Few things make a bad scientific theory more entertaining than a perfectly presented correlation.

A graph rises. A pattern appears. The numbers seem to line up beautifully. A confident researcher steps forward and announces that a hidden relationship has finally been discovered.

There is just one small problem.

The two things being compared have absolutely nothing to do with each other.

This is the world of spurious correlations: situations where two sets of data appear connected, but the relationship is accidental, meaningless, or caused by a completely different factor. While genuine scientific research requires careful investigation to determine whether a correlation represents a real connection, bad science has a habit of stopping at the graph and declaring victory.

For BAHFest, this makes correlation one of the richest sources of inspiration. A questionable pattern, a dramatic explanation, and an unnecessary amount of confidence are all the ingredients needed for a spectacularly wrong hypothesis.

What Is a Spurious Correlation?

A correlation occurs when two variables appear to change together.

For example:

• One measurement increases while another increases.

• One measurement decreases while another decreases.

• Two trends appear to follow the same pattern over time.

A correlation can be useful. Scientists use correlations every day to identify possible relationships worth investigating.

The problem begins when someone assumes:

“These two things happened together, therefore one must have caused the other.”

That leap is where comedy—and bad science—often begins.

A correlation may reveal an interesting pattern, but it does not automatically prove a cause-and-effect relationship.

A great BAHFest theory takes that mistake and runs with it.

The Classic Trap: Correlation Does Not Equal Causation

The phrase “correlation does not imply causation” has become one of the most important warnings in statistics.

It reminds researchers that two things can appear connected for many reasons.

A correlation might exist because:

• One factor actually causes the other

• Both factors are influenced by a third variable

• The relationship is purely accidental

• The data has been selected in a misleading way

• The sample size is too small

A responsible scientist investigates further.

A BAHFest scientist immediately prepares a presentation with 47 slides proving that the connection was obvious all along.

The Ice Cream and Shark Attack Problem

One famous example of correlation involves ice cream sales and shark attacks.

The pattern appears alarming:

• Ice cream sales increase

• Shark attacks increase

Does this mean ice cream attracts sharks?

Should beaches ban chocolate cones to protect swimmers?

Fortunately, the answer is no.

The hidden factor is the weather. During warmer months:

• More people buy ice cream

• More people visit beaches

• More people swim in the sea

The real connection is temperature and seasonal behaviour.

But without looking deeper, the graph creates a wonderfully misleading story.

This is exactly the kind of pattern that makes a perfect bad scientific theory:

1. Find two things that move together.

2. Ignore the obvious explanation.

3. Invent a dramatic hidden mechanism.

4. Present the discovery with confidence.

The Famous Spurious Correlation Database

The internet has produced countless examples of amusing correlations that demonstrate how easily patterns can be found in unrelated information.

One popular collection of examples compares unrelated statistics that happen to follow similar trends.

These include bizarre relationships such as:

• Cheese consumption appearing connected with unusual events

• Movie budgets appearing linked with unrelated social measurements

• Certain household activities matching completely unrelated national statistics

The humour comes from the fact that the relationships look meaningful when displayed visually.

A graph gives information authority.

A line moving upward feels like it must mean something.

The human brain is naturally drawn to patterns—even when the pattern is meaningless.

Why Humans Love Finding Patterns

The tendency to find connections is deeply human.

Our brains are designed to identify patterns because recognising relationships helped our ancestors survive.

If rustling grass might indicate a predator, assuming a connection was useful.

In modern life, however, the same instinct can lead us to see meaning where none exists.

We are especially vulnerable when:

• The data looks scientific

• The explanation sounds complicated

• The pattern matches our expectations

• The conclusion confirms something we already believe

This makes correlation an excellent tool for comedy because it exposes one of our most common reasoning mistakes.

How BAHFest Turns Correlations Into Comedy

A typical BAHFest presentation begins with something real.

The speaker notices an observation:

“People who do X often experience Y.”

At this stage, everything seems reasonable.

Then comes the crucial scientific leap:

“Therefore, X must be causing Y through an undiscovered mechanism involving an extremely complicated process.”

The explanation becomes increasingly elaborate.

Perhaps:

• A hidden biological system is responsible

• A forgotten evolutionary adaptation is involved

• A mysterious environmental force is influencing behaviour

• A previously unknown law of physics explains everything

The theory grows stronger with every slide, despite becoming less believable.

The Power of the Impressive Graph

Graphs are essential tools in genuine science, but they are also incredibly effective comedy devices.

A good BAHFest graph has three important features.

A Serious Title

“Relationship Between Human Behavioural Patterns and Atmospheric Conditions”

sounds scientific.

“Things That Happen at the Same Time”

does not.

Carefully Chosen Data

A convincing graph does not need to prove anything.

It only needs to make the audience briefly wonder:

“Wait… is there something here?”

A Dramatic Conclusion

After presenting a complicated graph, the speaker delivers the inevitable discovery:

“The data clearly demonstrates that our original completely unreasonable hypothesis was correct.”

The confidence is what makes it work.

When Coincidence Becomes a Theory

The most entertaining bad scientific theories often begin with a real coincidence.

History is full of examples where unexpected observations led to important discoveries.

The difference is what happens next.

A good scientist asks:

• Is this relationship real?

• Can it be repeated?

• What alternative explanations exist?

• What evidence would prove me wrong?

A bad scientist in a BAHFest presentation asks:

• How many additional slides can I make about this?

The Danger of Data Without Understanding

Spurious correlations highlight an important lesson about modern information.

Today, enormous amounts of data are available. With enough measurements, it is possible to discover countless apparent relationships.

The challenge is not finding patterns.

The challenge is understanding which patterns actually matter.

A thousand matching trends do not necessarily reveal a hidden law of nature. Sometimes they reveal nothing more than coincidence.

Why Bad Science Helps Explain Good Science

BAHFest celebrates incorrect theories, but the joke works because audiences understand what real science requires.

Good science depends on:

• Careful measurement

• Appropriate analysis

• Testing explanations

• Considering alternatives

• Accepting when evidence disagrees

Bad science often skips these steps and jumps directly from observation to conclusion.

That jump is where the comedy lives.

The Beautiful Absurdity of Being Wrong

Correlation is powerful because it can create the illusion of discovery. A simple graph can make unrelated events appear connected and encourage us to invent explanations that do not exist.

For BAHFest performers, this is a gift.

A suspicious pattern becomes a breakthrough.
A coincidence becomes evidence.
A mistake becomes a scientific revolution.

The greatest bad hypotheses are not created from complete nonsense. They are created by taking a tiny piece of reality, misunderstanding it spectacularly, and presenting the result with absolute confidence.

And sometimes, when the graph looks convincing enough, even the most ridiculous idea can seem almost scientific.

Almost.