---
title: Compare Classification Thresholds Without Hiding Error-Cost Tradeoffs — Free
description: Two thresholds can trade false positives for false negatives while their accuracy percentages look reassuringly similar. Use a fixed classroom dataset and an explicit cost assumption to inspect that tradeoff instead of calling one threshold best without a criterion.
category: learning
last_verified: 2026-10-01
freshness: green
published: 2026-10-01
canonical_url: https://gratis-ai.pages.dev/articles/classifier-confusion-cost-ledger.html
companion_app_url: https://gratis-ai.pages.dev/apps/classifier-confusion-cost-ledger.html
reading_time_minutes: 10
tags: learning, confusioncost, local-first, explicit-assumptions, no-account
---

# Compare Classification Thresholds Without Hiding Error-Cost Tradeoffs — Free

Two thresholds can trade false positives for false negatives while their accuracy percentages look reassuringly similar. Use a fixed classroom dataset and an explicit cost assumption to inspect that tradeoff instead of calling one threshold best without a criterion.

**Free and complete.** This guide is a standalone free tool — no account, no card, anywhere. The only paid option is one optional extra outcome: $4 one-time. Nothing in the free tier is gated.

## Who it's for

Learners checking classification evaluation tradeoffs

## What's free

- one complete unwatermarked confusion-cost board run in the app
- three labeled samples and unlimited article-demo calculations
- complete result text and a user-triggered input-state link

## Optional paid add-on

**$4 one-time.**

- unlimited subsequent confusion-cost board app calculations after a local convenience signal
- print the confusion-cost board with visible assumptions and its safety boundary
- CSV of the displayed cost units per 100 records and row workings; conditional estimate uses your manual-retyping minutes
- five local reading summaries (headline/count/time), not saved full inputs or cross-device backups

## Sources & verification

- Tool: GratisAI local-first instrument (no key, no account)
- Primary source: the free tool itself (local-first, no external dependency)
- Last verification: 2026-10-01 (green)

## Use it

- Interactive guide: https://gratis-ai.pages.dev/articles/classifier-confusion-cost-ledger.html
- Companion app: https://gratis-ai.pages.dev/apps/classifier-confusion-cost-ledger.html
- This machine-readable mirror: /mirrors/classifier-confusion-cost-ledger.md

*Generated free at https://gratis-ai.pages.dev/articles/classifier-confusion-cost-ledger.html — every guide on GratisAI is a complete, working tool. $0, no account, no card, anywhere.*