What is a logistic regression probability?

Logistic regression models the probability that an outcome is 1 (e.g., disease present, click, churn). It maps a linear combination of predictors through the sigmoid function so the output is always between 0 and 1.

What is log-odds (logit)?

Log-odds is the natural logarithm of the odds: ln(P / (1-P)). In logistic regression, the linear predictor b0 + b1*x1 + ... equals the log-odds. Positive values mean P > 0.5; negative values mean P < 0.5.

How is the sigmoid function applied?

P = 1 / (1 + e^(-z)) where z is the log-odds. This squashes any real number into (0, 1), giving a valid probability.

What does an odds ratio represent?

The odds ratio for a predictor equals e^(coefficient). It tells you how much the odds of the outcome multiply for a one-unit increase in that predictor, holding others constant.

Can I use this for multi-class models?

No. This calculator is for binary logistic regression (one outcome class vs. another). For multinomial or ordinal logistic regression you would need separate log-odds per class.

Logistic Regression Probability Calculator

Name: Logistic Regression Probability Calculator
Availability: InStock
Author: Nham Vu

Enter log-odds (or build a linear combination from coefficients) to get the predicted probability from a logistic regression model.

Model Inputs

Enter the intercept, then add as many predictors as needed.

Intercept (b₀)

Or enter a known log-odds directly

Results

— P(Y=1)

Log-odds (logit) —

Odds (e^log-odds) —

Probability P(Y=1) —

Probability P(Y=0) —

Linear combination

—

Quick interpretation guide

z = 0P = 0.50 (decision boundary)

z = 2P ≈ 0.88 (strong positive signal)

z = -2P ≈ 0.12 (strong negative signal)

z = 4P ≈ 0.98 (near-certain positive)

Summary

Enter log-odds (or build a linear combination from coefficients) to get the predicted probability from a logistic regression model.

How it works

Enter the intercept and one or more predictor coefficients with their values.
The tool computes the linear combination: log-odds = b0 + b1*x1 + b2*x2 + ...
The sigmoid function converts log-odds to a probability: P = 1 / (1 + e^(-log-odds)).
Read the probability, odds ratio, and log-odds directly from the results panel.
Add or remove predictors dynamically to model any number of variables.

Use cases

Verify logistic regression model predictions by hand.
Explore how changing a coefficient shifts the predicted probability.
Teach the relationship between log-odds, odds, and probability.
Quick sanity-check during binary classification model development.
Convert a known log-odds score to an interpretable probability.
Compare probabilities for different covariate combinations.

Frequently Asked Questions

Last updated: 2026-06-11 · Reviewed by Nham Vu