Wednesday, March 7, 2012

Underbust Survey- Part 3: Bust Sizes

Part 1, Part 2, Part 3, Part 4, Part 5

This was probably the most difficult part for me to write so far because there are a lot of dependent factors here and I wasn't sure of the best way to present the data.  Hopefully, they'll still be some good point to take away from this. :)

-  I had 179 respondents who included their bust size standing up and had an underbust size that I could use. 

What I wanted to see is how cup sizes are distributed across band sizes and if there were any conclusions we can draw from there.  In previous sections I counted women who were in between band sizes as half points in each band size.  In this case I doubled counted them because I was more interested in the distribution across the band size itself (so if a woman has a 27" band her cup size will be shown both for the 26 band range and the 28 band range).

For the cup sizes I simply subtracted the bust from the band size.  Clearly there is an art to fitting one's cup size and some women need to go up or down one cup size.  According to this article it sounds like women actually universally need to go up at least one cup size from their measured size if they have a D cup or above so if anything the cup sizes might be lower than they should be.

Here's the total distribution (I only used a range of 24-38 bands because I didn't have enough above and below that- you can even see pretty clearly that for women in the 24 and 28 band region that the data starts to thin out):

What I find interesting here is there is almost no correlation whatsoever with the cup size and band size.  We do have a  peak centered roughly around 7-9 for each band size (although it's easier to see below) but beyond that even in the smaller band sizes we see women having a 13 or 14" difference between their bust and underbust.  Actually, the handful of women who I had with a a difference of over 15" (not shown on the above graph because as you can see below the percentages were really low) all came from 32 bands or lower.

Note: In this case I simply subtracted the underbust from the bust without any assumptions about band size so there is no issue of double counting. 
So what can we take away from this?  About the best I can say is that we need a a very wide range of cup sizes.

Extending this point further, you often hear "Curvy" being used interchangeably with "plus-sized" but if you compare BMI to the difference between Bust and Underbust if anything you see the opposite
Like I said above the distribution is just insane.  You certainly cannot make any linear relationships here (unlike in Part 1 where you saw a very clear increase of the underbust size with BMI even if the underbust size was much smaller than previously thought).  However, what's interesting is that you don't see a correlation.  Many, Many women at the lower end of BMI need bras with large cup sizes.  Also, there are women at the high end of BMI who don't need large cup sizes.  There are also many women in between all of this.

I'd say the moral of the story is that we need many more offerings of cup sizes across all brands.  The practice that you often see where 28 bands stop at a smaller cup size doesn't appear to make sense in this context.  Now, I'm curious to hear back from the lingerie shop owners and what their experience is when it comes to band sizes vs. cup sizes.  Do you see any correlation at all between the two?  Or is it pretty much just random white noise like I'm seeing in my graphs? 

Just to let you all know.  Some of the topics that I still plan are covering:
- Bust variance (how bust measurements vary depending on standing, leaning over, breast perimeter, and laying back)
- Bust/Underbust/Waist measurements.  Do these converge to a method for standardized clothing?
- Is there a correlation between frame size and underbust measurement (I'm not 100% sure if I have enough data to see this but I'll sure try!)

Ok, now off to bed! :D


I want to also mention that in regards to bands being stretchy and how this might be affected as you go up/down in band size.  Bratabase has done some wonderful stats up looking at the stretchiness of bands, which you can see here. It looks like the larger band you have the more it stretches (with the exception of a 38 band but I wonder if that's more due to the very low number of 38 bands that were reported).

Another thing that becomes clear is that the bands stretch to the number shown on the label.  So even though the average 28 band is about 22" at rest, when you stretch it, it reaches 28" plus some change.

However, that's not all.  There was an interesting article in the Daily Mail that found that going by your underbust and bust measurements alone won't help you find your perfect fitting bra.  Ok, that might not seem that surprising because we all know that trying on many bras in person is the best solution BUT what IS surprising is that

For example, a back size of  34in and a bust measurement of 38in give a bra size of 34D. But according to the academics, this method is out of date, because it was only designed to go up to a D cup. Half of women in the UK now exceed that size.
In the University of Portsmouth study, 45 women were measured using both the traditional method and the best fit approach.
The researchers found that using the traditional method led to overestimating the back size in 76 per cent of cases and underestimating the cup size in 84 per cent.
SO, what that's saying is not only is plus-four wrong but in 76% of cases women with larger busts may need to go down another band size (so now we have the 2- method). 

That begs the question... where are all the 26 bands??? 


  1. Just wanted to throw this out there~ I love you!! This is all so amazing, all the data collected and analysis. So much work went into this, and it is so appreciated! I'm horrible with math and statistics was the worst course I took for my degree, but I follow exactly what you are saying.
    I've spent my evening looking at special order bras in 26 bands and feeling hopeless and like maybe I'm going about this all wrong. I had a comment on my blog today saying I have a case of "inverted letterphobia" and I was doubting my sizing. Anyway, this was incredibly helpful to me tonight and I really look forward to seeing what others have to say.

    1. Thanks! I've really tried to make this easy to read and understandable so I'm so glad to hear that you found it to be the case. :D

      I hear you on feeling upset about your size at times. I just dropped another inch on my underbust this past week, which puts me at 28". By my estimates that's going to probably put me in a 26 back by the time I'm done losing weight, so I'm definitely freaking out a bit! I'm hoping this will help bra manufacturers realize that there IS a market for small backs/large cups.

  2. Hi there, just letting you know I answered the survey again, this time with all the measurements. Hope you can rule out my previous answers.

  3. Fabulous, June! Looking at my records:

    * 28/30 Bands: Most of my sales are between F and G (which perfectly correlates with your 7 to 9" difference) although there are a few DD and E cup sales for the 30 band. So far, I haven't seen any woman higher than a G cup in those sizes.

    * 32 Band: I have a large concentration of sales between DDD/E and G with a small amount in D, DD, and J. The J cup sales are especially interesting because I have no sales in GG, H, or HH in this band!

    * 34 Band: Aside from no D cup sales, I do have a pretty nice dispersion in all the other sizes we carry. There area few DDs and DDDs but, again, the bulk of sales are between an F and G cup. There's also a not significant number of sales in the GG-K range.

    * 36 Band: The sales for this band are very similar to 34 with a slightly stronger concentration in the DDD/E through G range and less in the GG-K cups.

    * 38 Band: Again, a strong showing in the F-G range with a fair concentration in the DDD/E, GG, and H cups.

    You've done an amazing job with the survey!

    1. Thanks so much for sharing your statistics so far! It's interesting that it looks like we have fairly similar results. I wonder too how much your results are affected, though, by women who don't want to size down a band? It is interesting, though, that it seems like the largest cup size sales are in the 32-34 band range. But then again, it shouldn't be that surprising since that's what you said had the highest number of sales so most likely there's the largest amount of variation there. ;)

      Too bad I probably won't be going to your neck of the woods anytime soon because I'd break your 28 band stats. :D

  4. Penny said:

    More excellent work! While I think the main reason for small band sizes stopping at large cup sizes is the usual ignorance of womens body sizes, I expect there is also an element of them being aimed at teenagers, who actually will have smaller breasts. What's really odd is it used to be relatively easy to find 28A/AA and 28D+ but almost
    no-one did 28B or C!

    I actually expected this survey to indicate the average cup size to be larger for smaller band sizes as the same sized breast on a smaller woman is a larger cup letter. Maybe this is balanced out with the age variation, or maybe there is some truth in smaller breasts tending to
    be on smaller women, and the cup letter system actually compensates for it perfectly. If only manufacturers would realise that!

    1. Thanks Penny!

      I did to some extent see this but possibly it only occurs in a larger statistical sample. Like I mentioned in the post all the women who have 15" or above differences were in the 32 band or under range. I also saw that in the 38+ bands there were as many larger bands but I just didn't have a good enough sample size to say if this was just a coincidence or not. But you can see that cup size appears to go down slightly as you go to higher BMIs.

  5. Awesome work! Just wondering: which bust measurement did you use to determing cup size? standing up or leaning down?

    1. Thanks in this post I used standing because it was the most reported. However, in Part 4 (it's up now!) I discuss the variance in bust size depending on the method of measuring your bust. :)

  6. It makes sense that larger cup sizes occur across all band sizes. I wonder if there is some way of comparing the circumference of each individual breast to underbust measurement, rather than using the full bust measurement. If the difference between the bust and underbust gives us an estimate of the cup size, ie the relative volume of the breast tissue, then maybe looking at the absolute volume of the breast tissue would shed some light on the reason for the number of large cup sizes on slim women.

    I think the quote from the daily mail article is referring to the +4 method when it talks about measurements overestimating back size. When they say "back size of 34 inches" I suspect that is after they have added 4 inches. That said, I would expect the figure to be almost 100% not 76%!

    1. Zoggi. It would be interesting to see if there was some sort of way to estimate the volume of women's breasts. As is, it's quite difficult because breast shape varies so much from women to women. The best I could do for now is compare the variance in bust measurements (see part 4). However, maybe someone with more time and money as myself could do something like place breasts in a mould and measure their volume using that. Just a thought. ;)

      You know, I was quite surprised to see the daily mail article saying underbust=band size. However, it was a UK article so I gave it the benefit of the doubt. Also, I do think that it happens quite often that women need to go up a cup size (or more) from their standing bust measurement so maybe it is correct?

    2. It's easy to get a breast volume measurement without worrying about their different shapes. Use a water displacement measurement.

      To estimate the volume of a woman's breast:
      1. Find a semicircular bowl. Make sure this bowl will be large enough to accommodate your breast without spilling or tipping over.
      2. Fill the bowl half-full with water. Measure as you pour water in so you know its volume. Use tape to carefully mark the height of the water.
      3. Carefully, hold one breast flat against your ribcage and tilt slightly to one side. Carefully lower your breast into the water until it is covered with liquid. (Note: This is not an exact measurement. There's a little give and take at the edges of your breast. Also, if the water doesn't cover your breast, add more water and remark the water level.) A spotter can direct you to the proper depth to hold your breast in the water for an adequate reading.
      4. Without you moving, the spotter will carefully tap the new water level.
      5. You can now get out of the water and dry off. The water level will fall back down to the original level. Make sure the original level hasn't changed.
      6. Carefully measure out oil to fill the bowl to the higher level. If you pour slowly in small increments and keep track, there will be little mixing, and the oil should be reusable. If you measured carefully, you will hit the line and have an accurate measure of the volume your breasts took up.

      Realistically, while any woman could do this, how many want to dunk a breast into water and do conversions and subtract to get their breast volume? But it can be done! And with no higher math! The hardest part (in the US) would be making sure your measurements are in liters or cubic centimeters.