Filter whale prints from the raw trade tape in Python.

Fetch and filter

One call returns the recent tape. is_buyer_maker=true means the seller crossed the spread (aggressive sell); false means an aggressive buy. Multiply price by quantity, keep everything above your threshold, and aggregate by side.

import pandas as pd
import requests

BASE = "https://www.getoptiom.com/api/tradinghub"
SYMBOL = "BTCUSDT"
MIN_USD = 1_000_000  # whale threshold

res = requests.get(
    f"{BASE}/trades",
    params={"symbol": SYMBOL, "max_rows": 50_000},
    timeout=60,
)
res.raise_for_status()

df = pd.DataFrame(res.json()["rows"])
df["ts"] = pd.to_datetime(df["ts"], utc=True, format="mixed")
df["usd"] = df["price"] * df["qty"]
# is_buyer_maker=True means the SELLER hit the bid -> aggressive sell
df["side"] = df["is_buyer_maker"].map({True: "sell", False: "buy"})

whales = df[df["usd"] >= MIN_USD].sort_values("ts")
print(f"{len(whales)} whale prints >= {MIN_USD:,} USD out of {len(df)} trades "
      f"({df['ts'].min():%H:%M} → {df['ts'].max():%H:%M} UTC)\n")

print("Aggressive flow (whales only):")
print(whales.groupby("side")["usd"].agg(["count", "sum"]).round(0))

print("\nLast 10 whale prints:")
print(whales.tail(10)[["ts", "side", "price", "qty", "usd"]].round(2).to_string(index=False))

# net whale pressure over the window
net = whales.loc[whales["side"] == "buy", "usd"].sum() - whales.loc[whales["side"] == "sell", "usd"].sum()
print(f"\nNet whale pressure: {'+' if net >= 0 else ''}{net:,.0f} USD")

Reading the output

Repeated same-size prints seconds apart (like the 79.9 BTC clips in the sample output) are an execution algo slicing a large parent order — one player, not many. Net pressure over a window tells you which way size is leaning; it is most useful at levels you already care about, like the walls from the depth heatmap or the hot zones from liquidations.

Going further

Poll the endpoint every minute, keep a rolling window, and alert on prints above $5M or on net pressure crossing a threshold — the same pattern as the Telegram alerts tutorial. On Pro history you can join whale flow with subsequent returns and measure whether size actually predicts anything on your pairs.