What Even Is It?

Sweaters designed with a raglan sleeve can be identified by the uninterrupted diagonal “seam” — which is often not a real seam but rather a line of increases or decreases that creates the appearance of a seam — stretching from the underarm to the collarbone.

Unlike many sweaters, raglans have no shoulder seam. Instead, the top of the sleeve becomes the shoulder of the sweater. This means the “shoulder” of a raglan sweater is fairly loosely defined, and fairly relaxed and easy to wear.

Since it allows for greater freedom of movement than set-in-sleeve or drop-shoulder construction, raglan construction is often used to make sportswear. The “baseball T-shirt,” with three-quarter-length colored sleeves and a gray or black body, is probably the most famous raglan style, but a lot of sportswear is made with raglan construction because the sleeve and shoulder construction allows for easy movement. If you start looking for them, you’ll notice them everywhere — in winter coats, cozy sweatshirts, exercise clothes, and a huge number of contemporary sweater patterns.

How Raglan Is Accomplished

The most common way that knitters construct raglan sweaters these days is in one piece from the top down. In these patterns, the first row contains foundation stitches for the entire sweater: back, front (in two pieces), and two sleeves. But the foundation row consists primarily of stitches for the back neck, with just a handful of stitches cast on for the very top of the raglan sleeves and even fewer (often just one stitch) for each side of the front.

The first row also often includes four “seam” stitches that are marked so the knitter knows where to place the raglan increases.

In order to create some depth for the neckline on the front, the sweater is knit back and forth for at least a few inches; after a few inches, the two fronts will be joined. The knitter increases on both sides of every seam stitch on right-side rows until the front neckline depth is reached, and the two sides of the front are joined together. Then the knitter carries on, working now in the round, until the sweater reaches the desired length to the underarm.

All of the parts of the sweater grow on a diagonal away from the cast-on row, so that the sweater gets wider and the stitch count increases on every right-side row.

If the sweater were continued in this manner for thirty inches or so, it would become a poncho. 

Instead, when the knitter reaches the underarm, both sets of sleeve stitches are taken off the needle and put on waste yarn, and the front and back are joined together at the armpit with a set of new cast-on underarm stitches added in between them. These new stitches create ease of movement at the underarm.

The body is then knit downward until the desired length is achieved. It might be widened, for a kimono-style shape; it may be knit straight down, for a standard pullover silhouette; or it may be narrowed with waist shaping. Length options range from cropped just beneath the underarm to thigh or dress/robe length or anything in between.

Here’s a summary. Note, though, that the picture shows the body and sleeves flat, but in reality they are tubes.

Limitations of a Constant Rate of Increase

In the most straightforward possible raglan design, the knitter increases one stitch on each side of every seam on every right-side row to the underarm. This means that the front grows by two stitches on every right-side row, the back grows by two stitches on every right-side row, and each sleeve grows by two stitches on every right-side row.

In this approach, with the rate of increase held constant on all pieces of the sweater, the ultimate width of the front, back, and sleeves at the underarm — usually their widest point — is determined by (a) how many stitches were cast on in the first row and (b) how many rows the knitter completed to reach the underarm. That is, if you need a wider front, back, or sleeve, in this method the only way to get it is to either start with more stitches or knit a deeper armhole.

You can see in the illustration to the right what this means for sizing. The first, straight-size L pullover, has an armhole depth of 9 inches. To achieve a wider body for a 1X sweater, the second pullover, with the same rate of increase, has an armhole depth of 11 inches. And for a 3X sweater, the third pullover, with the same rate of increase, has an armhole depth of 13 inches.

Note, too, that the arms grow in circumference along with the armholes, so that the largest sweater has proportionally larger arms.

To avoid having such deep armholes while maintaining the same rate of increase, another option would be to cast on more stitches in the foundation row — which would create a wider neck opening.

The problem, though, is that someone with a larger body does not necessarily have a larger neck, larger arms, larger shoulders, or a longer distance from neck to underarm than someone with a smaller body. All bodies are different. The constant rate of increase found in “classic” raglan sweater accommodates only a limited range of difference.

Changing the Raglan Increase Rate

Luckily, there is no rule that says raglan increases must happen at the rate of one stitch on each side of every seam on every right-side row. The rate of increase can be different for every part of the sweater: sleeves may be increased every fourth row, for instance, while the back and front are increased every other row. Moreover, the rate of increase can speed up or slow down for every part of the sweater: the fronts can start out increasing every fourth row and then, when they get closer to the fullest part of the chest, can be increased every other row, or can increase two or more stitches every time.

You can probably imagine that these approaches will affect the shape of the raglan line. It won’t proceed at a precise 45-degree angle from underarm to collarbone if one piece of the sweater is growing at a different rate than its neighbor. Due to the stretch and malleability of knits, however, the difference is not always as visually apparent as you might expect.

Bodies that are infrequently accommodated well by off-the-rack clothing are unlikely to be accommodated well by standard raglan shaping. A customized approach to the rate of raglan increasing requires additional planning and math, but it also produces better fitting sweaters. It’s worth taking the time to figure out your desired measurements and work through the planning in order to get the garment you want — which is where I’ll pick up next time.

It’s funny, really. We are awkwardly configured blobs of limb-studded flesh wrapped around articulated skeletons with hundreds of individual bones. Our bodies are shaped by our genetics, our choices, and our experiences (both voluntary and involuntary), and for many of us, this makes buying off-the-rack fashion a complex proposition.

I can’t buy a pair of jeans off the internet with any kind of reliable success. I have a deep curve at my lower back; a high, round butt; powerful thighs; calves so impressive that if I buy denim insufficient elastic, I can’t pull the legs up past my knees. I know my sizes, or at least what they’re supposed to be. I buy jeans made by brands that provide size charts, and I match them dimension-to-dimension to other pairs of jeans I already own and wear, and yet when I tear my new pants out of their mailers and attempt to yank them onto my body, nine times out of ten, it’s a no go.

We are miraculously complicated creatures, is what I’m saying. Our clothes are approximations, echoes of the shapes of us that we zip and button ourselves into, hoping they will feel the way we want them to, touch us and not touch us how we want them to, move with us how we want them to, and, most of all, look the way we want them to.

Sometimes they fulfill our expectations. Often, though, the clothes we buy are not quite what we thought they would be. They don’t fit, or they fit but not how we wanted, or they fit how we wanted but move wrong, or wear wrong, or feel wrong. We’re used to this. We expect it.

Let me say that again. We expect for there to be a gap between the dreams we have for new garments and those garments-in-reality. We expect to be forced to adjust — so much so that the adjustment usually happens without comment. We don’t get angry when a sweater we bought off a picture and a size chart on the internet turns out to be a little shorter than we imagined. We adjust. We learn to live with what it is instead of what we wanted it to be.

And yet, when we knit sweaters, we often expect them to turn out exactly as we hoped. If they don’t, we consider them failures.

We apply this binary even knowing that knitting a sweater is more complicated than cutting and sewing knit fabric into the shape of a sweater. A cut-and-sew sweater has multiple pieces and relatively firm seams. A hand-knit sweater that is made in one piece from the top down is a single seamless chain of looped string, which we anticipate — outrageously — will be able to be draped over a body without changing shape in any way.

We expect our knitting to behave. In fact, we expect it to behave in a way that nothing does behave. We expect it to behave in a way that defies physics and human experience and everything we collectively understand about art. And we hang the weight of these unrealistic expectations on gauge.

Specifically, we hang our unrealistic expectations on The Gauge Swatch.

Gauge Is Not Magic

Conventional knitting wisdom says that if we knit a four-inch-by-four-inch square in the same stitch pattern as the garment in our mind’s eye, using the same size needles we’ll use for our garment, and if we then wash and block that square in the same way we will wash and block our garment, and measure it carefully, we can discover exactly how many stitches and rows of knitting we need to make to produce the garment. With our correctly executed Magic Gauge Swatch in hand, we can sally forth to plan a sweater that will turn out the size we intend it to be, after which point we will swan around in it, smug and delighted with ourselves and our superior brains.

I mean. Guys. There’s a reason knitters say that swatches lie. When you make a gauge swatch, wash and block and measure, plot and plan, and then end up with a garment that is eight inches wider and six inches shorter than you intended it to be, the feeling of betrayal is intense.

But I swatched! I knew my gauge! The gauge swatch must have LIED.

It did not. There was a lie, but the lie was not emitted by your swatch. The lie was the premise of the Magic Gauge Swatch itself. The lie was in the unrealistic hope that a four-by-four square of fabric lying flat on a table could tell you everything you needed to know to make a sweater that would fit your three-dimensional, alive, moving, complex human body.

It’s such a bummer that it’s a lie. But it is. I’m sorry.

So. What, then?

Obviously, we can’t give up on the notion of planning altogether. We do still want to make things that fit, and we want them to fit in particular ways, and we even want to be able to write down directions that will tell other people how to make things that fit them. I have a few thoughts on how to proceed, but first I want to resist what is so often the thing that comes after this question, the immediate next step my own brain wants to take, which is to propose a New Method.

Q: Well, if the four-by-four swatch isn’t enough, what DO I need to know to make a perfect sweater?

A. Oh, don’t worry. All you have to do is…

We shall sidestep that pit of vipers. It’s no good to replace one lie with another lie —and keep in mind, the lie that there is one way to achieve perfection is backed up by the lie that is the mirage of perfection itself. The lie is the idea that we can spend a week or a month or half a year poking around with sticks and string and then in the end not have to do the part where we adjust our expectations to reality.

The lie is the hope that if we work hard enough, if we try hard enough, if we plan hard enough, we’ll make something that is Actually Perfect.

I have been a maker for a long time, and I’ve known a lot of makers, and I’ve talked to quite a few makers, so I feel pretty confident in saying this: You will never make anything perfect. Even if it looks perfect at the moment you finish it. Even if seems perfect right at first. Even if you did every step of the making perfectly. Even if. You will never make anything perfect.

But. (There is always a but.) But, you will make things that turn out to be, if not perfect, then good. Good enough. Good in ways you hadn’t imagined. Good, not for you, but for the people you decide to give them to. Good, not for the body you had when you made them, but for the body you end up having five years in the future.

You will make things that are good because they satisfy you, good because you have such beautiful memories attached to them, good because they feed your soul, good because they kept your hands occupied when your mind was fretful and grieving, good because they remind you of someone or some place you never want to forget, good because the yarn was so gorgeous on the needles, good because you gave them to your kid, who gave them to her kid, who, long after you were gone, wore one of the things to made on an autumn afternoon at a park in a town you’ve never been to.

You will make good things. You will make so many good things, and they will bring you joy — if you can let go of the mirage of perfection.

Still, Planning Is Possible

So let’s say, for the sake of argument, that we have broken the spell, released ourselves from the mirage, and now we are ready to knit a sweater without the expectation of perfection.

I mean, we’re not ready. We are frail humans conditioned by decades of brainwashing by sexist, racist capitalism, and the best we can do most of the time is to recognize that we’re poisoned and write Progress, Not Perfection, Baby! and Smash the Racist Capitalist Patriarchy! on the whiteboards inside our brains.

Nonetheless, let’s pretend.

It seems pretty clear that we can’t throw out gauge altogether. At least, if we hope or intend to make something of a particular size, we still need a way to narrow down the possibilities of how many stitches to begin with, and knitting a gauge swatch gives us a starting point a lot more rapidly than, say, knitting an entire “test” sweater, washing it, measuring it, and then ripping it out and knitting it again to the right size, which no one would want to do, because it would be horribly boring. If we’re willing to commit an hour to making a four-inch-by-four-inch gauge swatch, washing, and measuring it, we’ll at least end up with some ballpark numbers to use in planning.

If we can stand to do it, we can even make that test swatch bigger — say, as big as one whole ball of yarn from the project — or we can take a card from Elizabeth Zimmerman’s deck and make a “test hat,” which gives us a chance to play with the yarn and the stitch pattern, and which will most likely fit someone even if it doesn’t fit us, and which produces something a lot more useful than a swatch.

We might not be able to stand it, though. I’ve made boatloads of swatches, but very few wider than four inches (sometimes they’re longer), and exactly zero test hats, despite thinking it’s a clever idea.

And anyway, no matter how much or how well we swatch, at some point we’re going to have to get comfortable inhabiting reality. The reality is, there’s only so much a swatch can tell us. This means we’re going to have to check in on our garment while we’re making it. If we’re knitting it in pieces, it probably means blocking and measuring those pieces after each one is finished to see if the size is working out. If we’re knitting our whole sweater in one piece, it means stopping to try it on and check how it’s fitting.

“Inhabiting reality” also means making some educated guesses about how our knitted fabric will behave after it’s all finished. Will it get bigger? How much bigger? Is it going to be heavy, and is the weight of it going to pull in any particular direction? Is the stitch pattern going to pull in even more when I’ve knit a whole big piece out of it? Does this kind of yarn have a habit of shrinking or growing in a notable way? (Silk and linen, I am side-eyeing you.)

I once read a knitwear design book that recommended pinning your swatch to a cork board for a few weeks to see how it behaves. It’s been thirteen or fourteen years since I read that, and I still haven’t done it, which suggests it isn’t a tip I’m going to adopt in my knitting practice. And that’s part of inhabiting reality, too — paying attention to what you are willing to do, how much you care about the exact fit of your finished object, and how much work you’re interested in taking on to get as close as you can to the size you want.

The truth is, you can knit and block an entire test sweater and wear it around awhile before ripping it out and re-knitting it to the size you truly want. It would be a tiny bit weird to do that, but no one is stopping you. Plenty of people knit the same sweater more than once, adjusting the fit of the second to make up for the deficiencies of the first. Plenty of people also use a favorite yarn over and over again, and the more they use it, the more information they have about how it behaves. Maybe doing that kind of thing would cramp your style, but you have no problem with taking your top-down sweater off the needles to block it when you get to the armholes so you can try it on and see if you’re good to go or if the fit is so far off the mark that you need to start over.

Which brings me to a question that connects to one of the more painful dimensions of inhabiting reality: How willing are you to rip your work out and try again? If you’re dedicated to the proposition of making sweaters that fit the way you want them to, and if you don’t have an unlimited yarn budget, it’s hard to get around it. Sometimes you have to go backward to make forward progress.

Say It with Me: “Process is My Jam”

I’m such a fan of forward momentum. Aren’t we all? But the longer I hang around the earth, and the older my kids get, and the more stuff I make — novels and sweaters, dinners and Instagram posts, couch pillows and backyard videos for the church pageant — the more I’m trying to embrace process.

Process is where it’s at, if you’re trying to make a sweater that fits. Process is everything.

So I’m going to give this a go. I’m going to make a sweater that fits (at least mostly) the way I want it to, using a process that involves a gauge swatch (see above) but does not pretend gauge is magic or that perfection is a real thing.

To make this journey spicier, I’m going to knit a raglan sweater knit in one piece from the top down, with a fancy cabled pattern on the body. Also, I have a limited quantity of yarn that is maybe not quite enough. (If you knew me, you would know that I always have a limited quantity of yarn that is maybe not enough.) This sweater will have a V-neck with deep ribbing and wide sleeves with hemmed or loosely ribbed cuffs. The front will be bigger than the back and will have short-row shaping at the bust. How long the body is will depend on how far my yarn goes and what my wife tells me she wants when I get to that point.

Before I start, I’m going to make measure my swatches, figure out my desired dimensions, and do some math. I’ll post about all that in case top-down raglan planning is something you’ve been missing in your life. After I start knitting, I’m also going to inhabit reality by keeping an eye on how it’s looking and by blocking the unfinished sweater when I get to the underarm. At that point, I’ll make my wife try it on, rip it back if I have to, and carry on from there.

Keep me honest, will you? Let’s see where the process takes us.

Most of the time, set-in sleeves are knit upward from the cuff. When you get to the underarm, you bind off at least an inch of stitches on each side, then decrease at each end of every right-side row until about three inches of stitches remain. At that point, you rapidly decrease the rest of the stitches (usually double the established rate of decrease) and bind off the final inch and a half. This makes a bell-shaped sleeve cap that has fewer rows than the arm opening, so that the flattened top of the sleeve cap can be eased into the shoulder.

If that doesn’t make a lot of sense to you, written out like that, well, join the club. There is something a tad mystical and perplexing about how set-in sleeves work and how they can be made to fit —at least for those of us (including myself!) whose ability to understand things spatially is somewhat deficient. Would a picture help?

There’s a lot to recommend the classic approach to knitting set-in sleeves. But what if, say, you’re knitting a sweater with a bag of random balls of yarn that you bought off eBay because you were charmed by the handwritten tags identifying them as the natural-dyeing experiments of a crafter from the 1960s?

And what if you’ve knit the front and back but haven’t weighed the yarn, and eyeballing it suggests you have enough for sleeves — probably — but definitely not enough for full-length sleeves, and you want to use every single meter of available yarn, but you super don’t want to run out of yarn a few inches from the top?

In that situation, top-down sleeves are your new best friend. Just knit those sleeves for as long as you can make it work, and when you start to run out of yarn, change gears and knit cuffs. Ta-da! You have a sleeve of exactly the right length — which is to say, you have a sleeve of a length that reflects however much yarn you happened to purchase.

Of course, when I say “you” here, I mean “me.” I know you didn’t buy that eBay yarn lot, because I did. I am the only person who would get myself into that particular situation. But you might find yourself wanting to maximize sleeve length without running out of some precious yarn of your own choosing, and top-down sleeves are perfect for that scenario.

Knitting set-in sleeves from the top down is also handy if you’re designing your own sweater and your visual planning is less than fantastic. I have made many sleeve caps, yet I can’t say that I exactly understand at a cellular level how they fit into armholes. I do have faith, however, in my ability to figure out as I go whether or not something is working. Knitting a sleeve cap right into the spot where that sleeve cap belongs is a very satisfying activity for me. When it isn’t working, I can see what’s gone awry, rip it back, and try again. And it’s only a sleeve cap, after all — hardly more than a swatch’s worth of material — so my experiments rarely take longer than an hour to go off the rails.

Hopefully I have convinced you to give top-down set-in sleeves a try.

All Right, Let’s Do This

First things first: How big do you want your sleeves to be? The measurement you need is the circumference of the sleeve around your upper arm, right at the underarm.

You can arrive at this measurement in a many different ways. If you’ve made a sweater in the past that you adore, you can take the upper arm circumference from that sweater’s pattern, or you can measure the sweater itself. You can also measure a non-handmade sleeve that fits the way you want your new sweater to fit. Even if your reference garment isn’t quite what you want for this project, it’s fine to add or subtract from the reference measurement to guesstimate the right fit. Or you can get wild and measure your actual arm, adding or subtracting from that measurement to adjust for how you want this sleeve to fit.

Got your ideal sleeve circumference figured out? Good. Write it down.

The second number you need is your stitch gauge. You probably already know this from your swatch, or from knitting the front and back of the sweater. If you don’t know it, go ahead and measure how many stitches you get per inch (or per centimeter — so long as you measure everything in the same units, it doesn’t matter which units you use), making sure this gauge reflects the yarn and the stitch pattern you’re planning to use for your sleeves. Don’t be afraid of fractions. If you get 21 stitches over 4 inches, for example, your stitch gauge is 5 1/4 stitches per inch. Write it down.

Now you’re going to multiply your desired sleeve circumference by your stitch gauge. This gives you the total number of stitches around your sleeve cap.

Desired Sleeve Circumference x Stitch Gauge

=

Total Number of Stitches around Sleeve Cap

Example: Say that my arm measurement at the underarm is 16 inches, but I want my sleeves to be big and blousy, so I am aiming for a sleeve that measures 6 inches bigger than my arm, or 22 inches at the underarm. My stitch gauge is 5 stitches per inch.

22 inches x 5 stitches per inch = 110 stitches

Again, write this number down. It is the number of stitches that you will pick up across your entire sleeve opening — the distance from underarm to underarm in the picture on the left, below.

With me so far? Excellent.

Next, you’ll want to halve the total number of stitches around the sleeve cap. This number is how many stitches you will pick up between the underarm and the top of the shoulder. That’s the opening at the right side of the sweater photo on the right, above.

total number of stitches in sleeve cap / 2

=

number of stitches from underarm to top of shoulder

Example: 110 stitches / 2 = 55 stitches to pick up on each side of arm opening

Again, write the number down.

Then, subtract from this number however many stitches you bound off at one side of the underarm on the body. Later, you are going to pick those up stitch-for-stitch. For the moment, we need to figure out how many stitches are above the bind-off, between the bind-off and the top of the shoulder on one side.

number of stitches on each side - bind-off stitches

=

number of stitches to pick up above bind-off on each side

Example: 55 stitches - 5 stitches at underarm bind-off = 50 stitches per side after bind-off

Got that one written down, too? Excellent.

So the reason we’re doing all this math is that stitches are wider than they are tall. This means you usually can’t just pick up one stitch for every single row around the armhole. Unless you want to knit a sleeve that is considerably wider than your arm measurement, you’re likely going to need to skip some rows. The question is how many? Do you skip every other row? Every third, fifth, seventh row?

The way to finish figuring it out is to count the number of rows in half of your arm opening, starting one row above the bind-off at the armhole and extending to the top of the shoulder. If you look at the picture on the right above, it’s easy to do, because the armhole is striped. Beginning at the orange stripe right above the bind-off, there are exactly 50 rows to the top of the shoulder. This means in my example sweater I will pick up 50 stitches over 50 rows, which I just told you wasn’t likely to be something you want to do, but it will work for me because I do want to make a sleeve that is substantially larger than my arm measurement.

Here’s how that math works:

number of stitches to pick up above bind-off / number of rows above bind-off

=>

round and reduce fraction

=

pick-up ratio of stitches to rows

Example: 50 stitches per side / 50 rows per side =>

1 / 1

pick up 1 stitch for every 1 row

But let’s say I wanted to make a sleeve that is closer in size to my arm measurement, just an inch larger around. My math would look like this:

17 inches x 5 stitches per inch = 85 stitches

85 stitches / 2 = 42.5 stitches per side —> round up to 43 stitches per side of armhole

43 stitches - 5 stitches at underarm bind-off = 38 stitches per side of armhole

38 stitches per side / 50 rows per side —> reduces to 19 stitches / 25 rows —> which is about 20 stitches / 25 rows —> which is 4 stitches / 5 rows

I will pick up 4 stitches for every 5 rows.

check in

These are the numbers you should have so far:

1. Total Number of Stitches in Sleeve Cap

2. Number of Stitches on Each Side of Cap

3. Number of Stitches to Pick Up above Bind-Off

4. Pick-Up Ratio of Stitches to Rows

Time To Knit!

STEP ONE: GET ALL YOUR STITCHES ON THE NEEDLE

If you made the front and back separately, they need to be joined together at the shoulders before you can make your sleeve.

You will be knitting back and forth. I use a circular needle because I like the flexible cord for working in this small, rounded opening, but straight needles also work.

First, pick up the stitches at the underarm. You will pick up one stitch for every stitch you bound off of the body.

Second, using the ratio you worked out above, pick up the stitches between the first row above the underarm and the top of the shoulder.

To make the sleeve in my first example above, I would pick up 1 stitch for every row above the underarm, for a total of 55 stitches (5 at the underarm bind-off + 50 picked up over 50 rows). Or, in my second example of the more fitted sleeve, I would pick up 4 stitches and then skip a row, then 4 stitches again, until I got to the shoulder, where I would count to make sure I had 43 stitches (5 at the bind-off + 38 picked up over 50 rows) on my needle.

Count to make sure you have the right number of stitches. You can fudge a little bit if you’re only off by one or two, but if you’re off by a lot, you might need to try again on picking up stitches (or recheck your math).

Continue down the other side, but in reverse: pick up the second half of your stitches according to your ratio, and finish by picking up the bound-off stitches one-for-one.

You should now have the “magic” total number of stitches for the sleeve cap on your needle.

Huzzah!

Concept: What You’re About to Do

The next step will be to make the sleeve cap. Remember, a set-in sleeve cap is shaped like a bell that is flat across the top. You’ll begin with rows that have many fewer stitches than the rows at the bottom of the bell, and then you will gradually widen your cap as you approach the bound-off stitches at the underarm.

To get the desired bell shape from the top down, you begin by working across the top third of the sleeve opening, centered over the shoulder seam. Then you use short rows to gradually increase the width of the sleeve cap until you reach the underarm bind-off.

Here’s a general overview:

The illustration below shows those one-third, one-half, and two-thirds points on a sleeve cap worked from the top down with stripes. See how the first stripe in the cap makes the bell shape, and the stripes gradually flatten out on the way down?

Let’s Break It Down, Tho

STEP TWO: SETTING UP YOUR SHORT ROWS

First things first: since you worked across the sleeve opening to pick up your stitches, you’ll need to work back on the wrong side for your first row of the sleeve cap. But you’re not going to work all the way back to the beginning of the row. Instead, you’re going to work across the wrong side to a point that is two-thirds of the way across the row.

You know what that means? More math!

Take the total number of stitches around the sleeve cap and divide that number into thirds. Add two of those thirds together. Write that number down. This is the number of stitches you will work before you stop and turn around.

total number of stitches around sleeve cap / 3

x 2

= stitches to get to two-thirds point of sleeve cap

Example: 110 stitches around sleeve cap / 3 = 36.3 stitches x 2 = 72.6 stitches

Round to 73.

I’m going to purl 73 stitches on my first row.

On your way across, place a marker at the halfway point. For me, that’s at 55 stitches.

Now you’re going to turn the short row. There are a lot of different ways to turn short rows in order to avoid making gaps in your fabric. My go-to is the wrap-and-turn method, which puts a little lasso around the base of the next stitch on the left needle. To wrap and turn on a purl row, I take the yarn to the back, slip the next stitch purlwise onto my right needle, bring the yarn to the front again, and slip the stitch back onto the left needle.

After turning the short row, turn the work so the right side is facing you. You should now have two-thirds of the stitches of your armhole opening on your left needle and one-third on your right needle.

Remember, your goal is to work across one third of your stitches and for those stitches to be centered over the shoulder seam. So however many stitches you’ve got on your left needle lying to the right of the stitch marker — which marks the shoulder seam — that’s how many stitches you need to work past the marker before you turn another short row. Or, to say the same thing another way, work across the top of the armhole, slipping your stitch marker, until a third of the total armhole stitches remain on your left needle. Wrap and turn. To wrap from the right side, you bring the yarn to the front, slip the next stitch purlwise onto your right needle, take the yarn to the back again, and turn. Short rows unlocked!

Now that you have the first two short rows in place, begin increasing the width of every row by one stitch. To do this, work across the next wrong side row until you get to the stitch that you wrapped (the one with the lasso around the base of it). Remember, this was the stitch one past the end of your row, so when you work it, you will be adding one stitch to your current row.

You can choose to incorporate that little wrap-lasso, or you can ignore it. You’ll get a slightly different effect depending on which way you go, so feel free to experiment. If you decide to ignore it, purl the stitch with the wrap, wrap the next stitch the way you did before, and turn your work. If you want to incorporate the wrap, you need to pick up the little wrap with your needle from the front side of the work, drape it over the tip of the needle so it is next to the stitch itself, and purl them both at the same time — the stitch and its wrap — as though you’re working a purl two together. Then wrap the next stitch and turn your work.

Do the same thing on the right side, making your row one stitch wider by incorporating or ignoring the wrap and then wrapping the stitch after it. To incorporate the wrap from the right side, pick the wrap up on the right side with your right needle from underneath and drape it over the left needle beside the stitch. Knit them together through the back loop, then wrap the next stitch and turn.

Keep going on like this, incorporating or blowing right past your wraps, adding one stitch to each row, and watching your sleeve cap grow until you have added back all the stitches of the original row to your needles except for the underarm stitches that you picked up from the bound-off ends of the row.

On the next row, incorporate or blow past your wrap and then keep working even to the end of the row. Turn and work all the way across to the end of this row, removing your stitch marker and incorporating the final wrap.

You’ve made a sleeve cap!

STEP THREE: NOW MAKE A SLEEVE

From here, you can finish the sleeve however you’d like. You have lots of options, but I’m only going to dash through a few of them because I have to end this tutorial sometime, and this seems like as good a place as any.

First decision: Do you want to continue working your sleeve back and forth, or do you want to switch to knitting it in the round? To work in the round, you’ll need a circular needle smaller than your sleeve circumference, as well as a plan (double-pointed needles, two circulars, or a circular with a very short cord) for dealing with the reduced circumference as you get toward the cuff. I generally avoid the hassle by knitting my sleeves flat and seaming them at the end, but I like seaming.

Second decision: What kind of sleeve do you want to make?

The easiest sleeve is probably the one I made, a sort of semi-balloon sleeve. I just worked straight down without any decreases until I got to the cuff. Then I decreased rapidly by working knit two, knit two together across one entire row. I did that on three sets of right-side rows until I had the number of stitches I wanted for the cuff. Then I worked the cuff in ribbing and bound it off.

If you don’t want a ribbed cuff, you can do a hem. Work to the end of your sleeve, purl a row, and then go down a couple needle sizes and work about two inches in stockinette. Your purl row becomes a turning row, and you sew the hem down on the inside.

If you want a more traditionally fitted sleeve, you’ll need to decrease the width as you work your way toward the cuff. As a general rule of thumb, work about two inches before you start decreasing. Then take off one stitch from each side about every inch until you reach the cuff. Sometimes you’ll have to decrease more rapidly as you get closer to the cuff — work a decrease row every half inch instead of every inch or decrease four stitches on every decrease row instead of two.

There are ways to apply more precise math to this situation, but you can also drape your sleeve-in-progress over your arm and eyeball it, and I find that works just as well or better a lot of the time.

Refinements

This approach to a top-down sleeve won’t work in all yarns, for all projects, or for all people. Here are a few refinements to think about if your cap is going off the rails.

• Reconsider the ratios. The division of the armhole opening into thirds might not work for your project. If you want a flatter, smoother top to your sleeve cap, try adding more stitches to that central portion. Or, if the top of the cap feels wide or bulky, try reducing the number of stitches you work there.

• Adjust the rate of increase. On a sleeve cap worked from the bottom up, it’s customary to decrease one stitch at each edge of every right side row until about three and a half inches of width remain, then decrease twice as rapidly over the next two inches. On a sleeve that has more ease and is knit using this method, the rapid decreases might begin lower down, when four and a half or even five and a half inches remain. The top-down method of this tutorial doesn’t include any rapid increases in the first few inches. You may find that it works for you to add two stitches per right side row for the first few inches before slowing down the rate of increase for the remainder of the sleeve cap.

• Stop increasing sooner. How many short rows you work determines the depth of the sleeve cap. You might like the look of your cap better if, instead of working short rows all the way up to the bound-off underarm rows, you stop the short rows an inch or two sooner, or incorporate those last sets of stitches before the underarm at a more rapid pace.

Final Thoughts

I hope this tutorial has given you the confidence to try a set in sleeve from the top down! The stakes are low. No one will die if you get it wrong. Have at it, and let me know how it works out!

If there are things you still want to know about top-down sleeves — or anything else — you can also submit questions through the Fit Desk form, and I’ll try to answer them in a future post.