How To Tell If We're Beating COVID-19

We were drowning in numbers. Every headline screamed a new case count, a new death toll, a new wave of fear. And none of it told us the one thing we actually needed to know... are we winning or losing?

That question haunted Henry Reich of MinutePhysics. And honestly, it haunts me in every arena of life. Not just pandemics. Every mission. Every fight against hopelessness. Are we making progress, or are we just spinning?

Here's what Henry and his collaborator Aatish Bhatia did. They stopped looking at the numbers the way everyone else was looking at them. They built a data visualization that stripped away the noise and revealed the signal.

Three ideas made it work.

Idea One: Scale It Right

They used a logarithmic scale. Not because it sounds fancy... because exponential growth demands it. On a normal scale, the early days of an epidemic look flat and harmless. By the time the curve shoots upward, it's already too late to react. A log scale treats growth equally at every level. Ten to a hundred looks the same as ten thousand to a hundred thousand. It lets you see the pattern when the pattern is still small enough to fight.

This is a principle that transfers. When you're tracking anything that compounds... debt, grief, disconnection, hopelessness... the early signals look tiny. Almost invisible. A logarithmic perspective trains your eye to catch the trajectory before the explosion.

Idea Two: Watch the Rate of Change

Most graphs showed cumulative cases over time. That tells you where you've been. Henry and Aatish plotted new cases per week instead... the rate of change. Not "how many total?" but "how fast is this growing right now?"

Think about it. If you're climbing out of a pit, the depth of the hole matters less than your velocity. Are you climbing faster today than yesterday? Are you slowing down? The rate of change is where the truth lives.

South Korea showed this beautifully. Their cumulative curve still looked scary halfway through... but their new-case-per-week chart revealed the turn. The growth was slowing. The interventions were working. You just couldn't see it without watching the rate.

Three months without food, three days without water, three minutes without hope. When you're in the middle of the fight, the only thing that keeps you breathing is detecting that the trajectory is bending. Even slightly. Even a little.

Idea Three: Remove Time From the Equation

This one blew me away.

They threw time off the x-axis entirely. Instead, they plotted new cases against total cases. Why? Because the disease doesn't care if it's March or April. It cares about two things only: how many cases exist now, and how many new cases are appearing.

When you plot it this way, exponential growth becomes a straight diagonal line. Every country in the world follows essentially the same trajectory... just shifted in time. And when a country's curve plummets off that diagonal? That's the signal. That's the emergency eject button being hit. That's China dropping off the line. That's progress made visible.

The graph doesn't need dates. It needs direction.

The Caveats Matter Too

Henry was honest about the limitations. Soooo honest. And that honesty is what makes this trustworthy.

- Log scales compress big numbers. Ten thousand looks close to one thousand. That can breed complacency. - The data shows detected cases, not true cases. Every country's testing capacity distorts the picture. More tests mean more detected cases, which can make growth look worse than reality... or better, depending on who's testing. - Using a weekly average introduces a natural delay. But that delay is actually a feature, not a bug. It makes the graph pessimistic by design. A downward trend on this graph is much more likely to be a real downward trend.

BAM... that's responsible science communication. Show the receipts. Show the cracks. Let people make informed decisions instead of emotional ones.

Why This Matters Beyond the Pandemic

This graph is a masterclass in a principle I keep coming back to: the way you frame a question determines what answers you can see.

Every daily news cycle was answering "how many cases today?" That's the wrong question. The right question is "what direction are we headed, and is that direction changing?"

Swap "cases" for anything. Student engagement. Donor fatigue. Personal healing. Organizational health. The question is never just "where are we?" The question is "are we bending the curve?"

Data literacy isn't just for scientists. It's survival gear for anyone trying to fight a war on hopelessness. Because when you can't see progress, you lose hope. And when you lose hope, you stop showing up. And when you stop showing up... the exponential wins.

Light doesn't fight darkness. It just shows up. But first you have to be able to see the light. Tools like this graph make the light visible when everything else is noise.

So here's the invitation. Next time you're drowning in data... in numbers that feel overwhelming and contradictory... stop asking "how much?" Start asking "which direction, and how fast?" Strip away the noise. Find the rate of change. Because that's where hope hides. Not in the total count of your scars, but in whether the bleeding is slowing. That's the graph worth watching. 💙

--- Source: https://www.youtube.com/watch?v=54XLXg4fYsc