT", "price": "64523.45000000" }

Example: Binance 24h Ticker

GET https://api.binance.com/api/v3/ticker/24hr?symbol=ETHUSDT

Response: {
  "symbol": "ETHUSDT",
  "priceChange": "45.20",
  "priceChangePercent": "1.42",
  "lastPrice": "3245.67",
  "volume": "584231.45",
  "highPrice": "3289.00",
  "lowPrice": "3178.23"
}

DEX On-Chain Data

For tokens that aren't listed on CEXes (the long tail), DEX pools are the only price source.

Reading DEX prices:

1. Direct RPC — Call pool contracts to read reserves/sqrtPrice
2. The Graph — Query indexed swap events via GraphQL
3. DexScreener API — Aggregated DEX data across 80+ chains

DexScreener Example

GET https://api.dexscreener.com/latest/dex/tokens/0xD74f5255D557944cf7Dd0E45FF521520002D5748

Returns: SPA pair data across all DEXes

How CEX and DEX Prices Stay in Sync

Arbitrage bots. If ETH is $3,200 on Binance but $3,210 on Uniswap:

1. Bot buys on Binance (cheaper)
2. Bot sells on Uniswap (expensive)
3. Price difference narrows until it's less than gas + bridge costs

This is why prices across venues are usually within 0.1–0.5% of each other.

Layer 2: Oracle Networks

Oracles aggregate CEX + DEX data and make it available on-chain for smart contracts.

Chainlink

The dominant oracle network. ~1000+ price feeds across EVM chains.

How it works:

1. Multiple independent oracle nodes fetch prices from CEXes and DEXes
2. Each node submits its observation to the aggregator contract
3. The contract takes the median of all observations
4. Updates when: deviation > threshold (typically 0.5–1%) OR heartbeat time passes

Reading a Chainlink feed on-chain:

// ETH/USD on Ethereum: 0x5f4eC3Df9cbd43714FE2740f5E3616155c5b8419
// SPA/USD on Arbitrum: Available via Chainlink

function getPrice() {
    (, int256 price,,,) = priceFeed.latestRoundData();
    return uint256(price); // 8 decimals
}

Key feeds:

Sperax Context: USDs uses Chainlink price feeds via its MasterPriceOracle to value collateral (USDC, USDT) and maintain its peg.

Pyth Network

Pull oracle — prices update sub-second but consumers must "pull" the latest update on-chain.

GET https://hermes.pyth.network/api/latest_price_feeds?ids[]=0xff61491a931112ddf1bd8147cd1b641375f79f5825126d665480874634fd0ace

Returns: { price: "6452345000000", expo: -8, conf: "2500000" }

Chainlink vs Pyth

Layer 3: Free Aggregator APIs

These combine CEX + DEX + oracle data into convenient APIs.

DeFi Llama (Best Free Option)

Zero rate limiting. Zero API key. Most generous free API in crypto.

Price query example:

GET https://coins.llama.fi/prices/current/arbitrum:0xD74f5255D557944cf7Dd0E45FF521520002D5748

Returns: USDs price from on-chain sources

CoinCap

Simple, clean API for top 2000 coins.

GET https://api.coincap.io/v2/assets?limit=10

Returns: Top 10 coins by market cap with price, volume, change

CoinPaprika

Decent free tier with ATH data and social links.

GET https://api.coinpaprika.com/v1/tickers/btc-bitcoin

Returns: Price, ATH, ATH date, volume, market cap, beta, supply

CryptoCompare

Good for OHLCV (candlestick) data.

GET https://min-api.cryptocompare.com/data/v2/histoday?fsym=BTC&tsym=USD&limit=30

Returns: 30 days of daily OHLCV data

Building Without CoinGecko

A practical multi-source approach:

Waterfall strategy (what production apps use):

1. Try Binance (fastest, most liquid)
2. Fall back to CoinCap (broader coverage)
3. Fall back to DexScreener (on-chain tokens)
4. Fall back to DeFi Llama (universal)

SperaxOS uses a similar multi-provider waterfall: Binance → CoinCap → DexScreener → CoinGecko for maximum reliability.

Price Calculation Methods

VWAP (Volume-Weighted Average Price)

How aggregators compute a single price from multiple exchanges:

VWAP = Σ(Price_i × Volume_i) / Σ(Volume_i)

Example:

VWAP = (3200×2B + 3205×500M + 3198×100M) / (2B + 500M + 100M)
     = $3,200.96

TWAP (Time-Weighted Average Price)

Used by Uniswap V3 oracle and some protocols:

TWAP = Average price over a time window (e.g., last 30 minutes)

More resistant to manipulation than spot price.

Agent Tips

1. Never rely on a single source — always implement fallbacks
2. CEX data is freshest for major tokens — use Binance/Coinbase first
3. DEX data is essential for tokens not listed on CEXes
4. DeFi Llama is the best free API — zero rate limiting, comprehensive
5. Chainlink feeds are the gold standard for on-chain price verification
6. Always show the data source — users should know where prices come from

Links

• DeFi Llama API: https://defillama.com/docs/api
• Binance API: https://binance-docs.github.io/apidocs/
• Chainlink Feeds: https://data.chain.link
• DexScreener: https://docs.dexscreener.com
• CoinCap: https://docs.coincap.io
• Pyth Network: https://pyth.network
• Sperax (uses Chainlink for USDs pricing): https://app.sperax.io