And no, I don't remotely buy that 1/3 octave smoothing in that phone app is going to give you any kind of useful data to fix room modes. It simply is not the proper tool for the job (room modes can be as narrow as 1/12 octave). People should spend the $100 on microphone and use REW to measure and optimize the room properly. That will make a huge difference in how clean and open the system will be (fixing room modes).
Depends on the app and the frequency, and the phone mic.
For my phone, I would not trust ~45 hz and below. Because it is wrong. However, for higher than that the app I used to use DID show big spikes. So a section might look pretty flat, but above that it might say "129" or "65", which is the strong peak reading in that area. That was good enough for some simple eq and to check the effects. It gave no data on the magnitude of those spikes, btw. Just that something was there.
Those spike numbers were accurate in terms of frequency. Confirmed with REW at a later date.
So I think it can help. It certainly helped me, in a relatively accurate way (meaning better than by ear), before getting a calibrated mic and REW.
But you have to know the limits of the tool. For phone apps and the mic on my phone, the limits mean I should NEVER use the phone app for sub measurements.
That said, I have to agree that sub integration is way trickier than people who have never done it might think. And much harder to learn how to do than people who now do it easily might remember.
---
If I were to critique the video, which I skimmed at 2x speed, I would say front load the speaker and speaker set up, push the sub to the end, and mention that it will need some work to integrate (a second video would not be out of line, or pointing to the many sub set up videos that exist would work too.). Get the 2 channel right, then add sub.
I think the main issue is that the video did not go basic enough.. A lot of knowledge seemed to be assumed. Given this is likely most useful for new people, you can't go too basic.
If I were to make such a video? Likely 2 channel set up using experimentation and ear. That's first. Second, measurement of that set up and comparison with the a second set up (maybe toed in slightly and then right at dead center, or right against the wall and out 3 feet). And I would talk about phone apps, and non-calibrated mics with REW (what can be done and what cannot be done with those methods*) as well as what you could do with REW and a calibrated mic (briefly). Finally, adding a sub.
My reason for that is trying to think about where people who might follow such suggestions are coming from, and thinking about how to get them to follow a path that leads to better and better results. So choice and set up (basic speaker positioning information), measuring and simple EQ (seeing what's going on, how position affects that, and how to do some corrections), sub integration.
*short version: you have to treat the measurements as reliable (test retest gives the same results) but not particularly valid. So you can probably use them to look at variations in the 200-300 hz range, but never compare 200 to 5000hz.