Fitness Trackers vs. Smartphones: Why Wearables Win

Hey guys, I'm Brent Rose,

tech writer and unlicensed plastic surgeon.

Today, we're gonna be talking about wearables.

A few weeks ago, a study came out that kind of

blew everybody's mind.

This study was conducted by researchers at the

Perelman School of Medicine and the Center for Health

Incentives and Behavioral Economics at

the University of Pennsylvania

or as you probably know them the P-S of M

and the C for H-I and B-E at U-P.

What this study claimed is that the phone in

your pocket is just as if not more accurate

at tracking as the wearables on your wrist.

A study like this is cat net for journalists.

It spells the death of fitness trackers,

and who wouldn't want fitness trackers to die.

They're ugly, they're everywhere,

and your friend who wears won't shut up about it.

The problem is the science behind the study

wasn't very good.

He used devices that were two years old and frankly,

the conclusions they drew were really misleading.

Let's break this down.

Today we're gonna be doing our own test

with all new gadgets.

We've got the Withings Pulse O2, the Basis Peak,

the Jawbone UP MOVE, Google's Nexus 6,

Apple's iPhone 6, the Garmin Fenix 3

and Fitbit Charge HR.

All slowly drowning in a sea of arm hair.

But the only thing that study looked at was the accuracy

of step counting on a treadmill.

We're gonna do the same test but in real life.

I'm gonna count 1,000 steps.

Here I go, one, two, three, four, five.

(upbeat music)

98, 99, 300, one, two.

(upbeat music)

99, 1,000.

Let's write everything down.

Three, four, are you kidding, wow.

Okay, so let's see what we got here.

The Nexus 6 came in at only 914 steps.

The Withings Pulse which was in my pocket came

in at 1,047 steps.

The UP MOVE was 1,011 steps as well as the Basis Peak.

The Fenix on my wrist here was only off by seven steps.

The iPhone came in at 995.

The real surprise is the Fitbit, which was on my wrist,

came in as the most accurate, 1,001 steps.

So what we're seeing here is that there's not a huge

amount of difference between whether or not it was

on my wrist or in my pocket,

whether it was a dedicated fitness tracker

or whether it was a phone.

They're all within a pretty small range.

So this is kind of flying in the face of

what the study's saying already.

That said, let's see if we can trip up

the wrist-worn trackers.

(clacking)

So a lot of people think that certain activities

will trigger false positives in the step counting.

So to simulate that, we put all four

wearable trackers on one wrist.

We decided to do about five minutes of whittling.

We chose whittling because it's a pretty good analogue

for some very common activities

such as peeling carrots, peeling potatoes, and yes,

obviously what you guys are all thinking, pic-amon.

Holy shit, that I did a lot of running.

Just then 540 false steps,

611, 618, 722 false steps.

So that was actually pretty damning.

These wrist-worn trackers were definitely subject

to a lot of false positives.

They thought I took on average 622 steps

while I was just sitting here whittling.

So that's really bad but here's the thing.

Steps counted is a really poor metric

for measuring your fitness levels anyway.

It doesn't know whether you're taking these steps

quickly or slowly, whether you're walking or dancing,

and obviously whittling is a serious issue.

And there's another problem.

Phones and other accelerometer-based trackers

are absolute garbage when it comes to tracking

anything that doesn't involve steps.

A much more useful metric for people is caloric burn.

You wanna know how many calories you're expending

so you know how much you can eat.

This is where the new breed of fitness trackers

really start to set themselves apart.

We're talking about the ones with built-in

heart rate monitors like the Fitbit Charge HR,

the Basis Peak, the Moto 360 Smartwatch

and even the forthcoming Apple Watch.

Let me show you what we mean.

Okay so both the Fitbit and the Basis Peak

are showing 74 beats per minute right now.

Even if these phones, and the accelerometer trackers

thought that each crunch was a single step,

obviously it's a lot harder than single steps.

So 101, 99, so they know I'm working.

These other ones have no idea what's going on.

This is a pretty extreme example

because I'm completely still and yet I'm working very hard.

As far as the phones are concerned,

and the non-heart rate monitor activity trackers,

I might as well be asleep, sitting on a train somewhere.

This one knows however my heart rate is already up

at a hundred.

This one says 98, 111, 113, 114 now.

Couldn't Fitbit, this one again no idea what's happening.

Obviously biking is probably the most popular

non-walking form of exercise.

What we're gonna do is see how many calories I burn

at the end of a 10-minute bike ride.

Most of these monitors have no idea what I'm doing.

The ones that have a heart rate monitor should show

a higher caloric burn.

Unfortunately our two phones don't do caloric estimates.

They wonked out right away.

Yoo-hoo.

All right, my heart's still

around 140 beats per minute.

I think if I peak around 148,

let's see what the calories are.

So this was exactly what we expected.

The Withings was way low at 54.

The Garmin was low at 91.

The Jawbone was pretty low at 163

but the Fitbit and the Basis were the two highest.

226 for the Fitbit and 243 for the Basis.

That makes a lot of sense because they were watching

my heart rate, and they could tell how hard I was working

whereas these other ones were doing it

just based on my motion.

So what did we learn here today?

While your phone is awesome in a lot of ways,

and it's less prone to false positives and step counting,

step counting isn't really that important

in the grand scheme of your overall health.

The more important metric is caloric burn

and for that, you gotta have a heart rate monitor.

So until your phone can monitor your vitals

through your pants' pocket, wearables aren't going anywhere.

This study was conducted by researchers

at the university, no.

So a few weeks ago, study was published in the journal

called the Medical German, Perelman School of Medicine

and the Center for.

The P-S of M and the C.

The P-S of M.

The P-S of M and the C for.

Why is that so hard for me?

It's my own (beep) line.

Who wrote this shit?