6 min read

Reliable Beats Accurate: Why Your Tracker Doesn't Have to Be Perfect

A client says to me, "My ring told me I burned 1,500 calories in that session, so I figure I earned the pizza." And I get it. You worked hard. The number is big. The reward feels deserved. But that number is mostly a lie, and the part that matters most isn't even the part you're looking at.

Let me untangle the thing that took me years to actually understand, because once it clicks, tracking stops feeling like a chore and starts feeling like a superpower.

Two words that fix everything: reliability and validity

These are technical terms, but they're simple, and they change how you should think about every number your devices spit at you.

Validity is whether a measurement is actually measuring what you think it's measuring. Is the number TRUE?

Reliability is whether a measurement gives you the same answer under the same conditions. Is the number CONSISTENT?

Picture a bathroom scale. You step on, it says 188. You step off, step back on, it says 188 again. That's reliable; it gives you the same number every time. But if nobody zeroed it out, you might actually weigh 184. So it's reliable and NOT valid. It's consistent, but it's consistently off.

Here's the part nobody tells you: for almost everything we track day to day, reliability matters more than validity.

Why "wrong but consistent" still works

Your wearable is not great at validity. The research backs this up; the calorie burn it estimates, the exact step count, those carry real error. A step tracker might be off by ten percent in either direction. Ten thousand steps could really be nine thousand, or eleven thousand. That sounds like a dealbreaker.

It isn't. As long as you use the SAME method every single time, that error stops mattering.

Think about it. If your tracker is always off by roughly the same amount in the same direction, that error becomes a constant. It's baked into every data point. And when you're looking at a trend line over weeks and months, a constant error just shifts the whole line; it doesn't change the SHAPE of the line. The question you actually care about is "am I trending up or down, am I hitting my average," and a consistent error answers that question perfectly fine.

The longer your timeline, the less any single day's variance matters, because that day is a smaller and smaller slice of the whole picture. The trend smooths it out.

Where people wreck this is by switching methods. One day you log steps off your ring, the next day you forgot to charge it so you use your phone, and surprise, the two disagree. Now you've introduced new error that isn't constant, and your trend gets noisy. Pick one method. Use it relentlessly. I track off my ring because I'm far more likely to have it on than to have my phone in my pocket. The accuracy isn't why I chose it; the consistency is.

The calorie-burn trap

Back to that 1,500-calorie workout. Here's the math people miss.

When your heart rate monitor estimates burn, it asks for your age, weight, and gender up front, and uses that to guess your basal metabolic rate, the calories you'd burn just existing. Then it adds the bump from your elevated heart rate. So that 1,500 includes the roughly 1,000 you would have burned sitting at a desk doing nothing. Only about 500 came from the actual work.

So when you "eat back" 1,500 calories, you're double-counting. Your daily intake target was already built off your basal rate. Now you're adding it again on top. This is why I don't have clients add calories based on what their device says they burned. We'll add fuel on heavy training days, sure, but it's deliberate, not pegged to a number the device made up.

The quiet superpower: track how you FEEL

Here's where it gets good. The numbers (steps, calories in, sleep hours, resting heart rate) are your quantitative data. Objective, measurable. Useful.

But the qualitative stuff is just as important, sometimes more. Every week, rate these one to ten: stress, hunger, recovery, energy, digestion, sleep quality, mood, cravings, training quality, and for those who track it, menstrual cycle and sex drive. One being easy breezy, ten being get away from me.

These feel fuzzy and subjective, and on any single day they are. But aggregate them over time and the subjectivity smooths out, same as the step count does. And the patterns they reveal are gold. If I see a client's stress elevated all week, hunger weirdly low, but cravings climbing, that combination is telling me their body is adapting to a deficit. Hunger hormones got turned down, but the body still wants calories, so it routes the signal through cravings instead. I can read that without a single lab test.

That's the whole point. Cortisol tests and lab panels are the truly valid measurements, but they're expensive and you can't run them every day. Rating your stress from one to ten costs nothing, and tracked consistently over months it gives you a window into the same biology. Reliable enough, often enough, beats valid but unaffordable.

What to actually do

Pick your metrics. Pick ONE method for each. Use that method every day, even when you suspect it's a little off, because the "off" is your friend now. Log the feelings alongside the numbers. Then stop staring at individual days and start reading the weekly and monthly averages.

The more honest data points you feed in, the clearer the picture gets, and the better any decision built on it becomes.

One caveat, because this is your health: everything here is a general principle for reading your own trends, not a prescription for your specific situation. If something in your data looks off, loop in a coach or a medical professional who can see the whole picture with you.

That's the real reason I lean on tracking, and the real reason I lean on my own coach. Two weeks ago felt like a total mess to me; travel, a sick kid, an anniversary, my wife's birthday, all of it. Then I turned in my data and my coach said, "These were good weeks. Your numbers look great." It didn't feel that way. The data knew better than my mood did. That's what good data buys you: a truth that's steadier than how you happen to feel on a Tuesday.

Listen to the full episode here.

If you want help building a tracking system that actually informs your decisions instead of just nagging you, reach out and apply. You don't have to hire me to get pointed in the right direction; I've never turned anyone away over money, and I'm always happy to send you to the right person if it isn't me.

Vokse eller dø. — Coach Jordan