Pattern Reviews,  Sewing Tips

Good Drafting but Poor Grading?

Apologies in advance for my erratic posting schedule…or lack of schedule! Sometimes I have one big idea that requires a lot of analysis and research. Other times a single concept has several smaller aspects and it seems better to break it down. That’s when the posts come at you more frequently…but they’re shorter. This is one of those times:  one sewing pattern (SLPCo “Haiti) prompted several thoughts simultaneously.

When a sewing pattern has fit issues, people ask if the problem is in the drafting, or the grading. Drafting is creating the shape and proportions of the pattern in a central fit-approved size, then grading is the next step of developing a size range. In a previous post about drafting, I stated that “Poor grading may coincide with poor drafting, but not necessarily.”

Is it possible to have a poorly-drafted pattern with beautiful grading? Sure, a patternmaker could draft a pattern with wonky armscyes, an ill-fitting neckline, out-of-proportion collar, an imbalanced front-to-back crotch ratio…all sorts of fit problems….and yet send it out to a grading service that perfectly prints out a nest of sizes. You can’t fault the grader for the original drafting fit issues. For example, I have  a stack of McCall’s children’s patterns that are accurately graded, but the original draft was way off (inside-necklines are enormous, armscyes are huge)….so the whole pattern is bad.

In my own experience with indie patterns, it’s more often that the drafting is fine, and then the grading is off.  My guess is that while industry patterns are graded by dedicated graders, or sent out to a grading service, indie patternmakers often do their own grading in-house….and their grading skills may not be as polished as their drafting ability.  Unfortunately a pattern well-drafted in the central fit-size can be made inaccurate in the rest of the size range by inconsistent grading.  The result will be a pattern that “runs small” in some sizes while it simultaneously “runs large” in other sizes….often causing lots of argument in online pattern groups, because everybody is right! They’re just looking at things from a different perspective, depending upon which size they’ve sewn.

An example of a pdf that is nicely drafted but inconsistently graded to the body size chart is the FREE Haiti dress pattern that I reviewed previously. I did mention before that the grading is uneven (there is less fitting ease in the larger sizes than in the smaller sizes, and the armscye has  excessive depth in the small sizes) but I did not elaborate on it at the time because it truly wasn’t necessary to adjust the grading for the purpose of participating in the charity cause,  since a wide variety of sizes are needed. That said, if you are using the pattern to sew for a particular child, more consistent grading would be helpful.


The Haiti dress is the easiest style to analyze as it has just a single pattern piece, and is meant for one type of fabric (woven cottons), and has no fitted areas (waistline,  darts) where stitching would slightly alter the fit. (In a future post I will explain how stitching, lining, fabric choice, construction details can all affect measurements of the finished garment).

The Haiti dress pattern pattern:

  • “runs big” at the small end of the size range
  •  “runs small” in the larger sizes 

This is true for both the armscye depth, and the chest width.

In children’s pull-over dresses or tops, it’s critical that the fitting ease is adequate at the chest width, because not enough ease can result in a garment that cannot go over the shoulders. The reason for this is the undeveloped bustline, which means that the chest measure runs a few inches smaller than the around-shoulder measure…and the garment needs to pull over the shoulders. As the bust develops, the chest measurement increases in relation to the around-shoulder measurement. A full adult bust may have a larger measurement than the around-shoulder. But for a child, we’re looking at this situation:


For fitted tops and dresses with a waistline, a placket going into the skirt will add the extra width to allow the garment to slide over the shoulders.  A slim dress such as a shift or sheath can have a zipper to make getting dressed easier. But in an unfitted style  such as the pillowcase dress (no zipper or placket), there must be enough ease in the pattern itself at the chest measure to slip over the shoulders. Generally there must be 15% to 20% of ease.

Note that ease is measured as a percentage, not as a specific amount. As Kathleen Fasanella explains in her Entrepreneur’s Guide to Sewn Product Manufacturing:  “as people get bigger, they get proportionately bigger“. (See previous post for an explanation of this.)

Back to the Haiti pattern: here’s how to go about making the ease consistent.

It’s a bit difficult to figure out what’s happening with the grading because the pattern is graded with a stacking point at the neckline….so the armscye points appear a bit jumbled (yes I understand they wouldn’t be if you print layers separately, however I always print the entire range to check the grading):


It is easier to see the differences between sizes if a pattern is graded by using the bust level (or waist level) as the stacking point:


But that’s okay, I can work with this.  I see two issues:

  1. The ARMSCYE DEPTH  grade is minimal between sizes (I prefer a grade rule of  1/4″ between sizes)
  2. The CHEST WIDTH grading is “jumpy”, and negligible between  sizes 4,5,6:


So I’m going to adjust the grading first for armscye depth, and then for chest width.

Also, let me clarify that this is not the “real” way to grade a professional pattern, it’s a short-cut method that works only with simple unfitted designs…such as this one with just a single pattern piece, no darts, no fitted waist or neckline.  “Real” grading shifts the fit-approved-sized pattern pieces along the X and Y axis according to grade rules for each measurement point, to create a “nest” of graduated sizes.  More detailed patterns have multiple grade-points to consider, and so multiple grade rules…and often require thoughtful consideration  of how each body area “grows” (for example, neckline width doesn’t grow nearly as much as bustline width).

The purpose of showing this quick method is to clarify what the end result of consistent grading should be, not necessarily the best method to get there. I just want you to see why your dress might “run small” or “run big”….depending on which size your are making.

Armscye: to my eye, the size 7 has a good depth, so starting there, I marked the size 7 depth by drawing in a horizontal line:


Then I marked the rest of the horizontal lines 1/4″ apart to create the new armscye depths going up and down the size range:


Chest width:

The chest width in the pattern is “jumpy”: wearing ease is inconsistent between sizes.  Here I’ve made a chart of the chest size of the body measurements given on the pattern (blue) , and the finished garment measurements calculated from the pattern pieces (pink).  (Note: if this were a fitted style with a waist seam, the actual garment would be slightly tighter after stitching, as each stitch “pulls the garment into spec”…but since this is a loose unfitted style, the circumference of the pattern pieces will be almost identical to the circumference of the finished garment).  If you subtract the body Chest measure from the finished Garment measure, you’ll have the amount of Ease (green). Dividng the Ease by the Chest measure gives you the percentage of wearing ease by size (yellow).  It ranges from 22% at the small end of the size range, down to just 10% at the upper end:


As a general rule, a child needs at least 15% garment ease at the chest, to fit around the shoulders when pulling a garment over the head, if the garment has no stretch and no zipper or placket opening. 10% (in size 12) is not enough ease to pull a garment over the head. The dress could get stuck on her shoulders. A more realistic percentage of ease is 18″.  Here is what the garment measurements would be with 18% wearing ease incorporated across the size range (yellow shows the percentage of ease, green shows the inches of ease, and pink shows the new pattern measurements):


I prefer using the industry-standard of a consistent grade rule between sizes, so I decide to regrade with a 1/4″ rule (total circumference 1″) between sizes 2 through 7, and a 1/2″ rule (total circumference 2″) between size 8 and 12….here are the revised pattern measurements:


Since the garment has the same width front and back, it’s simple to divide the pattern chest circumference by 4 (left/right/front/.back) and use these measurements for drafting. The new pattern measurements are here in pink:


The next step is to transfer the chest pattern measurements to the paper pattern.  The black dots show the new chest measurements from the center-front (or center-back) to the armscye, at the new armscye depth level….the colored vertical lines show the revised side-seams:


Last step, re-draw the armscyes (using a French curve):


And then the new side-seams.  There is now more distinction between the sizes, and the fitting ease is consistent through the smaller size range, and then through the larger size range. :


The revised pattern will now glide overhead easily at every size, and have a more consistent fit when worn.


Remember this is a FREE pattern! And also there is still plenty of time to participate in the Real Hope for Haiti initiative (info on the Simple Life site) which gives you a $5 credit towards a future pattern.

Happy Sewing!

Best, Janet


  • Bunny

    Another fabulous post, Janet! You teach us all so much and raise the bar for what we should expect in our patterns. Thank you!

  • Suzy

    I just love how simple you make this all sound. Thank you so much for your blog, it’s a joy to read!

  • Saskia

    You make this all look so simple! I really enjoy reading your blog posts and I’m taking so much from your experience. Like someone else had mentioned, if you wrote this in a book, I’d certainly buy it.

  • [email protected]

    Hi Saskia! Well yes it’s simple this time, only because it’s a simple pattern. Grading a detailed pattern with lots of fit areas is very complicated and an artform in itself. It’s well worth the investment for a patternmaker to send out a pattern for professional grading, if at all possible. I see soooooo many lovely patterns from indie designers, that are then rendered not-so-great by inaccurate grading.

    Thank you for your sweet feedback!! Best, Janet

  • Andrea

    How much ease should there be in knitted items? Tightly fitted knits like boxers or loosely fitted knits, like play shorts with knit binding at the waist and leg opening? I get that there is negative ease in tight fitted items, but to what extent?
    How do we true a pattern in places like at the crotch seam of knit boxers where three or more parts come together? (Can you tell what I’m working on these days? haha!)

  • [email protected]

    Every knit fabric has its own stretch qualities, so the ease depends on how the fabric behaves. In industry, EVERY new knit planned for production gets its own pattern, and its own fit-approval. There could be negative ease, as in four-way stretch knit for swimwear or shapewear that stretches to fit the body. There could be small percentage of ease, as in jersey meant for teeshirts or play shorts. If the knit is an interlock with minimal stretch, the ease will be more like that for wovens. All I can advise is test, test, test! As far as trueing with multiple pattern pieces, as long as they all add up to 360 degrees the pattern will work. Simple patterns with 2 pieces on one side and 2 pieces on the other side (such as left front/back and right front/back) measure 180 degrees plus 180 degrees (= 360 degrees)…but with additional pattern pieces you’re just splitting up that 360 degrees in a different configuration.

  • rita penner

    There seems to be a swell of interest in your creating a book encompassing your advice and knowledge. I feel this blog is already this book. All you need to do is create a table of contents, perhaps an index of topics pointing to different books. Thanks for all your hard work!

Leave a Reply

Your email address will not be published. Required fields are marked *