Your findings are consistent with much of what you saw on the Internet and also what my experience was.
I’ve come to the conclusion that there is no defined answer on the topic of the “ideal” temperature in as much as it is a range of temperatures where the tradeoff is a triangle drawn between stringiness, layer lines and strength.
The only other tip I might share that I saw recently on a YouTube video was a very unscientific approach that the YouTuber admitted was unscientific but better than nothing at all. What he did was to take a small screwdriver and break off the tip of the tiny tower inside of the temp tower. Based on the “feel” that would let him know what temperature produced a stronger bond. Now Stefan over at CNC kitchen uses his DIY strength meter and that is a lot more scientific but I mean C’mon now, how anal-retentive can one be? This is supposed to be fun not work. Nevertheless, I love Stefan’s videos because of the fact that he goes that extra mile that I just simply won’t do.
Here’s a tower that I ran last night for Sunlu Transparent PETG. These are the tips that the YouTuber said to “feel” the breaking point to gauge relative strength. Didn’t work for me, I guess my fingers aren’t calibrated for force. 
But if you look at both ends of the temp spectrum on that one tower, all it shows is that this particular PETG is going to string at either end of the Spectrum. The MFG calls out 220-250°C. Well to me they look like cr@p at both ends so needless to say, I’m underwhelmed and I have to resign myself to stringiness. Now others will quickly point out that I could mess with the retraction settings. Indeed I could and waste 5 hours of my life with only marginal benefits.
Look at these two closeups and you tell me if they look very different. These were done at a 0.20 layer height. Like I said, underwhelming results.
