Why We Built the Argument Strength Meter

Here is a question you face every single day, probably dozens of times: Should I be convinced by this? Someone makes a claim — a politician, a colleague, an ad, a headline. You feel a pull. It sounds right. But is it actually a strong argument, or does it just feel like one?

This distinction — between feeling convinced and an argument actually being strong — is one of the most underappreciated skills in critical thinking. And it is the reason we built the Argument Strength Meter.

Confidence is not evidence

The human brain is remarkably bad at separating persuasion from proof. Three cognitive tendencies make this especially difficult:

Authority bias. If a confident expert says something, we tend to believe it — even when they offer no supporting evidence. The white coat, the title, the assured tone: these are signals of status, not of argument quality.

The fluency effect. Arguments that are easy to process feel more true. A well-crafted sentence, a catchy slogan, a clean metaphor — our brains interpret processing ease as a signal of truth. This is why misinformation that rhymes spreads faster than corrections that don't.

Confirmation bias. When a claim aligns with what we already believe, we barely evaluate it at all. It just slots in. We save our skepticism for claims that challenge us — which means the arguments we scrutinize least are the ones most likely to be wrong in ways we would never notice.

These aren't character flaws. They're features of a brain optimized for speed, not accuracy. But they mean that our gut feeling about argument quality is systematically unreliable.

What actually makes an argument strong?

Stripped of rhetoric, an argument's strength comes down to a handful of factors:

• Evidence relevance. Does the evidence actually bear on the claim? Citing a study about mice to prove a claim about human psychology is evidence — but not relevant evidence.
• Evidence sufficiency. Is there enough evidence? A single anecdote can illustrate, but it cannot prove. One study can suggest, but it cannot settle.
• Logical validity. Does the conclusion actually follow from the premises? "Sales went up after we changed the logo, therefore the logo change caused the increase" has a premise and a conclusion — but the connection is assumed, not demonstrated.
• Alternative explanations. Has the argument considered other possible causes? Strong arguments don't just build a case for their conclusion — they actively address why competing explanations are less likely.
• Appropriate hedging. Does the argument's confidence match its evidence? Claiming certainty from preliminary data is a red flag. Strong arguments say "suggests" when the evidence suggests, and "proves" only when it does.

Three arguments, three scores

To make this concrete, here is how the Argument Strength Meter might rate three common types of claims:

"This policy created 2 million jobs." A political claim. Score: 3/10. It asserts a causal relationship without controlling for other factors (economic cycles, global trends, previous policies). It uses a precise number to imply precision of analysis. No alternative explanations are considered. This is a common pattern: specific numbers create an illusion of rigor.

"9 out of 10 dentists recommend this toothpaste." An advertising claim. Score: 2/10. The survey methodology is unknown. "Recommend" is undefined — over what alternatives? The statistic is designed to sound scientific while providing no actual scientific content. This argument's strength is almost entirely rhetorical.

"A meta-analysis of 47 studies found a moderate effect (d=0.4) of spaced repetition on long-term retention." A scientific claim. Score: 8/10. It specifies the method (meta-analysis), the evidence base (47 studies), the effect size (d=0.4), and hedges appropriately ("moderate effect"). It doesn't claim more than the data supports. The main limitation — which a truly strong presentation would note — is whether the included studies are themselves methodologically sound.

How to use the Argument Strength Meter

The app is simple by design. Paste or type any argument — a paragraph from a news article, a talking point from a debate, a claim from social media. The Argument Strength Meter breaks it down across the dimensions above and assigns a score from 1 to 10, with a clear explanation of why.

It is not a fact-checker. It does not verify whether claims are true. What it does is something different and, we think, more fundamental: it evaluates whether the reasoning is sound. A factually true claim can be supported by terrible reasoning, and a false claim can be wrapped in impeccable logic. The Meter focuses on the logic.

Over time, using it builds intuition. You start to notice when an argument skips over the "why should I believe this?" step. You catch yourself being swayed by fluency rather than substance. The goal isn't to make you suspicious of everything — it's to make you a better judge of when suspicion is warranted.

Try the Argument Strength Meter — and see how the arguments you encounter every day hold up under scrutiny.