Mental Math Tricks That Actually Work (and Why)

You have probably seen the videos. Someone multiplies two three-digit numbers in their head before the audience can pull up a calculator, and the comments fill with "I am bad at math, this is wizardry." It is not wizardry. The people doing it are using a small set of techniques that work because of how the number system is built. Once you see the structure, the magic disappears and the speed remains.

This article is not a list of party tricks. It is a short tour of the mental math shortcuts that actually pay off in daily life and on timed tests, with a brief explanation of why each one works. If you understand the why, you will remember them; if you only memorize the steps, you will forget by next Tuesday.

Why Mental Math Still Matters

In an era of phones and Desmos calculators, it is fair to ask whether mental math is worth the effort. The honest answer is yes, but for a different reason than your fourth-grade teacher gave you.

The point is not to outrun a calculator. It is to keep your working memory free during multi-step problems. Every time you have to stop and think hard about whether 7 times 8 is 54 or 56, you burn a slot in working memory that should be holding the structure of the larger problem. Students who are fluent in basic arithmetic solve harder problems faster, not because the arithmetic is faster, but because the harder problem stays intact in their head while they work on it.

This is the same reason fluent readers comprehend more than slow readers, even when both eventually decode every word. Fluency frees the mind for meaning.

Trick 1: Multiplying by 11 (Two-Digit Numbers)

To multiply any two-digit number by 11, add the two digits and put the sum between them.

• 23 times 11: split the 2 and the 3, add them (5), put it between: 253.
• 45 times 11: 4 plus 5 is 9, put it between: 495.
• 72 times 11: 7 plus 2 is 9: 792.

If the sum is 10 or more, carry the 1 to the leading digit.

• 67 times 11: 6 plus 7 is 13, write the 3, carry the 1: 6 plus 1 is 7, so 737.

Why it works: Multiplying by 11 is the same as multiplying by 10 and then adding the original number. 23 times 10 is 230. Plus 23 is 253. The "split and insert" trick is just a compact way to add a number to a shifted version of itself.

Trick 2: Squaring Numbers Ending in 5

For any number ending in 5, the square follows a fixed pattern. Take the digit (or digits) before the 5, multiply it by itself plus one, and append 25.

• 15 squared: 1 times 2 is 2, append 25: 225.
• 25 squared: 2 times 3 is 6, append 25: 625.
• 35 squared: 3 times 4 is 12, append 25: 1225.
• 65 squared: 6 times 7 is 42, append 25: 4225.

Why it works: Any number ending in 5 can be written as 10n plus 5. Squaring it gives 100n squared plus 100n plus 25, which factors into 100 times n times (n plus 1), plus 25. That is exactly the pattern: n times (n plus 1) goes in the hundreds place and above, with 25 appended at the end.

Trick 3: Multiplying Two Numbers Just Below 100

For two numbers close to 100, work out how far each is below 100, then combine.

• 96 times 97: the deficits are 4 and 3. Subtract one deficit from the other number (96 minus 3, or equivalently 97 minus 4) to get 93. Multiply the two deficits (4 times 3) to get 12. Stick them together: 9312.
• 98 times 95: deficits 2 and 5. 98 minus 5 is 93. 2 times 5 is 10. Result: 9310.

If the product of the deficits is a single digit, pad with a zero.

• 99 times 98: deficits 1 and 2. 99 minus 2 is 97. 1 times 2 is 2, padded to 02. Result: 9702.

Why it works: Write each number as 100 minus a deficit. The product expands into 10000 minus 100 times the sum of deficits, plus the product of deficits. The first two terms equal 100 times (one number minus the other deficit), which is the "subtract across" step. The last term is the appended product.

Trick 4: Percentages Are Commutative

This one is not a trick so much as a reframing, but it saves time constantly. The percent operator is symmetric: x percent of y equals y percent of x.

• 4 percent of 75 looks annoying. 75 percent of 4 is obviously 3.
• 18 percent of 50 looks annoying. 50 percent of 18 is 9.
• 8 percent of 25 is 25 percent of 8, which is 2.

When you face a percentage problem, ask whether swapping the two numbers makes one side trivially easy. Often it does.

Why it works: "X percent of Y" means X over 100 times Y. Multiplication does not care about order, so it equals Y over 100 times X, which is "Y percent of X."

Trick 5: Doubling and Halving

To multiply two numbers, you can double one and halve the other without changing the answer. This is gold when one of the numbers is awkward and the other is even.

• 16 times 25: halve 16 to get 8, double 25 to get 50. Now the problem is 8 times 50, which is 400.
• 14 times 35: halve 14 to get 7, double 35 to get 70. Now it is 7 times 70, which is 490.
• 12 times 75: halve 12 to get 6, double 75 to get 150. Now it is 6 times 150, which is 900.

You can repeat this. 24 times 25 becomes 12 times 50, which becomes 6 times 100, which is 600.

Why it works: Multiplying one factor by 2 and dividing the other by 2 leaves the product unchanged, because the two operations cancel. You are restating the same multiplication in a friendlier form.

Trick 6: The "Round and Adjust" Move

For mental subtraction and addition involving numbers near a round value, round first and adjust at the end.

• 472 minus 199: round 199 up to 200, subtract to get 272, then add back the 1 you over-subtracted: 273.
• 583 plus 297: round 297 up to 300, add to get 883, then subtract the 3 you over-added: 880.
• 845 minus 398: round 398 to 400, subtract to get 445, add back 2: 447.

This is the single highest-leverage technique on this list. Real-life arithmetic is full of awkward numbers near round ones (prices ending in 99, percentages ending in 9, time intervals near a whole hour), and the round-and-adjust move handles all of them.

Why it works: You are exploiting the associative property: a minus b equals a minus (b plus delta) plus delta. Rounding b changes one ugly subtraction into one easy subtraction plus a tiny correction.

Trick 7: Estimation as a Sanity Check

This is the trick that keeps the others honest. Before you commit to an answer, do a five-second estimate using rounded numbers and check whether the precise answer is in the right ballpark.

If you compute 47 times 23 and get something like 1081, ask: "47 is roughly 50, 23 is roughly 20, so the answer should be near 1000." 1081 is reasonable. If you had gotten 10810 or 108, you would catch the slipped decimal or extra zero immediately.

Estimation is the most underrated mental math skill because it does not feel like math. It feels like common sense. But on every standardized test ever written, the wrong-answer choices are designed to look plausible to a student who skipped the sanity check. A two-second estimate eliminates trap answers faster than any algebra.

How to Practice These Without Burning Out

Reading about these tricks will not make them automatic. They become useful when they are faster to deploy than long arithmetic, and that takes deliberate, low-stakes repetition. Two principles:

Practice in tiny doses. Five minutes a day of mental arithmetic produces faster fluency than a one-hour session per week. The reason is the same as for spaced repetition in general: your brain consolidates skills between sessions, not during them.

Mix the tricks. Do not drill multiplication-by-11 problems for ten minutes straight. Mix in squaring, percentages, and round-and-adjust problems in the same session. Interleaved practice feels harder in the moment, but it is what builds the recognition skill of "which trick applies here?", which is the actual point.

If timed practice makes you tense, start untimed. There is no value in drilling speed under a stress response that is killing your accuracy. See our piece on math anxiety for what to do when the timer itself is the obstacle.

Where Math Zen Fits In

Mental math fluency is one of the most natural use cases for short, frequent app sessions. Math Zen's Zen Mode is a calm space to drill arithmetic without a clock when you are still learning the trick; once a technique feels automatic, Timed Mode gives you a way to test whether it actually saves you time under pressure. The bucket progression keeps the difficulty in the productive zone, so you are not wasting time on problems that are too easy or grinding through problems that are too hard.

For most learners, ten minutes a day for two to three weeks is enough to make the tricks above feel natural. After that, they become invisible: you stop noticing that you are using them, which is exactly when they start paying off in every other math problem you face.

The Bottom Line

The best mental math tricks are not memory feats. They are small restatements of arithmetic that exploit how numbers behave: multiplying by 10 plus the original, splitting near round values, swapping percentages, doubling and halving. Each one is short, each one has a reason, and each one becomes automatic with a few minutes of daily practice.

Pick two from this list to start. Use them in the wild for a week, in your head while you wait in line or check a receipt. By next month, they will be background skills, and you will move on to harder problems with more of your working memory free for the actual thinking.