How To Get An Average Of Two Numbers

You stare at the spreadsheet. Two cells. Worth adding: two numbers. You need the middle ground. The balance point. The average.

Most people freeze for a second. That's why is it addition first? Here's the thing — division first? In real terms, does order matter? It’s one of those things that feels like it should be obvious — until you actually have to do it in front of someone.

Let’s clear it up once and for all.

What Is an Average of Two Numbers

At its core, the average of two numbers is the value sitting exactly halfway between them. " Exactly halfway. Not "roughly halfway.If you drew a number line and dropped a pin at the first number, then another at the second, the average is the spot where the distance to the left pin equals the distance to the right pin.

Mathematicians call this the arithmetic mean. For two values, the formula collapses into something almost stupidly simple:

Add them together. Divide by two.

That’s it. Now, no weighting. No sigma notation. No geometric means unless you’re doing something very specific (and if you are, you already know you’re not reading this article).

The Formula Written Out

If your numbers are a and b:

Average = (a + b) / 2

You can swap a and b. The result doesn’t change. Even so, addition is commutative. Division by a constant preserves that. The order genuinely does not matter.

A Quick Concrete Example

Say you have 14 and 22.14 + 22 = 36
36 / 2 = 18

Eighteen is four away from 14. Plus, eighteen is four away from 22. Symmetry achieved.

Why It Matters / Why People Care

You might wonder why a whole article exists for this. Fair question. The truth: people mess this up constantly. Not because the math is hard, but because the context trips them up.

Grades and GPA

A student gets an 84 on the midterm and a 92 on the final. That’s a weighted average — a completely different beast. Worth adding: " The average is (84 + 92) / 2 = 88. That's why the syllabus says "exams count equally. But if the midterm is worth 30% and the final 70%, you can’t just average the two scores. Simple. Confusing the two is the single most common grading error I see.

Financial Midpoints

You’re negotiating salary. On the flip side, the range posted is $60,000 to $80,000. The midpoint — the number you’d cite as "market average" — is $70,000. That said, recruiters know this. Candidates who don’t often anchor too low or too high.

Data Cleaning

Two sensors. But if one sensor is known to drift, a simple average introduces bias. 7. So one reads 102. Average them: 102.3, the other 101.You need a single value to log. Plus, 0. Again — context changes the tool.

Everyday Estimation

Gas prices. Two stations near your house: $3.49 and $3.On top of that, 69. Now, quick mental average: $3. 59. You decide if the drive to the cheaper one saves enough to justify the extra mileage. People do this math in their heads at the pump constantly.

How It Works — Step by Step

Let’s walk through the mechanics slowly. Not because you can’t add, but because the how reveals where errors hide.

Step 1: Identify the Two Values

Sounds trivial. But in a messy dataset, "the two numbers" might be buried in different columns, different sheets, or different formats. In real terms, one might be a percentage (0. 87) and the other a raw score (87). If you average those without converting, you get garbage.

Rule: Both numbers must be in the same unit and scale before you touch a calculator.

Step 2: Add Them

Use a calculator. Use Python. Still, use mental math if the numbers are friendly. But write down the sum. Don’t just hold it in working memory — that’s how transposition errors happen (writing 36 as 63).

Step 3: Divide by Two

Divide the sum by 2. That said, not by the number of data points in the whole dataset. By 2. Not by the count of numbers you think* you have. Because there are two numbers.

Step 4: Sanity Check

Look at the result. Is it between the two original numbers? If not, you made a mistake. On the flip side, the average of two numbers must* lie between them (inclusive, if they’re equal). No exceptions.

Doing It in Spreadsheets

Excel, Google Sheets, LibreOffice Calc — they all handle this the same way.

Continue exploring with our guides on how to find range of a data set and what time will it be in 15 minutes.

Method A: The Formula

=(A1+B1)/2

Type it. Hit enter. Done.

Method B: The AVERAGE Function

=AVERAGE(A1:B1)

This is safer if you might later expand to three or four numbers. The function ignores text and empty cells. The manual formula does not.

Method C: Status Bar Highlight Highlight the two cells. Look at the bottom right of the window. Sheets and Excel both show "Average: [value]" automatically. Zero typing. Fastest way to verify.

Doing It in Code

Python:

avg = (a + b) / 2

Or:

import statistics
avg = statistics.mean([a, b])

The statistics module handles edge cases (like Decimal or Fraction types) more gracefully than raw division.

JavaScript:

const avg = (a + b) / 2;

Watch for integer division in languages that enforce it (C, Java, Go with int types). Cast to float first if you need precision.

Mental Math Shortcuts

For two-digit numbers, you don’t always need a calculator.

Trick 1: Meet in the Middle
If the numbers are 37 and 43, the middle is obvious: 40. They’re symmetric around 40.

Trick 2: Adjust One, Adjust the Other
38 and 44. Add 2 to 38 → 40. Subtract 2 from 44 → 42. Now average 40 and 42 → 41. Same result. Less carrying.

Trick 3: Use the Difference
Difference is 6. Half the difference is 3. Add 3 to the smaller (38 + 3 = 41). Or subtract 3 from the larger (44 - 3 = 41). This works every time.

Common Mistakes / What Most People Get Wrong

Averaging Percentages Directly

You scored 80% on a 50-question test and 90% on a 100-question test. Your overall percentage is not (80 + 90) / 2 = 85%. The tests have different weights. You need total points earned divided by total points possible. This is the weighted average trap. It catches everyone once.

Averaging Rates

You drive to work at 30 mph. Your average speed is not 45 mph. You drive back at 60 mph. 60 miles / 1.Average speed is total distance divided by total time. In practice, 5 hours = 40 mph. If the distance is 30 miles each way: 1 hour there, 0.Now, 5 hours back. The harmonic mean applies here, not the arithmetic mean.

Forgetting Units

Forgetting Units

Mixing units silently corrupts results. But 26 what? 20°C = 68°F. That's why convert first. Neither Fahrenheit nor Celsius. Even so, you’re averaging temperatures: one reading is 32°F, another is 20°C. Now (32 + 68)/2 = 50°F. Plugging those numbers directly gives (32 + 20)/2 = 26. Always verify units match before averaging.

Rounding Too Early

You’re averaging 1.Think about it: 234 and 2. Plus, 345. If you round each to one decimal place first (1.This leads to 2 and 2. So naturally, 3), your average is 1. 75. Still, the correct answer is (1. Day to day, 234 + 2. 345)/2 = 1.7895, which rounds to 1.And 8. Premature rounding introduces systematic error. Keep full precision through the calculation, round only the final result.

Treating Averages as Exact Values

The average of 10 and 20 is 15. Day to day, averages hide distribution. But that doesn’t mean every value was 15, or that 15 is representative. Two datasets can have the same mean but wildly different spreads. Always pair averages with context: range, standard deviation, or a quick mental check of whether the numbers cluster or scatter.

Conclusion

Finding the average of two numbers is deceptively simple. So the core operation — add and divide by two — rarely fails you. But the surrounding context is where mistakes live: unit mismatches, premature rounding, percentage traps, and rate miscalculations. Whether you're typing a formula into a spreadsheet, writing code, or doing mental math, the key is matching your method to your use case. So use =AVERAGE() when you might expand later. Use the status bar for quick checks. In practice, use mental shortcuts to build intuition. And always, always do a sanity check — if your average doesn't fall between the two original numbers, something went wrong.