You need to get over the paranoia, and either just get on with the job, or get to grips with off-camera metering, and work out how much of what you can do, is actually useful.
I have no idea about smart phone's, let alone smurphone 'apps'. I did NOT get on with touch screen things, my fingers is too big, and the do it all, in a nutshell, jack-of-all trades master of none, thing about them made me want to chuck the thing at the nearest wall! Like when daughter said 'Here Dad, take a photo of us!" and it took her longer to find the frigging camera mode for me to use, than it would for me to find a proper camera, and a film, in the attic! But still...
There are two main means of metering; 'Incident' metering; measuring the ambient light falling 'on' your subject, and 'reflected' metering, measuring the light reflected 'off' the subject. Most cameras with an inbuilt meter, take reflected meter readings. Hand-Held accessory meters can usually do both.
With a reflected meter reading, the niggle is that it will give you a suggested Exposure Value, EV, based on how much light it sees. So if you are pointing the meter at a white rabbit on a ski-slope, it will give a higher EV than if you are pointing it at a black cat in a coal hole, because its measuring the amount of light being bounced back off the subject. If you take an incident metering, you are measuring the light falling on the subject, so you will get the same EV for the rabbit on the ski slope as the cat in the coal hole, and it may be argues that what you get as a result is more accurate or more natural... so which is more appropriate for what you hope to achieve? Here starts the art.
Now you go from decisions over incident vs reflected, to some sort of compromise or 'average' taking both an incident and reflected reading; looking at your subject, and then trying to work out some sort of average to use in the middle. Here starts the 'Craft'.
BUT the key is to look at the subject, and work out what you hope to achieve and then what sort of metering method would be more appropriate.
Tear-Away tot.... running in and out of shadows.... tricky; you likely want to keep the shutter speed up, or the kid will tend to streak; and you have changing light conditions that would have you chasing around where tot be, to take an incident reading for each shot, which tends to suggest that reflected readings may be the more appropriate... b-u-t depends. I have used f16-sunny and a couple of hand held incident readings to come up with base settings to use when shooting O/H's tear away granddaughter, and then tweeked then a tad, when she's hidden under the climbing frame, or leapt onto the shiny slide. It C-A-N be done, and you don't need to get too bogged down in the technicalities...
But that is, I think the danger here, getting yourself tied in knots, trying to do it all, shot by shot, like the computer in a widgetal camera tries to. More b-u-t, it's a case of understanding metering methods, and more importantly, working out where and when any particular method may be more or less helpful to you. As said, it's 'craft'; technique over technology.