Article

HomeArticles › Percent Change

How to Calculate Percent Change Without Getting the Direction Wrong

Percent change is the arithmetic you use every day without thinking — sale prices, stock moves, poll swings, inflation numbers. It is also the arithmetic people get wrong more often than any other, because the formula is asymmetric and the denominator is invisible. Here is how to do it right, and the four traps that flip answers if you're not careful.

Last updated: May 4, 2026.

The formula

Percent change between an old value and a new value is:

Percent change = ((new − old) / old) × 100

Three things to notice:

  • The numerator is the raw difference. A positive numerator means an increase, negative means a decrease.
  • The denominator is always the old value — where you started, not where you ended.
  • The result is multiplied by 100 to turn a ratio into a percent. Forgetting the ×100 is how you end up reporting "0.5% growth" when you meant 50%.

Simple example: a price goes from $80 to $100. That's ((100 − 80) / 80) × 100 = (20 / 80) × 100 = +25%. The price rose by 25% of its original value.

Why the denominator matters: 100 → 150 is not the mirror of 150 → 100

Here's the mistake almost everyone makes: they assume percent change is symmetric. It isn't.

  • 100 → 150: ((150 − 100) / 100) × 100 = +50%
  • 150 → 100: ((100 − 150) / 150) × 100 = −33.3%

Same two numbers, but the drop is only 33.3% — not 50%. Why? Because when you go down from 150, you're measuring the $50 loss against a starting base of $150, not $100. A third of 150 is 50. A half of 150 would be 75.

This matters every time you see a stock or a price "recover" from a drop. A stock that falls 20% needs to rise 25% to get back to even, not 20%. A stock that falls 50% needs to double (+100%) to recover. The bigger the drop, the more asymmetric the math gets.

The thing most people get wrong: round-trip asymmetry

This is the one that trips up even people who are good at math. If an investment gains 50% and then loses 50%, where are you?

Not zero. You're down 25%.

Work it out on $100:

  • Start: $100
  • After +50%: $100 × 1.50 = $150
  • After −50%: $150 × 0.50 = $75

You lost $25 on a $100 start, which is a 25% drop. The same thing happens in reverse: a 50% loss followed by a 50% gain also leaves you at $75, not $100. Order doesn't matter; the asymmetry is baked into the multiplication.

The general rule: any +X% followed by −X% leaves you with 1 − (X/100)² of what you started with. For ±20% round trips you lose 4%. For ±30% you lose 9%. For ±50% you lose 25%. For ±100% — well, you can't recover from a 100% loss, so that one is obvious in the other direction.

This is why volatility is expensive even when the average return looks fine. A portfolio that goes +30% one year and −30% the next has an "average" return of 0%, but a real return of −9%.

Percentage points vs. percent: the headline trap

A percentage point is the absolute gap between two percentages. A percent is a relative change. They are not the same thing, and every political poll and interest-rate headline exploits the confusion.

Example one — interest rates. The Fed raises rates from 5% to 6%. That is:

  • 1 percentage point higher (6 − 5)
  • 20% higher relatively ((6 − 5) / 5 × 100)

Both are correct. A reporter who writes "rates jumped 20%" is technically right but dramatically misleading — readers will think rates went to 25%, not 6%.

Example two — polling. A candidate polls at 40% this week and 44% next week. That's a 4 percentage-point gain, or a 10% relative gain. If the headline says "Support surges 10%!" it sounds like the candidate went from 40 to 50. They didn't.

The fix: use "percentage point" or "pp" whenever you're describing the gap between two percentages. Reserve "percent" for relative changes in non-percentage quantities (prices, populations, votes in raw numbers).

When percent change misleads: the small-base problem

Percent change scales with the starting value. A $1 increase is a huge percentage when you're starting from $10 and nothing when you're starting from $1,000,000.

  • $10 → $11 is +10%
  • $1,000,000 → $1,000,100 is +0.01%

Same absolute change, wildly different percentages. This is why you should be suspicious whenever a press release leads with a huge percentage and buries the raw numbers. "Startup revenue grew 400%!" might mean $2,000 became $10,000. "Cases of the disease tripled" might mean 1 became 3.

The small-base problem also shows up in ratios of ratios. A cancer drug that cuts risk from 2% to 1% is a 50% reduction in relative risk, but only a 1 percentage-point reduction in absolute risk. Which one matters to a patient depends on how many patients you're comparing; both are true, and neither alone is enough.

Rule of thumb: if the base is under 100 of anything, demand the raw numbers before you take the percentage seriously.

Quick reference: common reversals

Memorize these. They come up constantly.

  • A 10% drop needs an 11.1% gain to recover.
  • A 20% drop needs a 25% gain.
  • A 25% drop needs a 33.3% gain.
  • A 33.3% drop needs a 50% gain.
  • A 50% drop needs a 100% gain (a double).
  • A 75% drop needs a 300% gain (a quadruple).
  • A 90% drop needs a 900% gain (a 10x).

The pattern: recovery from a drop of X needs a gain of X / (1 − X). It blows up fast as losses get bigger, which is the mathematical argument for not losing money in the first place.

A worked example: a sale on a sale

A jacket lists for $200. The store knocks 30% off, then runs a Black Friday promo of an additional 20% off the sale price. What's the final price, and what's the total discount?

Not 50% off. Work it in order:

  • Start: $200
  • After 30% off: $200 × 0.70 = $140
  • After another 20% off: $140 × 0.80 = $112

Total discount: ((200 − 112) / 200) × 100 = 44%, not 50%. Chained discounts always produce less than the sum of their parts, for the same reason round-trip gains and losses don't cancel: each percentage is applied to a smaller base than the last.

Frequently asked questions

Which value goes in the denominator?

Always the old (starting) value. If you put the new value in the denominator you'll get a different answer, and it won't be wrong by a constant factor — it'll be wrong differently depending on the direction of change. 100 to 150 divided by 150 gives 33.3%, which is the reverse calculation (150 → 100), not the forward one.

If I gain 10% every year for 10 years, am I up 100%?

No — you're up about 159%. Ten compounded 10% gains give 1.10¹⁰ ≈ 2.594, which is +159.4%, not +100%. Adding percentages only works when they're all applied to the same base; once the base changes each period, you need to multiply the growth factors.

How do I undo a percent change to get the original price?

Divide, don't subtract. If a price is $126 after a 5% markup, the original was 126 / 1.05 = $120. Subtracting 5% of $126 gives $119.70, which is 30 cents off — close, but that's exactly the kind of error that compounds when you chain calculations.

Can percent change be more than 100%?

Upward, absolutely. A value that triples has a +200% change; a tenfold increase is +900%. Downward, you cap at −100%, which means the value dropped to zero. A "120% decrease" is mathematically nonsense — you can't lose more than everything.

Should I use percent change or percentage-point change?

Depends on what the underlying number is. For raw quantities (dollars, people, units sold), use percent change. For quantities that are themselves percentages (interest rates, poll shares, unemployment rates), prefer percentage points to avoid ambiguity. When in doubt, report both.

Try it yourself

Run a few of your own numbers through the basic calculator. Pick two values you care about — a price last year and this year, a weight six months ago and now, a bank balance — and work out the percent change both forwards and backwards. Watching the denominator flip the answer makes the asymmetry stick in a way that reading about it never quite does.

This article is for general education. Numbers here are illustrative, not financial or medical advice.