Next-Day Movers: A Practical Guide

The Problem Every Volatility Trader Knows

Every evening, thousands of traders scan for the same thing: which stocks are going to move tomorrow?

Most of them rely on earnings calendars, news feeds, or gut instinct. Some build screens around unusual options activity or implied volatility rank. But the fundamental question is simple — out of thousands of liquid equities, which ones are most likely to deliver outsized realized movement in the next session?

That's exactly what the Next-Day Movers dataset answers.

What This Dataset Actually Does

Each trading day at 3:30 PM ET, a predictive model evaluates a universe of liquid, optionable U.S. equities and selects the names most likely to experience large price moves the following session.

The model ranks candidates using features rooted in volatility structure — implied volatility levels, realized volatility behavior, and term structure dynamics. The output is a short list of tickers: the highest-conviction next-day movement candidates.

Once published, these selections are stored point-in-time. They're never changed after the fact. The next trading session plays out, and the realized return is recorded alongside the original selection.

That means you get two things in one dataset:

1. A daily signal — names the model expects to move big tomorrow

2. A built-in scorecard — the actual realized returns, so you can study whether the model's picks delivered

Why Implied Volatility Tells You Something About Tomorrow

The core intuition here comes from the relationship between implied and realized volatility.

Implied volatility is the market's best estimate of how much a stock will move over a given period. It's derived from option prices — when options are expensive, implied vol is high, and the market is pricing in a big move.

Realized volatility is what actually happens. It's the observed magnitude of price changes after the fact.

These two quantities are related, but they don't always agree. When implied volatility is elevated relative to recent realized behavior — or when the term structure of implied vol exhibits specific patterns — it often signals that the market is bracing for something. A catalyst is expected. Positioning is asymmetric. Liquidity providers are demanding wider premiums.

The Next-Day Movers model uses this kind of information (among other features) to surface the names where the probability of a large move is highest — regardless of direction.

This is important: the dataset is non-directional. It doesn't predict whether a stock will go up or down. It predicts whether a stock will move a lot. That distinction opens up a different category of trading strategies.

How Traders Use This

Volatility selling with an edge

If you sell options — straddles, strangles, iron condors — your biggest risk is the underlying moving more than the premium you collected. The Next-Day Movers dataset lets you identify which names the model thinks will move most, so you can either:

• Avoid selling premium on those names (risk management)

• Sell premium on names the model did NOT select, where realized movement may be lower than what's priced in

Directional bets on movement

If you have a directional view and want to maximize the probability that the stock actually moves enough to make the trade worthwhile, this dataset gives you a short list of candidates where large movement is most likely. Pair it with your own directional signal and you've got a combined edge: you think it's going up, and the model thinks it's going to move big.

Straddle/strangle buying

The simplest application: buy a straddle or strangle on the selected names. You're paying for vol — the question is whether realized movement exceeds the implied move priced into the options. The dataset gives you historically stored selections and realized outcomes, so you can backtest whether this approach has been profitable, and under what conditions.

Screening and prioritization

Even if you don't trade based on the model directly, it's a powerful daily filter. Instead of scanning 4,000 tickers every morning, you get a focused list of names where something interesting is likely to happen. That's valuable even for discretionary traders.

What the Data Looks Like

A typical response:

{
  "date": "2026-03-09",
  "ticker": "FIGR",
  "return": 21.29,
  "absolute_move": 21.29
}

• date — the day the model made the selection

• ticker — the equity selected

• return — the signed return in the following trading session (positive = up, negative = down)

• absolute_move — the magnitude of that return, regardless of direction

For the most recent selection, return and absolute_move will be null — because the next session hasn't happened yet. Once it closes, the values populate and remain fixed.

Backtesting Without Lookahead Bias

One of the most important properties of this dataset is that it's stored point-in-time.

The selections you see for March 9th are exactly what the model produced on March 9th. They weren't adjusted on March 10th after seeing what happened. The realized returns are appended after the fact, but the selections themselves are frozen at publication time.

This means you can use this dataset in systematic backtests without worrying about lookahead bias — the selections were made with only information available at the time.

Key Details

Getting Started

import requests

url = "https://api.alphanume.com/v1/next-day-movers"
params = {
    "date_gte": "2026-03-01",
    "api_key": "your_api_key"
}

r = requests.get(url, params=params)
data = r.json()

for obs in data["data"]:
    print(f"{obs['date']} | {obs['ticker']} | return: {obs['return']}%")

You can filter by single date, date ranges, or pull the full history. See the API documentation for complete parameter details.

The Bottom Line

Markets price in expected movement through implied volatility. But some names are more likely to deliver on that expectation than others. The Next-Day Movers dataset identifies those names daily — giving you a focused, model-driven list of the stocks most likely to experience outsized moves in the next session.

Whether you use it to size positions, select strikes, avoid risk, or build a systematic vol strategy, the core value is the same: knowing where movement is most likely to show up, before it does.