Guides / API Guide

ProbSights API Rate Limiting: Best Practices

Rate limits are per-day and per-key. Free keys get a small daily allowance; paid plans raise it substantially, and the exact ceiling for your plan is reported as dailyLimit on your dashboard alongside requests used this month and how many were rate-limited. Check that number rather than assuming one.

Where the requests actually go

Most accounts that hit a limit are not pulling too much data — they are making too many requests for the same data. In rough order of impact:

Pattern	Fix
Re-fetching the category or market list every run	Cache it; it changes slowly
One request per timestamp	Pull the range once, index locally
Narrow windows in a loop (per-minute, per-hour)	One wide range plus `pagination_key`
Re-downloading history you already have	Persist it; settled markets never change
Polling for updates on a fixed timer	Use webhooks where available

Widen the range, don't multiply the calls

A day of 1m snapshots is ~1,440 rows — two requests at limit=1000. Fetching it hour by hour is 24 requests for the same rows, and risks gaps at every seam. Always prefer the widest range you need with proper cursor paging.

Back off on 429

import time, random

def get_with_retry(path, attempts=5, **params):
    for i in range(attempts):
        res = session.get(f"{BASE}{path}", params=params)
        if res.status_code != 429:
            res.raise_for_status()
            return res.json()
        # Exponential backoff with jitter — a fixed sleep re-synchronizes
        # parallel workers into the same retry spike.
        time.sleep(min(2 ** i, 30) + random.random())
    raise RuntimeError("rate limited after retries")

Honor Retry-After when present, and cap concurrency: parallel workers share one key's budget, so eight threads exhaust it eight times faster.

Cache with the right lifetime

Settled markets — permanent. Once resolved, history never changes.
Category and series lists — hours.
Active market prices — seconds to minutes, matched to your strategy.

Related guides

Data from Kalshi and Polymarket. Search live API docs Pricing