Pattern adjustments; or, the devil is in the details

I’ve been thinking a lot lately about a pullover pattern by Sari Nordlund, called Onnen:

© Sari Nordlund

It checks a number of boxes for me at the moment – easy to knit, worsted-weight yarn, raglan, interesting but not too complicated cables, vertical pattern, comfortable, comforting, wearable. After looking through literally thousands of patterns over a period of many weeks, I found that I kept coming back to this one. I even have the yarn at hand to knit this with.

So, I bought the pattern and spent some time looking it over, and immediately hit a conceptual snag. The gauge is listed as follows (this is available right on the pattern page on Ravelry, so I’m not giving anything away):

Gauge:
19 sts and 25 rounds = 10 cm over stockinette stitch on 4.5 mm / US 7 needles, in a round, after blocking.
25 sts and 27 rounds = 10 cm over cabled pattern on 4.5 mm / US 7 needles, in a round, after blocking. Measurement taken approximately over round 19.

So far so good. Looking at the pattern, however, I became confused. The pullover is knitted bottom-up in-the-round. For the size L (47″), it asks that you cast on 244 stitches. Markers are added to divide the front and the back, which is divided evenly so that both back and front have 122 stitches. However, the back is knitted entirely in stockinette stitch which has a gauge of 19 stitches to 4in/10cm, and the front is composed of a central block of 68 stitches in cable pattern knitted at a gauge of 25 stitches to 4in/10cm, plus 54 stitches (27 either side) knitted in stockinette at a 19 stitch gauge. I have to admit that this sets up warning bells for me.

I did some math. It turns out that if you are hitting the recommended gauges, for the size L, the back will measure 25.68″ and the front will measure 22.24″. (It goes like this: 19 stitches per 10cm/4″ equals 1.9 stitches/cm or 4.75 stitches/inch. 25 stitches per 10cm/4″ equals 2.5 stitches/cm or 6.25 stitches/inch. To continue in inches, the back will thus be 122 stitches at 4.75 per inch, or 25.68″. The front will be 54 stitches at 4.75 stitches/inch (11.36″) PLUS 68 stitches at 6.25 stitches/inch (10.88″) which equals 22.24″. Yikes! It’s worse than I imagined.

(At this point, I asked Doug to read this, and I did his head in. “What’s the problem?”, he asked. “The circumference works out right and its knitted in the round.” The problem is with the placement of the markers, and thus the side seams, and thus the placement of the sleeves and raglan shapings. If the side seams are placed so that both back and front have an equal number of stitches, then the sleeves will be in the wrong place, and the front of the sweater will be bunched over the chest, while the back of the sweater will be too wide.)

Why would you knit a sweater and place the raglan sleeve shapings unevenly skewed towards the front, so that the front measures significantly less than the back? By this time I thought maybe I was over-thinking things and I wrote to the designer (on Ravelry, which perhaps she doesn’t monitor) and I never heard back.

I thought about forgetting the pattern altogether, and spent a few more weeks searching around for something else to knit, but in fact, I like this pullover. (I have quite a few of Nordlund’s newer patterns on my favourites list.) I do think the pattern photos look just a bit off around the raglan shapings, but of course that makes sense if they are fundamentally in the wrong place. So, being a bit OCD, I checked every single photo of this sweater which exists on Ravelry. There are currently 98 of them. Of those 98, only 2 photos show the back of the sweater being worn on an actual person (there are a number of photos of the back of the sweater, while it is hanging on a hanger). Of those two, one of them is obscured by a bunch of long hair and a slightly angled shot, and the other looks, frankly, much too wide across the back.

I then spent a long time re-calculating the widths of back and front as I moved the markers towards the back one stitch at a time, and discovered that if I moved the markers back 4 stitches on each side, so that the overall stitch count is the same, but the back has 16 fewer stitches than the front, then the finished pullover measurements should be roughly equal for the front and back.

I won’t go into the details too much here, but I also freaked out about the fact that the ribbing was even across the bottom of the sweater, even though the central cabled portion of the sweater had a much tighter gauge. For those of you interested in the nitty gritty details, I ended up casting on 240 stitches for the ribbing, and then increasing 4 stitches evenly across the cabled block of stitches at the front, while at the same time, placing the markers so that there are 114 stitches on the back and 130 stitches on the front.

Am I being unduly obsessive? Will this work out in the end? Only time will tell.

21 thoughts on “Pattern adjustments; or, the devil is in the details

  1. I would be at least as obsessive. In fact, I see no point in designing such a sweater to be knit in the round in the first place. Much more straight forward to design it to be knit it flat in four pieces (back, front, two sleeves) and then – assuming it has schematics, and I dismiss any designer who doesn’t provide these for sweaters – you could analyze the fit and adjust it as needed before you pick up the needles. I love knitting in the round when it comes to stranded-colorwork projects, but not much else.

    Crossing my fingers for you!

  2. I think you’re right. It’s disappointing that the designer didn’t respond. That’s a pretty big gauge issue, and a reasonable solution.

  3. I know the cabling would cause the front to be smaller horizontally, but that doesn’t seem to excuse making the back wider.

    • I think that the back being wider is just a consequence of placing the sleeves in the wrong place – according to the number of stitches rather than the actual knitted measurements. Hopefully, it will all work out!

  4. It sounds to me like you got it right. Shouldn’t there have been increases above the ribbing to accommodate the cable crossings? It is a lovely looking sweater. Good luck.

  5. Gorgeous sweater. Good for you to work out all the math! Can’t wait to see the finished product – front and back

  6. I agree with your reasoning – cables draw the fabric in quite a bit, so equivalent stitches for back and front would definitely turn out wonky. Looking forward to your version!

  7. I recognise your panic for sure, who wants to embark on a project that doesn’t make sense. I have in the past written to people on Revelry who were recent finishers of the same pattern and asked them directly about it. All but one person came back to me and those that did all seem to have made the same modification themselves and had also seen the same flaw in the pattern. Part of the reason I now keep knitting Heidi Kirrmaier’s designs is each of her patterns has a query thread and you can ask dumb questions and get lovely replies. To be that’s worth it, plus I love her designs. So I would try emailing a recent English speaking finisher and ask them.

    • I wouldn’t describe it as panic, more like being finicky. I often write to people who have knitted projects on Ravelry and ask them questions. I didn’t do it in this case, but I agree with you that people are generally very nice and generous with their time in responding.

  8. Facing an issue such as this is challenging and so time consuming. My first thought would be to run from this pattern, so I really admire your tenacity! This designer has so many beautiful designs on Ravelry and it is unfortunate that you didn’t hear back from her as to her thoughts and feedback behind this pattern design. You’ve already spent so much time analyzing the pattern and I know that your changes will make an amazing garment. Look forward to hearing more!

    • I agree that she has many beautiful designs coming out, and there are quite a few I can imagine casting on. I have a feeling this design is an early one. I imagine she is using other channels for communicating, perhaps Instagram which I don’t follow. In any case, I can report that so far, it looks great.

  9. I typically go even further — even if a sweater has the same allover pattern or stockinette on both front and back (ie, no patterns with different gauges), I still modify the pattern to place the sleeves about an inch further back, to ensure the front will be larger than the back. On a design like this one, where the cabled gauge will compress the front, that modification is absolutely necessary!

    It’s poor design not to account for the different gauge of the cables, whether in the distribution of front and back stitches, or not reducing the number of ribbing stitches just below the cable.

    • This is really interesting; I am glad to hear that you’ve had success with this, because it makes a lot of sense to me. I’ve done something similar once with a seamed garment, knitting a smaller size for the back than for the front. Thank you for commenting.

  10. Your logic seems quite reasonable. Thanks for sharing. I’ll have this in the back of my mind if I ever knit a cabled sweater again. I’ve only done one and I just followed the pattern. Didn’t check to see if the math worked. I don’t recall it even gave a stockinette as well as a cable gauge.

Leave a Reply