Numbers Lie to You
- By: Thiago Munck
- On:

Have you ever seen an exaggerated statistic in the news and thought to yourself:
“There’s no way this is true!”
You were probably right. But how can it be a lie if it’s based on statistical data?
Today, we’re going to talk about some of the most common strategies used to deceive customers and make results look more appealing than they really are.
1 – Misleading Graph Scales
Anyone familiar with math or physics knows that graphical analysis requires consistent scales and breakpoints. However, that rule is often ignored to manipulate visual perception.
At first glance, a graph might suggest that interest rates in 2012 were dramatically higher than in other years. But when you look at the actual numbers, the difference is only 0.012%. A small difference appears massive simply because the graph’s scale was distorted.
2 – Relative vs. Absolute Numbers
Another common strategy is using percentages to make numbers sound more impressive than they are. Just like with graphs, it’s about making something small look big.
For example, a pharmaceutical company might advertise that their drug reduces the risk of heart attack by 50%. That sounds dramatic—until you find out that, in reality, the risk only drops from 2 in 100 people to 1 in 100.
So yes, it’s a 50% reduction… but it only benefits one person out of a hundred.
3 – Numbers Without Context
This tactic is one of the most misleading. Numbers without proper background or explanation can be twisted into saying almost anything.
Case 1: The Toothpaste Ad
You’ve probably seen claims like “9 out of 10 dentists recommend this product.” But how was this data collected?
Was the survey representative? How many dentists were consulted?
Also, marketers know that saying “90% of professionals recommend it” sounds more believable than saying “10 out of 10,” which could feel suspiciously perfect.
Case 2: Football Player Comparison
Let’s compare two well-known footballers: Harry Maguire and Virgil van Dijk.
Maguire has often been criticized as error-prone, while van Dijk is praised as one of the best defenders of his generation. But if we look only at raw stats—like blocks, interceptions, and tackles—we’re missing important context.
Maguire’s teams were often under more pressure, meaning he had more opportunities to rack up defensive stats. On the other hand, van Dijk played for dominant Liverpool teams that faced fewer attacks.
So while the numbers may seem to favor Maguire in some categories, that doesn’t automatically make him the better defender. Context matters.
4 – Bias in Data Collection
Bias during data collection or interpretation can generate numbers that appear meaningful, but don’t reflect reality. Two major types of bias that distort data are:
Survivorship Bias
During World War II, U.S. analysts studied bullet holes on aircraft that had returned from battle to figure out where to add extra armor. At first, it made sense to reinforce the areas with the most damage.
But statistician Abraham Wald from Columbia University pointed out the problem: they were only analyzing the surviving planes. The missing data—the planes that were shot down—wasn’t included.
The real weak points were the areas without bullet holes on the survivors, because hits in those zones likely caused the planes to crash.
This is survivorship bias: drawing conclusions based only on the data that “survived” the process, while ignoring what was lost.
Voluntary Response Bias
Online ratings are another example. A product might have a high average score based on dozens or hundreds of reviews. But are those reviews truly representative?
In many cases, the ratings come from people with strong opinions—either very positive or very negative. The majority of users, who may have had neutral or average experiences, usually don’t leave reviews.
This is called voluntary response bias: the data is skewed because only certain people choose to participate. As a result, the numbers may not reflect how the average person actually feels.
Final Thoughts
Just because something comes with a number doesn’t mean it tells the truth.
Statistics can be used to reveal reality—or to hide it. Whether it’s distorted graphs, out-of-context data, or biased sources, always ask:
“Where did this number come from?”
“What’s missing from the picture?”
Being aware of these tactics helps you see through the numbers—and get closer to the truth.