No comment 
.
It looks like the BRI only needs the waist and they use the hip measurement for Visceral Adipose Tissue estimates. But you’re right that many don’t know how to measure the waist. It’s not the same as the pants size.
I’m at 2.3, which it says is right in the middle of the healthy zone for my sex/age.
Oh - how to measure waist circumference
It’s easy to measure your waist. And it’s not just about your clothing size. Your waist circumference is a clue to whether you’re at higher risk for type 2 diabetes, high blood pressure, high cholesterol, and heart disease. And all you need is a tape measure:
.
Straight like parallel to the horizon or straight like following the shortest path around the contours of your body?
Like parallel.
Huh. What site are you using? Per the cited article:
FWIW I’m BRI 4.1 and BMI 27.2 (had dropped to 25 in May when peak marathon miles). If I measured correctly.
The one above. Its results for me:
Percent Body Fat: 19.6%
Visceral Adipose Tissue (VAT): 0.9%
Total VAT Mass: 1.1 lbs
Body Roundness Index: 2.3 (in the healthy zone)
Body Mass Index (BMI): 20.2
Looking at their references I’m not sure which one they would be using to declare what is the healthy zone. Per the article I reference you are potentially on the at risk low side.
From the article:
In this national cohort, we noticed that very low BRI was associated with a significantly increased risk of all-cause mortality, especially in individuals aged 65 years and older. This association seems plausible, as BRI was identified as a potential proxy measure associated with nutritional status,26 and very low BRI can be accompanied with malnutrition, fatigue, reduced activity tolerance, and muscle atrophy.25 The reasons behind the association between BRI and mortality may be epidemiologically and clinically plausible
Everything that I’m seeing is saying that a healthy BRI is between 4 and 6. I’m also seeing different formulas for it.
This would seem to indicate that it’s non-standardized - which could be problematic if Weight Watchers doesn’t step in to get people on the same standard.
The entry-point or the floor?
Seems… unlikely. If I had to pick a fitness deficiency, it would be in upper-body muscle mass. But BRI doesn’t take weight into account, and adding a few pounds of upper-body muscle wouldn’t affect my waist. The graphic shows:
Seems to indicate I’m right in the middle.
I’d have to grow a serious beer belly to hit a BRI of 4.
Statistically at higher risk. An association. But of course no individual is the pooled statistics. Likely, like BMI, it will, if used widely, be worthwhile as a screening device. Not a be all end all. Maybe an improvement though.
FWIW by BMI alone your low number (20.2) is associated with higher mortality than someone with a BMI of 25.
Not sure I understand the question.
Does that control for things like drug use, eating disorders, etc.? The BRI paper above seems to just look at all-cause mortality. The irony is that a healthy weight could be considered an abnormal indicator if the average person is expected to be overweight.
When the covid vaccine was still in short supply, they used BMI rather than any more accurate measure of being over weight) as one of the qualifying criteria. Every state used slightly different rules, so ymmv, but my state, and most states, used BMI. I bet that BRI would have been better.
Measuring my waist and hips takes about as much time as measuring my weight. Tape measures are likely more accurate than many home scales, and take up less space. And my circumference doesn’t change as much during the day as my weight does.
I weigh myself every morning at the same time and same point in my digestive cycle. It has about 1.5% noise, and probably less than 1% if I sorta mentally average over time. But my waist measurement is probably up to an inch off, which is more like 3%.
Controlled for possible confounders as past as possible. Smoking status actually reported as different curves. There are several studies but here is possibly the best.
I don’t see eating disorders or drug abuse as listed potential confounders however. Smoking drinking education and poverty levels … trying to control for non apparent pre existing disease, yes.
My digestive cycle is not as regular as yours. 
It took some training on my part. Part of my 4S morning routine: shave, shit, scale, shower.
Should I measure his center or his knees?
Yes it looks at all cause mortality. Can you explain more what you mean about that irony though?
The point of these studies are to determine which range of the indices actually are associated with lowest all cause mortality rate (as a meaningful definition of being “healthy”). As made very clear by @DrDeth’s cite, the 25 cut off for “overweight” was very arbitrary. What BMI and BRI ranges are associated with the lowest mortality risk? In the cited article BRI ranges as listed seemed to have more predictive value.
Suppose you had a person who would be overweight under normal conditions–like most, they eat more than they should and exercise less. That isn’t great, but only represents a mildly increased mortality risk.
So what would counteract that natural tendency? Smoking. Other drugs. Bulemia. Cancer. Etc. They might reduce the weight to a “healthy” level but not in a healthy way.
How it actually plays out in the statistics I don’t know. But since these conditions represent a large mortality risk, even if they’re relatively rare they could dominate any mild effects from being slightly overweight.
The paper above indicates that a BMI of 20-22 is optimal when they had a long-term follow-up (in never-smokers). I’ve had the same BMI since high school and no chronic pulmonary diseases (or acute ones if you exclude mild cases of colds and COVID). That would seem to exclude the possibility of pre-clinical diseases.