Find the percentage decrease between an old value and a new, smaller value.
Find the percentage decrease between an old value and a new, smaller value.
Enter values above and click Calculate — results will appear here with the formula explained.
Percentage decrease measures how much a value fell relative to where it started. Subtract the new value from the original, divide by the original, then multiply by 100. Using the absolute value of the original handles negative baselines (common in profit/loss and temperature contexts) without sign confusion.
The denominator must stay the original. A price cut from $200 to $150 is a 25% decrease, but $50 divided by the new $150 gives 33% — the same dollars, wrong reference. Always anchor to the pre-change value, which is why this calculator shows the original as divisor and labels increases separately when the new value is higher.
From zero is undefined, not 0% or infinite in any useful sense: every positive number is infinitely larger than zero as a percentage, which conveys no decision information. When a baseline is zero (first sale, new metric), quote the absolute drop — for example '$0 to $10,000 then back to $7,000' — and resume percentages once the base is non-zero.
Sequential decreases multiply, never add: 20% off then 10% off is not 30% off. $100 → $80 → $72, a 28% total fall (0.8 × 0.9 = 0.72). Chain the factors instead of summing them; compound losses like inflation over years or repeated markdowns all compound multiplicatively, and adding them understates the true decline — sometimes dramatically.
Decrease versus percentage-point decrease is the headline trap behind most misleading stats: a rate falling from 30% to 20% is a 10-point drop but a 33% relative fall (10 ÷ 30). Polls, unemployment and discount claims live on this distinction — always check whether a claim uses points or percent, and anchor every 'down X%' to its money or count base before acting.
Use percentage decrease for discounts, shrinkage, salary cuts and portfolio losses, but pair it with absolute change for context: a $2 drop from $10 is 20% but $2, while a $2 drop from $1,000 is 0.2% and barely material. Reporting both prevents the small-base inflation where tiny absolutes look dramatic as percentages. In finance, drawdowns are quoted as percentage decreases from peaks for this reason — a $10,000 portfolio falling to $8,500 is a 15% drawdown ($1,500), while the same $1,500 on a $100,000 peak is 1.5% and an entirely different risk story.
Find the percentage decrease between an old value and a new, smaller value. Formula: decrease % = ((original − new) ÷ |original|) × 100. Example: A TV drops from $400 to $340: the fall is $60 and 60 ÷ 400 = 0.15, so the discount is exactly 15%.
Percentage decrease measures how much a value fell relative to where it started. Subtract the new value from the original, divide by the original, then multiply by 100. Using the absolute value of the original handles negative baselines (common in profit/loss and temperature contexts) without sign confusion.
The denominator must stay the original. A price cut from $200 to $150 is a 25% decrease, but $50 divided by the new $150 gives 33% — the same dollars, wrong reference. Always anchor to the pre-change value, which is why this calculator shows the original as divisor and labels increases separately when the new value is higher.
From zero is undefined, not 0% or infinite in any useful sense: every positive number is infinitely larger than zero as a percentage, which conveys no decision information. When a baseline is zero (first sale, new metric), quote the absolute drop — for example '$0 to $10,000 then back to $7,000' — and resume percentages once the base is non-zero.
Sequential decreases multiply, never add: 20% off then 10% off is not 30% off. $100 → $80 → $72, a 28% total fall (0.8 × 0.9 = 0.72). Chain the factors instead of summing them; compound losses like inflation over years or repeated markdowns all compound multiplicatively, and adding them understates the true decline — sometimes dramatically.
Decrease versus percentage-point decrease is the headline trap behind most misleading stats: a rate falling from 30% to 20% is a 10-point drop but a 33% relative fall (10 ÷ 30). Polls, unemployment and discount claims live on this distinction — always check whether a claim uses points or percent, and anchor every 'down X%' to its money or count base before acting.
Use percentage decrease for discounts, shrinkage, salary cuts and portfolio losses, but pair it with absolute change for context: a $2 drop from $10 is 20% but $2, while a $2 drop from $1,000 is 0.2% and barely material. Reporting both prevents the small-base inflation where tiny absolutes look dramatic as percentages. In finance, drawdowns are quoted as percentage decreases from peaks for this reason — a $10,000 portfolio falling to $8,500 is a 15% drawdown ($1,500), while the same $1,500 on a $100,000 peak is 1.5% and an entirely different risk story.
A TV drops from $400 to $340: the fall is $60 and 60 ÷ 400 = 0.15, so the discount is exactly 15%. A two-step sale shows compounding: $400 → $320 (20% off) → $288 (10% more off) — the second step is 32 ÷ 320 = 10%, but the total from $400 is 112 ÷ 400 = 28%, not 30%. The reverse check: rising from $340 back to $400 is (60 ÷ 340) × 100 = 17.6% increase, revealing the asymmetry — −15% down is not undone by +15% up.
Formulas are standard public references (see our methodology). External standards are cited in the text where they apply.
Last reviewed: September 2026 · Report an error