How to Get a Daily List of Predicted Stock Movers via API

A short, practical guide to pulling Alphanume's Next-Day Movers feed into your own scanner or backtest.

You already know the kind of signal you want: a daily, pre-computed list of stocks likely to move hard next session, delivered as data you can program against rather than a screen you read by eye. This guide shows how to get that from the Alphanume Next-Day Movers API and what to do with it once you have it.

The endpoint

Everything runs through a single GET request:

GET https://api.alphanume.com/v1/next-day-movers

Authenticate with your key as a query parameter (?api_key=alp_your_key) or with an X-API-Key header. The response is JSON shaped as { "count": N, "data": [ ... ] }, with one object per ranked ticker.

Pulling today's list in Python

import requests

r = requests.get(
    "https://api.alphanume.com/v1/next-day-movers",
    params={"api_key": "alp_your_key"},
)
movers = r.json()["data"]
for row in movers:
    print(row["date"], row["ticker"])

That is the whole integration. Each row carries date, ticker, return, and absolute_move. The model runs daily at 3:30 PM ET, so a single morning call gives you the current watchlist.

Filtering by date for backtests

Because the dataset is point-in-time, you can reconstruct the exact list that was published on any historical date and test it honestly. Most endpoints accept date plus the range filters date_gte, date_lte, date_gt, and date_lt:

GET https://api.alphanume.com/v1/next-day-movers?date_gte=2026-05-01&date_lte=2026-05-31&api_key=alp_your_key

Pull a month of history, join the predicted tickers to the realized return and absolute_move columns, and you can measure hit rate and average move size directly. No survivorship bias and no lookahead, because the prediction was frozen before the outcome was known.

A simple workflow

• Each morning, call the endpoint and take the top names as your movement watchlist.
• Cross-reference against your own liquidity and borrow filters.
• Size positions using realized volatility so a 2 percent name and a 12 percent name are not treated the same.
• Periodically re-pull history and re-check that the signal still earns its place.

Access and limits

Next-Day Movers is on the free tier as a rolling 30-day delayed window, with the current session reserved for Pro. Rate limits on Pro run to 600 requests per minute, and over-limit calls return HTTP 429. Grab a free Alphanume API key and read the API documentation to get started, or explore it on Alphanume first.

Where it fits in a daily workflow

The feed slots in at the top of a morning routine. A single call returns the day's ranked names, which you treat as a movement watchlist rather than a set of trade signals on their own. From there you apply your own liquidity, borrow, and event filters, and size positions by realized volatility so a quiet name and a jumpy one are not treated the same.

Because every historical list is point-in-time, the same endpoint doubles as a backtest source. Pull a few months with the date range filters, join the predicted tickers to their realized return and absolute move, and you can measure hit rate and average move size before you ever risk capital on the live feed.

Frequently asked questions

What is the endpoint for predicted movers?

It is GET https://api.alphanume.com/v1/next-day-movers. Pass your key as ?api_key=alp_your_key or an X-API-Key header. The response is JSON shaped as { "count": N, "data": [ ... ] }.

How do I pull only a date range?

Use the range filters date_gte, date_lte, date_gt, and date_lt, in any logically valid combination. Dates always use the YYYY-MM-DD format.

How many requests can I make?

The Pro tier allows 600 requests per minute. Going over the limit returns an HTTP 429 response, so back off and retry when you hit it.

What fields come back per ticker?

Each row carries date, ticker, return, and absolute_move. The two outcome fields are null for the most recent date until that session resolves.

Do I need a paid plan to start?

No. A free key gives you a rolling 30-day delayed window, which is enough to wire up the integration and validate the signal before upgrading to Pro for current-session data.