---
title: Introduction
description: "Introduction to the statistical methods Statsig uses to analyze experiments, including hypothesis testing, confidence intervals, and variance reduction."
product: general
lang: en
last_updated: 2025-09-18
token_estimate: 729
---
# Introduction

> For AI agents: a documentation index is available at [/llms.txt](/llms.txt). Append `.md` to any page URL for markdown, or send `Accept: text/markdown`.

The Statsig Stats Engine is the set of statistical calculations behind the experiment results you read in [Pulse](https://docs.statsig.com/experiments/interpreting-results/read-results). It turns exposure and metric data into a metric delta, confidence interval, and p-value for each metric, applies variance reduction, and supports frequentist, SPRT, and Bayesian analysis methods. Use these pages when you need the formula or assumption behind a number in your results.

## What the Stats Engine calculates and how

- [Metric deltas](https://docs.statsig.com/experiments/statistical-methods/metric-deltas): The difference in mean metric value between the test and control groups, in absolute and relative terms.
- [Standard error and mean variance](https://docs.statsig.com/experiments/statistical-methods/variance): How much each group's mean would vary across repeated samples. This is the input to every confidence interval and p-value.
- [Confidence intervals](https://docs.statsig.com/experiments/statistical-methods/confidence-intervals): The gray, red, and green bars in Pulse, which represent the uncertainty in each metric delta.
- [p-values](https://docs.statsig.com/experiments/statistical-methods/p-value): The probability of observing a metric delta at least as extreme as the one measured, if the treatment has no effect on that metric.
- [Topline and projected impact](https://docs.statsig.com/experiments/statistical-methods/topline-impact): The daily effect of the experiment on a company metric, and the estimated effect if you launch the test group to all users.
- [Variance reduction](https://docs.statsig.com/experiments/statistical-methods/variance-reduction): The techniques Statsig applies so the same effect reaches significance with fewer users.
- [CUPED](https://docs.statsig.com/experiments/statistical-methods/methodologies/cuped): Uses each user's pre-exposure data to reduce variance. Statsig applies it automatically.
- [Pre-experiment bias](https://docs.statsig.com/experiments/statistical-methods/pre-experiment-bias): How Statsig detects groups that already differed before exposure and flags the affected metrics.
- [Frequentist sequential testing](https://docs.statsig.com/experiments/advanced-setup/sequential-testing): Adjusts p-values and confidence intervals so you can read results early without inflating false positives.
- [Sequential probability ratio tests](https://docs.statsig.com/experiments/advanced-setup/sprt): Compares the likelihood of your data under the alternative and null hypotheses, and can accept either one.
- [Bayesian experiments](https://docs.statsig.com/experiments/advanced-setup/bayesian): Reports chance to beat control and expected loss, with optional informative priors.