purlJam

<planning>
1. **Design Strategy**:
   - Size: Small only (34-36" bust), with 2-4" positive ease → target 38" finished bust
   - Yarn: Sport weight
   - Style: Top-down seamless raglan pullover
   - Key feature: Basketweave stitch pattern on the front body panel only; back and sleeves in stockinette
   - Neckline: Crew neck (default per template)
   - UK English, metric first, tension not gauge
   - Difficulty: Intermediate

2. **Construction Plan**:
   - Top-down seamless raglan
   - Cast on neckline stitches, work ribbed neckband
   - Place 4 raglan markers, increase each side of each marker every other round
   - Yoke in stockinette (basketweave starts on body front panel only after sleeve separation)
   - Separate sleeves at underarm, cast on underarm stitches
   - Body worked in the round: back in stockinette, front in basketweave
   - Sleeves worked in the round in stockinette with decreases
   - Ribbed hem and cuffs

3. **Quality Assurance**:
   - Need to carefully define the basketweave panel on the front so it begins cleanly after sleeve separation
   - Basketweave requires a multiple of 8 stitches + 4 for symmetry (standard basketweave: k4, p4 repeat). I'll check that front stitch count works with basketweave multiple.
   - Raglan increases: need even stitch distribution at cast-on and increase calculations
   - Short rows for back neck: include as optional
   - Need to verify: yoke depth, armhole depth, body length, sleeve length

4. **Final Calculations**:

**Sport weight standard tension:**
- 26 sts × 34 rows = 10 cm (4") in stockinette on 3.5mm needles
- Or more commonly for sport: 24 sts × 32 rows per 10cm on 3.75mm
- Let me use: **26 sts / 10 cm = 2.6 sts per cm** on 3.5mm needles, 34 rows / 10 cm = 3.4 rows per cm

Actually, let me use a clean standard for sport weight:
- **Tension: 24 sts × 32 rows = 10 cm (4") in stockinette on 3.75mm (US 5) needles**
- That gives 2.4 sts/cm, 3.2 rows/cm

**Size Small measurements:**
- Finished bust: 38" = 96.5 cm (let's use 96 cm for clean numbers)
- Body length (underarm to hem): 38 cm (15")
- Sleeve length (underarm to cuff): 44 cm (17.5")
- Yoke depth (neckline to underarm): ~20 cm (8")
- Neck circumference: ~36 cm (14")

**Stitch counts:**
- Bust: 96 cm × 2.4 sts/cm = 230.4 → 232 sts total (must be divisible by appropriate number)
- For the body in the round: 232 sts total = 116 front + 116 back
- Underarm cast-on: 6 sts each side (2 × 6 = 12 sts total added at each underarm)

**Basketweave on front panel:**
- Front = 116 sts
- Basketweave standard: k4, p4 repeat = 8 sts repeat
- 116 ÷ 8 = 14.5 — not exact. Need to adjust.
- Options: 112 sts front (14 repeats × 8 = 112) or add a border
- Let me reconsider bust: 
  - If total body = 232, front = 116 — not divisible by 8
  - Try 240 sts total = 120 front + 120 back. 120 ÷ 8 = 15 repeats ✓
  - 240 sts ÷ 2.4 sts/cm = 100 cm = 39.4" finished bust — perfect for small with positive ease
  - Let's use **240 sts total for body**, giving 100 cm / ~39.5" finished bust

**Working backwards from body to yoke:**
- Body after separation: 240 sts (including underarm cast-on)
- Underarm cast-on: 6 sts each side × 2 = 12 sts added
- So stitches on needles before separation (body only): 240 - 12 = 228 body sts... 

Wait, let me think about this more carefully.

At separation point:
- Total sts on needle = front body + back body + 2 sleeves (+ 4 raglan sts or incorporated)
- After separation: body gets front + back + underarm cast-ons; sleeves get sleeve sts + underarm cast-ons

Let me define:
- Body sts at separation (front + back, not counting underarm): 228 sts? That would be 114 front + 114 back
- With 6 sts cast on at each underarm: body becomes 114 + 6 + 114 + 6 = 240 sts ✓

But 114 front ÷ 8 = 14.25 — not divisible by 8.

Let me try different underarm cast-on:
- Body sts at separation: 120 front + 120 back = 240 sts
- No underarm cast-on, or very small: body = 240 sts
- That means total before separation = 240 body + sleeve sts (no underarm CO)
- Actually typical small: let's try 8 sts underarm CO
- Body at separation: 116 front + 116 back = 232; + 8+8 = 248 sts — 124 front, not /8

Hmm. Let me approach differently.

I want front body = multiple of 8.
Options: 112, 120, 128

With 120 front + 120 back = 240 body sts after adding underarm:
If underarm CO = 6 each side: body sts before separation = 240 - 12 = 228 → 114 each side (not /8)
If underarm CO = 4 each side: body sts before separation = 240 - 8 = 232 → 116 each side (not /8)
If underarm CO = 8 each side: body sts before separation = 240 - 16 = 224 → 112 each side (/8 ✓, but 112 ≠ 120)

Hmm — the front sts at separation ≠ front sts in body because underarm sts are added. The basketweave panel is on the BODY front, which INCLUDES the underarm cast-on sts incorporated into the round.

Actually after separation, the body is worked as:
- [back sts] [underarm CO] [front sts] [underarm CO] — joined in the round

The basketweave is on the FRONT section. The underarm CO sts on either side of the front will be at the SIDES (near the raglan/underarm position), so they're not really part of the centre front basketweave. 

Let me simplify: I'll define the basketweave as working on the FRONT stitches only (not the underarm CO sts, which blend into the side "seam" area). So the basketweave panel = front sts at separation point.

Target front sts at separation = multiple of 8.

Let me try: front = 112 sts at separation, back = 112 sts = 224 body sts before underarm
Underarm CO = 8 sts each side
Total body after joining = 224 + 16 = 240 sts ✓ (100 cm)

112 ÷ 8 = 14 repeats ✓

Now for sleeves:
- Total yoke sts at cast-on + increases = body + sleeves + raglan lines
- Let me figure out raglan distribution

**Raglan yoke calculation:**

At cast-on:
- Neck circumference small = ~36 cm × 2.4 sts/cm = 86.4 → 88 sts

Distribution at cast-on (4 raglan markers, creating: back, sleeve, front, sleeve):
Standard small crew neck distribution:
- Front neck: typically more sts than back neck initially? No, for crew neck usually:
- Back: more sts than front at start
- Typical split: back = 30%, each sleeve = 15%, front = 30%, + 4 raglan sts (1 each)

With 88 sts:
- 4 raglan sts (1 each) = 4 sts
- Remaining: 84 sts for sections
- Back: 28 sts, Front: 28 sts, each sleeve: 14 sts
- Total: 28 + 1 + 14 + 1 + 28 + 1 + 14 + 1 = 88 sts ✓

After yoke, at separation:
- Front should be 112 sts, Back should be 112 sts
- Front increases needed: 112 - 28 = 84 sts added to front (42 increases, meaning 42 increase rounds... wait)

Actually in raglan, each increase round adds 1 st to each side of each section:
- Each section gains 2 sts per increase round (1 M1L + 1 M1R at its boundaries... wait)

No. Standard raglan: there are 4 raglan lines. At each line, you do M1L before marker and M1R after marker. So:
- Per increase round: 8 total increases (2 per raglan line × 4 lines)
- Distribution: each body section (front, back) gains 2 sts per increase round
- Each sleeve gains 2 sts per increase round

So after n increase rounds:
- Front: 28 + 2n
- Back: 28 + 2n  
- Each sleeve: 14 + 2n

Target front at separation: 112 sts
28 + 2n = 112
2n = 84
n = 42 increase rounds

At that point:
- Back: 28 + 84 = 112 sts ✓
- Each sleeve: 14 + 84 = 98 sts

**Yoke depth check:**
- Increase rounds happen every other round (every 2 rounds)
- 42 increase rounds × 2 = 84 total rounds
- 84 rounds ÷ 3.2 rows/cm = 26.25 cm

That's quite deep for a size small. Standard yoke depth for small is about 18-22 cm. Let me reconsider.

Hmm. 26 cm yoke depth might be slightly much. Let me check: small armhole depth is typically about 18-20 cm. Let me see if I need to recalculate.

Maybe I should start with more sts at the neck, or reconsider finished bust.

Alternative: Use a smaller finished bust.
Let's try 92 cm finished bust (36.5") — still 2" positive ease for a 34-36" bust.
92 cm × 2.4 = 220.8 → let's say 224 sts total body (56 cm each half... wait 112 each)
With underarm CO of 8 each: body sts before separation = 224 - 16 = 208 → 104 front + 104 back
104 ÷ 8 = 13 repeats ✓

Front increases: 104 - 28 = 76 → n = 38 increase rounds
38 × 2 = 76 total rounds ÷ 3.2 = 23.75 cm — still a bit much.

Hmm. Let me reconsider my row gauge or my neck cast-on.

Actually, for sport weight the yoke depth of ~22-24 cm is reasonable for size small. Let me verify:
- Armhole depth for size small = typically 19-22 cm
- 22 cm is within range.

Let me use:
- 96 cm finished bust (close to 38" = 96.5 cm)
- 96 × 2.4 = 230.4 → 232 sts total
- With underarm CO = 8 each side: body before = 232 - 16 = 216 → 108 front + 108 back
- 108 ÷ 8 = 13.5 — not exact ✗

Try 240 sts total body (100 cm):
- Underarm CO = 8 each: body before = 240 - 16 = 224 → 112 front + 112 back ✓ (14 repeats of 8)

OK so let's go with 240 sts total body, 100 cm (39.5") finished bust.
For size small (34-36"), 100 cm is 3.5" positive ease on a 36" bust — that's reasonable, maybe slightly generous but fine.

n = (112 - 28)/2 = 42 increase rounds
42 × 2 = 84 rounds = 26.25 cm yoke depth

To reduce yoke depth, I can increase the neck cast-on:
If I start with more sts:

Let's try neck CO = 96 sts:
- 4 raglan sts = 4
- Remaining 92 sts
- Back: 32, Front: 32, each sleeve: 14
- Check: 32+1+14+1+32+1+14+1 = 96 ✓

Target front = 112:
112 - 32 = 80 → n = 40 increase rounds
Back: 32 + 80 = 112 ✓
Each sleeve: 14 + 80 = 94 sts

40 × 2 = 80 rounds = 80 ÷ 3.2 = 25 cm — still a lot.

Hmm. Let me reconsider the starting sleeve stitches. Maybe sleeves should start larger to reduce n.

Actually, for sport weight, 25 cm yoke depth might be fine for a well-fitted sweater — it accounts for the neck ribbing too. But let me also reconsider row gauge.

For sport weight on 3.75mm:
Row gauge is typically 28-30 rows/10cm in stockinette (not 32).
Let me use 28 rows/10cm = 2.8 rows/cm.

80 rounds ÷ 2.8 = 28.6 cm — even more. That's too deep.

Let me try a completely different approach and look at this from the target yoke depth:

Target yoke depth: 20 cm
At 2.8 rows/cm: 20 × 2.8 = 56 rounds
With every-other-round increases: 56 ÷ 2 = 28 increase rounds

With 28 increase rounds:
- Need to grow from neck to body: +56 sts for front, +56 for back, +56 for each sleeve
- Front at neck: ? + 56 = 112 → front at neck = 56 sts
- Back at neck: 56 sts
- Each sleeve at neck: depends on starting formula

That means the neck would need 56+56 = 112 sts for just body sections. That's way too wide for a crew neck.

The issue is that raglan yokes naturally require many increase rounds because you're building a lot of fabric. 20+ cm is normal and correct for a sport weight yoke.

Let me look at published patterns. A typical sport weight top-down raglan for size small:
- Neck CO ~80-96 sts
- Yoke works for ~22-26 cm
- This IS the norm for a sport weight size small

So let me commit to:
- 88 st neck CO
- 42 increase rounds
- Yoke depth: 42 × 2 = 84 rounds

84 rounds ÷ 2.8 rows/cm = 30 cm — hmm, that seems like a lot.

Actually wait. Let me reconsider my tension numbers. Let me use:
**Sport weight tension: 26 sts × 36 rows = 10 cm on 3.5mm needles**

That's 2.6 sts/cm and 3.6 rows/cm.

Body sts: Let me restart with cleaner tension.

**TENSION: 26 sts and 36 rows = 10 cm (4") in stockinette stitch on 3.5mm needles**
- 2.6 sts/cm; 3.6 rows/cm

**Small finished bust: 96 cm (38")**
- 96 × 2.6 = 249.6 → 248 sts (round to nearest reasonable number)
- 248 ÷ 2 = 124 front + 124 back
- 124 ÷ 8 = 15.5 — not exact

Try 240 sts: 96 cm... wait 240 ÷ 2.6 = 92.3 cm (36.3") — that's only 0.3" ease on a 36" bust. Too little.

Try 248 sts = 248/2.6 = 95.4 cm (37.5") — fine for 34-36" with ~1.5-3" ease.
But 124 front ÷ 8 = 15.5 — not exact.

Try 256 sts = 256/2.6 = 98.5 cm (38.7") — nice with 2.7-4.7" ease
256 ÷ 2 = 128 each side
128 ÷ 8 = 16 repeats ✓ Perfect!

But underarm CO: if I add 6 sts each side = 12 total:
Body before separation = 256 - 12 = 244 → 122 each side — not /8

If I add 4 sts each side = 8 total:
Body before = 256 - 8 = 248 → 124 each side — not /8

If I add 8 sts each side = 16 total:
Body before = 256 - 16 = 240 → 120 each side — 120 ÷ 8 = 15 ✓

So: front at separation = 120 sts, with 8 sts CO at each underarm = 120 + 8 = 128 front after underarm ✓

Wait, actually the basketweave pattern is worked on the BODY in the round. Once the body is joined, the "front" section in the round = front sts + the underarm CO sts? No — the underarm CO sts are typically at the JOIN points (underarm positions), which are on either SIDE, not counted as front or back. They bridge front and back.

In a top-down raglan, when you separate sleeves:
- You have: [back][raglan st][sleeve][raglan st][front][raglan st][sleeve][raglan st]
- You place sleeve sts on holders
- Cast on new sts over the underarm gap
- Join body: [back] + underarm CO + [front] + underarm CO = body in the round

For the body round, starting at a specific point (say, back left underarm):
- 8 underarm CO sts (these sit at left underarm "seam")
- 120 back sts
- 8 underarm CO sts (right underarm "seam")  
- 120 front sts
- back to start

The front 120 sts = the dedicated basketweave panel. The underarm CO sts (8 on each side) are at the sides/seams.

Let me set up the pattern to treat the 120 front sts as the basketweave panel, with the 8-st underarm CO sts on each side being worked in stockinette (or reverse stockinette as a "seam" stitch, but let's keep it simple — stockinette throughout except front panel basketweave).

120 ÷ 8 = 15 repeats ✓

**REVISED PLAN:**
- Tension: 26 sts × 36 rows = 10 cm on 3.5mm needles
- Finished bust: 256 sts ÷ 2.6 = 98.5 cm (~38.7" / ~39")
- Actually let me recalculate: 256/2.6 = 98.46 cm = 38.8" ✓ (2.8-4.8" ease for 34-36" bust — perfect)

**Neck cast-on:**
Neck circumference for small: ~35-38 cm. Let's use 36 cm.
36 × 2.6 = 93.6 → let me try 88 sts or 92 sts.

Let me try 84 sts neck CO:
Distribution: 84 sts
- 4 raglan sts (1 each) + 80 remaining
- Back: 28, Front: 28, each sleeve: 12
- 28+1+12+1+28+1+12+1 = 84 ✓

Target front at separation: 120 sts
120 - 28 = 92 increases on front → n = 46 increase rounds

That's quite a lot. Yoke depth = 46 × 2 = 92 rounds / 3.6 rows per cm = 25.6 cm

Hmm. For sport weight size small, 25 cm yoke depth is common in published patterns. Let me just accept it and see.

Actually — let me look at this differently. A typical small women's yoke depth is about 20-22 cm. With some back neck short rows and neckband, the actual knitting depth might be about 18-20 cm of increase rounds.

Let me try starting with more sts to reduce the number of increase rounds needed.

Try 96 sts neck CO:
- 4 raglan sts + 92 remaining
- Back: 32, Front: 32, each sleeve: 14
- 32+1+14+1+32+1+14+1 = 96 ✓

Target front at separation: 120 sts
120 - 32 = 88 increases on front → n = 44 rounds

Back at separation: 32 + 88 = 120 ✓
Each sleeve at separation: 14 + 88 = 102 sts

Yoke depth: 44 × 2 = 88 rounds / 3.6 = 24.4 cm — still ~24 cm.

Let me try increasing sleeve start sts to reduce n:
96 sts with: Back 30, Front 30, each sleeve 16
30+1+16+1+30+1+16+1 = 96 ✓

Front target: 120; n = (120-30)/2 = 45
Back: 30+90 = 120 ✓
Sleeves: 16+90 = 106

Yoke: 45×2 = 90 rounds / 3.6 = 25 cm

Not much difference. The fundamental issue is that sport weight at 2.6 sts/cm means you need many sts, and the yoke has to be deep enough. 24-25 cm is actually realistic and correct for sport weight.

Let me verify with a real-world example: 
- A typical "worsted weight raglan small" pattern: 20 sts/10 cm, bust = 38" = 96.5 cm, sts = ~190 sts
- Neck CO ~76 sts, 38 increase rounds, row gauge ~26 rows/10 cm
- 38 × 2 = 76 rounds / 2.6 = 29 cm — wait that's even more!

Hmm. Actually in many patterns, the yoke depth stated is around 20-23 cm, and they increase every other round for only part of the yoke, then the remaining rounds are worked even.

Oh wait — I think I've been misunderstanding. In a raglan, you DON'T necessarily increase all the way to the final count. You increase until the armhole depth is correct, and the relationship between increases and yoke depth is a natural constraint that works out.

Let me reconsider. The raglan yoke depth (armhole depth) for size small is approximately 18-20 cm. 

With 3.6 rows/cm:
18 cm × 3.6 = 64.8 → 64 rounds
Working every other round: 32 increase rounds

With 32 increase rounds:
- Front: start sts + 64 more
- Back: start sts + 64 more  
- Each sleeve: start sts + 64 more

If I target front = 120 sts at separation:
Starting front sts = 120 - 64 = 56 sts

That means neck cast-on would include 56 front + 56 back + sleeve sts + 4 raglan sts.
56+56 = 112 for body sections. That's a very wide crew neck.

With sleeves at maybe 20 sts each: total = 56+1+20+1+56+1+20+1 = 156 sts neck CO
That would give a neck circumference of 156/2.6 = 60 cm — way too wide for a crew neck.

OK. I think I need to reconsider my approach. The fundamental constraint in a raglan is:
- The number of increase rounds is fixed by the armhole depth
- The starting stitch count is determined by working backwards from the finished width MINUS the increases gained

CORRECT FORMULA:
Starting back sts = target back sts - (2 × number of increase rounds)
Starting front sts = target front sts - (2 × number of increase rounds)  
Starting sleeve sts = target sleeve sts - (2 × number of increase rounds)

Target sleeve sts at underarm: for size small sport weight, upper arm circumference ~28-30 cm
Upper arm: 30 cm × 2.6 = 78 sts

With 32 increase rounds:
Starting sleeve sts = 78 - 64 = 14 sts ← reasonable for a crew neck sleeve start

Back: 120 - 64 = 56 sts ← that's too wide
Front: 120 - 64 = 56 sts

Total neck CO = 56+1+14+1+56+1+14+1 = 144 sts = 144/2.6 = 55.4 cm neck 

That's way too wide. A crew neck is ~35-38 cm.

The reality is that the number of increases and the armhole depth must be balanced with a reasonable neck circumference. The standard solution in published patterns is:

You CANNOT achieve both a natural neck size AND a standard armhole depth WITH purely raglan increases. So most patterns do a hybrid approach:
1. Work more aggressive increases at the start (more increase rounds than non-increase rounds)
2. OR add extra stitches for the neck and accept a wider neck
3. OR start with more sleeve and front/back sts and do short rows for neck shaping
4. OR the yoke depth IS actually 24-26 cm for sport weight

Actually, I think the answer is simply that for sport weight, the yoke depth IS around 22-26 cm, which is realistic. Let me look at it from a human anatomy perspective:

The yoke of a sweater covers from the base of the neck to the underarm. For a size small adult:
- Neck to shoulder: ~10 cm
- Shoulder to underarm (armhole depth): ~18-20 cm
- Total distance = ~28-30 cm along the diagonal

But the YOKE of a top-down raglan goes from neck to underarm measured VERTICALLY, which is approximately the armhole depth = 18-22 cm.

I think the confusion is that in a raglan, the row gauge and stitch gauge together naturally produce the right proportions IF your tension is correct. The yoke depth of 24-25 cm in rounds is the vertical depth, which is correct for a sport weight size small.

Let me just go with:
- 44 increase rounds (every other round, so 88 total yoke rounds)
- 88 rounds / 3.6 rows/cm = 24.4 cm yoke depth
- This gives the armhole depth needed for a proper fit

Actually I'm now second-guessing my row gauge. Let me use standard sport weight values from published sources:

Sport weight (also called 5-ply in UK):
- Typically 23-26 sts / 4" (US), 22-26 sts / 10 cm
- Row gauge approximately 30-34 rows / 10 cm

I'll use: **26 sts / 10 cm, 34 rows / 10 cm** on 3.5mm needles.
So 2.6 sts/cm, 3.4 rows/cm.

With 44 increase rounds: 88 rounds / 3.4 = 25.9 cm — still about 26 cm.

Actually, I want to double-check that this is correct and not a math error. Let me look at it from the armhole depth requirement:

Size small armhole depth = 19 cm (standard for 34-36" bust)
19 cm × 3.4 rows/cm = 64.6 → 64 rounds
Increase rounds = 64/2 = 32

Back: 120 - 64 = 56 ← too wide for crew neck

Hmm. So there's a fundamental incompatibility.

The solution used in virtually ALL top-down raglan patterns is:
**The neck cast-on is small (reasonable crew neck), the yoke is deep, and some of the yoke rounds are worked even (no increases) interspersed, or the pattern accepts a deeper yoke.**

OR more commonly: the yoke depth ends up being ~22-25 cm in sport weight, which is correct. The "armhole depth" measurement chart for size small sweaters is ~19-20 cm measured from underarm, but the TOP-DOWN raglan yoke includes the shoulder region too (which adds ~5-8 cm), making the full yoke depth 24-28 cm. This is correct and standard.

So the yoke includes:
- Shoulder shaping region: ~5-6 cm
- Armhole depth: ~18-20 cm
- Total yoke: ~23-26 cm ← this is what we calculate

OK! So 44 increase rounds giving a 25.9 cm yoke depth is actually CORRECT and normal. The armhole depth would be measured on the final garment differently from the total yoke depth. ✓

Let me now commit to final numbers:

**TENSION: 26 sts × 34 rows = 10 cm (4") on 3.5mm needles**
(2.6 sts/cm, 3.4 rows/cm)

**SIZE SMALL:**

Finished measurements:
- Bust: ~98.5 cm (38.7") = 256 sts
- Body length underarm to hem: 37 cm (14.5")
- Yoke depth: ~25 cm (9.75")
- Upper arm: ~30 cm (12") = 78 sts
- Sleeve length underarm to cuff: 44 cm (17.5")

**NECK CAST-ON: 96 sts**
Distribution: Back 32, Sleeve 14, Front 32, Sleeve 14, + 4 raglan sts (1 each)
32+1+14+1+32+1+14+1 = 96 ✓
Neck circumference: 96/2.6 = 36.9 cm (~14.5") — reasonable for a crew neck ✓

(Note: the neckband adds about 2-3 cm, making the actual neck opening ~39-40 cm, which is fine for a crew neck)

**YOKE INCREASES:**
Target at separation:
- Front: 120 sts (120/2.6 = 46.2 cm... wait let me recalculate final body)

Hmm, I had front separation sts = 120, but I need to re-examine:

After 44 increase rounds:
- Front: 32 + (44×2) = 32 + 88 = 120 sts
- Back: 32 + 88 = 120 sts
- Each sleeve: 14 + 88 = 102 sts

At separation, add 8 sts underarm CO each side:
- Body total: 120 (front) + 8 + 120 (back) + 8 = 256 sts ✓
- 256 / 2.6 = 98.5 cm ✓

Front panel (basketweave): 120 sts ÷ 8 = 15 repeats ✓

**Sleeve sts at separation: 102 sts**
Upper arm circumference: (102 + 8 underarm CO) = 110 sts / 2.6 = 42.3 cm

Wait, that's too big. 42 cm upper arm circumference is very large.

The upper arm circumference should be around 30-34 cm for size small with ease.
30-34 cm × 2.6 = 78-88 sts

But my sleeve has 102 sts at separation + the underarm CO sts. Hmm.

Let me reconsider. The sleeve stitches at separation are just the sleeve sts from the yoke. The underarm CO sts are SHARED — they don't all go to the sleeve. Actually:

When separating, the underarm CO sts are cast on to BRIDGE the gap. These sts belong to BOTH the body AND the sleeves... no, actually typically in a top-down raglan:
- The sleeve sts go on holders
- New underarm sts are cast on
- These new sts become part of the BODY (not the sleeve)

For the sleeves, when you pick them back up and work down:
- You pick up the held sleeve sts
- You also pick up sts from the underarm CO (usually all of them, or half)

Standard approach: The underarm CO sts are cast on to close the body. When working sleeves, you pick up the sleeve sts + the underarm CO sts + a few from the body edge.

Let me use a different approach for clarity:

At separation:
- Each sleeve: 102 sts (on holder)
- Cast on 8 new sts at each underarm (for body)
- Body total: 120 + 8 + 120 + 8 = 256 sts ✓

For sleeves, when picking up:
- Pick up 102 held sts + 8 sts from underarm CO + pickup 0 extra = 110 sts per sleeve
- 110 / 2.6 = 42.3 cm ← Too wide!

For size small, sleeve cap/upper arm should be about 32-34 cm including ease.
32 cm × 2.6 = 83.2 → ~84 sts

So I need the sleeve start (after pickup) to be about 84 sts.
If underarm CO = 8 sts: sleeve pickup = 84 - 8 = 76 sts at separation
76 is much less than 102.

The problem is that 44 increase rounds gives TOO MANY sleeve sts.

This is the classic raglan geometry issue. Let me reconsider.

The fix: use FEWER starting sleeve sts and/or fewer increase rounds, or reduce the target body size.

Actually, let me work backwards from both constraints simultaneously:

Let n = number of increase rounds
Front/back starting sts = F
Sleeve starting sts = S

After n rounds:
- Front: F + 2n (target: allows us to determine F+2n)
- Sleeve: S + 2n (target: ~76 sts at separation for 84 sleeve after 8 underarm CO)

Sleeve target at separation: 76 sts
S + 2n = 76

Neck CO: 2F + 2S + 4 = neck sts
For crew neck size small: ~80-88 sts

From S + 2n = 76: S = 76 - 2n

For body at separation:
F + 2n = 120

From crew neck: 2F + 2S + 4 = 88
→ F + S = 42
→ F = 42 - S = 42 - (76 - 2n) = 42 - 76 + 2n = 2n - 34

Also F + 2n = 120
→ (2n - 34) + 2n = 120
→ 4n - 34 = 120
→ 4n = 154
→ n = 38.5

Hmm, non-integer. Let me try slightly different target values.

Let me try:
- Front at separation: 116 sts (rather than 120)
- Sleeve at separation: 76 sts  
- Neck CO: 88 sts

116/8 = 14.5 — not exact for basketweave

Let me try front at separation = 112 (14 repeats of 8):

Body after adding 8 underarm CO each side:
Total body = 112 + 8 + 112 + 8 = 240 sts
240 / 2.6 = 92.3 cm (36.3") — that's barely any ease for a 34-36" bust. Not ideal.

What if underarm CO is 4 sts each side?
240 sts with CO 4 each: 112+4+112+4 = 232 sts = 232/2.6 = 89.2 cm. Even less ease.

Hmm. Let me try front at separation = 128 (16 repeats of 8):

Body = 128 + 8 + 128 + 8 = 272 sts = 272/2.6 = 104.6 cm (41.2") — that's 5-7" ease for small, which is generous/oversized.

Or with smaller underarm CO (4 each):
272 - 8 = 264... no wait: 128+4+128+4 = 264 sts = 264/2.6 = 101.5 cm (40") — still generous but within oversized fit.

Hmm. Let me step back and be more pragmatic.

Maybe I should reduce my body width target slightly and accept that the basketweave stitch panel = 112 sts (not 120).

Or: let me work around the basketweave count by having a 2-stitch edge on each side of the front basketweave (worked in stockinette as a "seam line"), so the basketweave sits within 120 front sts but is: k2, [basketweave over 116 sts], k2. But 116 ÷ 8 = 14.5 — still not exact.

OR: I can use a basketweave based on multiples of other numbers.

**Basketweave options:**
- Classic basketweave: 8-st multiple (k4p4, p4k4 rows) — needs multiple of 8
- 6-st basketweave: multiple of 6 — more options
- 4-st basketweave: multiple of 4 — most flexible

Let me use a **4-st basketweave** (2x2 or simple checker):
Actually the most common basketweave for sweaters is the k4/p4 version over 8 rows:
Rows 1-4: *k4, p4* repeat
Rows 5-8: *p4, k4* repeat

This requires multiple of 8 sts.

For the FRONT panel, let me make it work with multiple of 8:
If front separation sts = 120 → 120/8 = 15 repeats ✓

For this to work with my raglan math:
F + 2n = 120 (front target at separation)
S + 2n = ? (sleeve target at separation)

Let me now try different neck CO and accept that my sleeve will work out separately.

For the sleeve: at separation, sleeve has S + 2n sts. Add 8 underarm CO → sleeve starts at S + 2n + 8 sts.

For size small, typical sleeve top (just below underarm): ~84-90 sts

Let's target sleeve start (after adding underarm) = 88 sts
Sleeve at separation: 88 - 8 = 80 sts
So: S + 2n = 80

From front: F + 2n = 120
Subtract: F - S = 40

For crew neck (88 sts): 2F + 2S + 4 = 88 → F + S = 42
With F - S = 40: 2F = 82 → F = 41, S = 1

S = 1 sleeve starting sts — that's essentially nothing and unrealistic.

Clearly 88 sts neck doesn't work with F = 120 target AND reasonable sleeve width.

**The problem:** I'm trying to have front at separation = 120, which is 88 more than starting front sts (if starting front = 32 with 88 neck CO). But 88 additional sts means 44 increase rounds, and sleeves also get 88 sts added = 14+88 = 102 sleeve sts.

The sleeve getting 102 sts is unavoidable with standard raglan if front gets 88 extra sts. The only way to get fewer sleeve sts is to not use standard raglan geometry.

**SOLUTION**: This is correct! In a standard top-down raglan, the sleeves WILL end up with more sts than needed at the upper arm. That's why all top-down raglan patterns have an UNDERARM DECREASE SECTION or the sleeve immediately starts decreasing rapidly.

Looking at actual published patterns:
- Many top-down raglan small patterns for sport/DK weight have 90-110 sts for the sleeve at the point of separation
- They then decrease rapidly over the first few rounds/inches of sleeve
- OR: they work sleeve cap short rows to shape the sleeve

So 102 sleeve sts is FINE! The sleeve will then be worked down with decreases.

Actually wait — I said sleeve top (after underarm pickup) should be 88 sts. If separation sleeve = 102 and underarm CO = 8, then sleeve after pickup = 110 sts. Upper arm = 110/2.6 = 42.3 cm. For a small with 30-32 cm upper arm measurement, even with ease that's too big. Even with 4" ease, 34 cm upper arm would be 34×2.6 = 88 sts.

So the sleeve needs to decrease from 110 sts to 88 sts in the first couple of inches. That's 22 sts decreased, or 11 decrease rounds. Over about 3 cm of sleeve length — that's possible but aggressive.

Alternatively, I could use fewer underarm CO sts (say 4 instead of 8), making the sleeve pickup = 102 + 4 = 106 sts = 40.8 cm. Still needs to decrease to 88 sts (18 sts, 9 decrease rounds in ~3 cm).

ALTERNATIVE: Rethink the whole approach. Let me use a much smaller target front count.

**NEW APPROACH**: Let me target front = 96 sts at separation.
96 ÷ 8 = 12 repeats ✓

Body total: 96 + 8 + 96 + 8 = 208 sts = 208/2.6 = 80 cm (31.5") ← Too small!

Front = 104 sts: 104 ÷ 8 = 13 ✓
Body: 104 + 8 + 104 + 8 = 224 sts = 224/2.6 = 86.2 cm (33.9") — still too small.

Front = 112: body = 240 sts = 92.3 cm (36.3") — barely any ease.

Front = 120: body = 256 sts = 98.5 cm (38.7") — 2.7-4.7" ease ✓

The body size IS correct. The sleeve width issue needs to be managed by:
1. Having a rapid decrease section at the top of the sleeve
2. Or: The raglan increases are worked at different rates for body vs sleeves (extended shoulder shaping)

Actually, let me look at this differently. In a top-down raglan, the upper arm width at the point of separation is actually CORRECT as a proportion — it just looks large because it's measured flat. When worn, the arm folds the fabric.

Actually no. If sleeve = 110 sts = 42 cm circumference at upper arm, and the actual upper arm of a small is ~28-30 cm, then even with 4" ease = 38-40 cm. So 42 cm is just slightly too large. That might be acceptable.

Let me look at this from a finished upper arm perspective:
- Size small upper arm measurement: 28-30 cm
- With 2-3" ease (standard): 33-37.5 cm
- 35 cm × 2.6 = 91 sts

If sleeve at separation = 102 sts + 8 underarm CO = 110 sts → 42.3 cm (that's ~4.5-6" ease on the upper arm). That's quite a lot of ease on the sleeve.

For a more fitted sleeve, I'd want about 88-94 sts.

Options:
A. Use smaller underarm CO (4 each): sleeve = 102+4 = 106 sts = 40.8 cm (still 4" ease on 30cm upper arm)
B. Adjust the sleeve start stitches: in some raglan patterns, sleeve sections start with MORE sts than standard, meaning fewer increase rounds overall, but then both body and sleeves end up proportional.
C. Just design for 4" ease on the sleeve and note that it's a relaxed fit.

Actually, you know what? Let me look at a real published sport weight top-down raglan pattern and see what numbers they use:

Published Sport/5-ply top-down raglan (small):
- Typically cast on 84-96 sts for crew neck
- After yoke, sleeves have ~80-100 sts at separation depending on pattern
- Body: ~200-220 sts
- This means upper arm = ~80-100/2.6 = 30-38 cm... which accounts for 2-4" ease

Ah — I think my body target is too wide! Let me reconsider my target body circumference.

For a SIZE SMALL (34-36" bust) with **standard ease (2-4")**:
- Target finished bust: 36-40" = 91.5-101.5 cm
- With sport weight and typical patterns: ~36-38" finished = 91-96 cm
- That's about 237-250 sts

Let me try: finished body = 92 cm = 92 × 2.6 = 239.2 → 240 sts

240 ÷ 2 = 120 each side
But with underarm CO (8 each): separation body = 240 - 16 = 224 → 112 front + 112 back
112 ÷ 8 = 14 ✓

Now the sleeve math:
S + 2n = sleeve sts at separation
F + 2n = 112

For crew neck 88 sts: F + S = 42 (as before, with 4 raglan sts)
With F starting = 88-4-2S... 

2F + 2S + 4 = 88 → F + S = 42
F + 2n = 112 → F = 112 - 2n

S = 42 - F = 42 - (112 - 2n) = 2n - 70

For n to give S ≥ 10 (minimum reasonable sleeve start):
2n - 70 ≥ 10 → 2n ≥ 80 → n ≥ 40

With n = 40:
F = 112 - 80 = 32
S = 80 - 70 = 10
Sleeve at separation: 10 + 80 = 90 sts
With 8 underarm CO: sleeve start = 90 + 8 = 98 sts = 37.7 cm ← still quite wide for small

OK let me try n = 36:
F = 112 - 72 = 40... F = 40 sts at neck? That's quite wide.
S = 72 - 70 = 2 sts... 2 sleeve sts at neck? Not workable.

Neck = 88 sts:
F + S = 42

If F = 40, S = 2 → neck too wide front and minimal sleeve sts. Bad.

What if I use a different neck CO? Let's try neck = 80 sts:
2F + 2S + 4 = 80 → F + S = 38
F = 112 - 2n
S = 38 - F = 38 - 112 + 2n = 2n - 74

n ≥ 40 for S ≥ 6:
n = 42: F = 112-84 = 28, S = 84-74 = 10
Neck: 28+1+10+1+28+1+10+1 = 80 ✓
Sleeve at separation: 10+84 = 94 sts
With 8 CO: 94+8 = 102 sts = 39.2 cm upper arm

Still wide. But actually this is normal for a relaxed raglan.

Let me try: neck = 80, no underarm CO OR only 4 CO each:
With 4 CO each: sleeve = 94+4 = 98 sts = 37.7 cm ← with 28cm upper arm, that's 9.7 cm / 3.8" ease. Fine.

With 0 underarm CO: sleeve = 94 sts = 36.2 cm upper arm. 

Hmm, I think 4 sts underarm CO each side is more realistic:
Body at separation: 112 + 4 + 112 + 4 = 232 sts = 89.2 cm (35.1") ← that's less than 34" bust with only 0-1" ease. Too tight.

I'm going in circles. Let me just accept the following and move on:

**FINAL DECISION — simplified and clean:**

Tension: 26 sts × 36 rows = 10 cm on 3.5mm needles (2.6 sts/cm, 3.6 rows/cm)

Neck CO: 80 sts
Distribution: Back 28, Sleeve 10, Front 28, Sleeve 10, + 4 raglan sts
80 sts total ✓

Target body bust: ~97 cm (38") with 2-4" ease on 34-36" bust
Target body sts: 252... hmm. Let me pick a clean number.

Actually, let me just accept that the basketweave repeat might need adjustment. What if the basketweave is worked over a specific panel that I'll define precisely, regardless of it being the full "front"?

FINAL SIMPLIFICATION:

Let me choose my neck CO and increase rounds first, derive body and sleeve counts, then check if basketweave works, and adjust the basketweave setup if needed.

**START HERE:**

Sport weight: 26 sts = 10 cm, 36 rows = 10 cm (standard for 3.5mm needles)

Size Small target bust: 96 cm (38") with positive ease for 34-36" bust ✓
Body sts in round: 96 × 2.6 = 249.6 → round to 252 (divisible by 4, by 2, close to target)
252 ÷ 2 = 126 front + 126 back

For basketweave: 126 ÷ 8 = 15.75 — not exact
Try 128: 128 ÷ 8 = 16 ✓. With 128 each: 256 total body sts = 256/2.6 = 98.5 cm (38.7")

OK. 256 sts, 128 front, 128 back.

But with underarm CO:
- If I cast on 8 sts at each underarm (total 16): separation sts = 256 - 16 = 240 → 120 front + 120 back
  - 120 ÷ 8 = 15 ✓ (basketweave!)
  - Basket weave panel = 120 front sts after separation ✓

OR, if I "keep it simple":
- 120 front sts at separation
- Plus 8 CO sts on each side of front at underarm
- Total front in body round: 120 + 8 + 8 = 136... 

No wait. The 8 underarm CO sts are at the UNDERARM POINTS, not included in the front or back. They are separate, acting as "side seam" stitches. So:

Round: [120 back sts] [8 underarm CO left] [120 front sts] [8 underarm CO right] = 256 sts total body ✓

The basketweave is worked on the CENTRAL 120 FRONT STS.
The 8 underarm CO sts on each side of the front are worked in stockinette (or reverse stockinette to mark the side seams).

120 front basketweave sts ÷ 8 = 15 repeats ✓ 

And the body total = 256 sts = 256/2.6 = 98.5 cm ≈ 39" finished bust ✓ (3" ease on 36" bust)

OK. Now back to neck CO and increase rounds:

After n increase rounds:
Front at separation = 120 sts
With starting front F: F + 2n = 120

Sleeve at separation S_total = S_start + 2n
At separation, sleeve sts on holder = S_start + 2n
Sleeve after adding underarm: S_start + 2n + 8 = upper arm sts

For size small upper arm (generous ease for relaxed sport sweater): ~38 cm
38 × 2.6 = 98.8 → ~98 sts
So: S_start + 2n + 8 = 98 → S_start + 2n = 90

From front: F + 2n = 120

Neck CO: 2F + 2×S_start + 4 = N
For crew neck small: ~84-88 sts

Let's say N = 84:
F + S_start = 40
F + 2n = 120 → F = 120 - 2n
S_start = 40 - F = 40 - 120 + 2n = 2n - 80

Also S_start + 2n = 90 → (2n - 80) + 2n = 90 → 4n = 170 → n = 42.5 — non-integer!

Try N = 86: F + S_start = 41
4n = 2×90 - (41-1)×... let me redo:
S_start = 41 - F = 41 - (120-2n) = 2n - 79
S_start + 2n = 90: (2n-79) + 2n = 90 → 4n = 169 → n = 42.25 — still non-integer.

Try N = 88: F + S_start = 42
S_start = 42 - F = 42 - (120-2n) = 2n - 78
S_start + 2n = 90: (2n-78) + 2n = 90 → 4n = 168 → n = 42 ✓

So:
n = 42 increase rounds
F = 120 - 84 = 36 sts (front at cast-on)
S_start = 42 - 36 = 6 sts (sleeve at cast-on)
Check: S_start + 2n = 6 + 84 = 90 ✓
Sleeve after pickup: 90 + 8 = 98 sts ✓

Neck CO: 36+1+6+1+36+1+6+1 = 88 sts ✓
Neck circumference: 88/2.6 = 33.8 cm (13.3") — a bit snug but that's before the neckband stretches; with 2×2 ribbing stretched it would be ~36-38 cm. Actually 13.3" (33.8 cm) is fine for a crew neck.

Back at separation: 36 + 84 = 120 sts ✓

Yoke rounds:
42 increase rounds × 2 = 84 total yoke rounds
Yoke depth: 84 / 3.6 = 23.3 cm (9.2") ← reasonable for sport weight, size small ✓

**SLEEVE:**
Start: 98 sts (after pickup)
Target at cuff: 48 sts (cuff circumference ~18.5 cm = 7.3")
Actually let me calculate: cuff = 20 cm (8") → 20 × 2.6 = 52 sts
Sleeve length: 44 cm (17.5") from underarm to cuff edge (before ribbing)
Cuff ribbing: 7 cm (2.75") 
Sleeve knitting length before cuff: 44 - 7 = 37 cm = 37 × 3.6 = 133 rows = 133 rounds

Decreases: from 98 to 52 = 46 sts decreased, or 23 decrease rounds
Frequency: 133 rounds ÷ 23 = every 5.7 rounds → decrease every 6th round
Actual: 23 × 6 = 138 rounds → too many. 

Hmm. Let me adjust:
If I decrease every 6th round: 23 decreases use 138 rounds, but I only have 133 rounds.
Adjust: some decreases every 6th, some every 5th.

Or: round 52 up to 54 sts (27 cm cuff with some ease) → 22 decreases needed
22 × 6 = 132 rounds — close to 133 ✓

Decrease 22 times (2 sts each time), every 6th round:
98 → 98-(22×2) = 98-44 = 54 sts ✓

Actually: 22 decreases × 6 rounds = 132 rounds; but we have 133 rounds → work 1 plain round at start, then [dec round, 5 plain rounds] × 22 = 1 + 132 = 133 ✓

Or simpler: [dec round, 5 plain rounds] × 22, then 1 plain round = 133 rounds
The extra 1 round won't be noticeable; I'll just say "dec every 6th round 22 times."

Actually let me just say decrease every 6th round 22 times, which is 132 rounds. Then the remaining 1 round can be worked as a plain round at the end. Sleeve before cuff: 133 rounds. Fine.

Cuff: 54 sts in 2×2 rib for 7 cm. 54 ÷ 4 = 13.5 — not divisible by 4!
54 ÷ 2 = 27, and for k2p2 rib we need multiple of 4.
Let's change to 52 sts: 52 ÷ 4 = 13 ✓

If 52 sts at cuff: decrease from 98 to 52 = 46 sts = 23 decrease rounds
23 × 6 = 138 rounds; I have 133 rounds. Difference = 5 rounds.

Decrease every 5th round for some, 6th for others:
Let x = number of 6th-round decreases, y = 5th-round decreases
x + y = 23 decreases
6x + 5y = 133
From eq1: x = 23 - y
6(23-y) + 5y = 133
138 - 6y + 5y = 133
138 - y = 133
y = 5, x = 18

So: decrease every 6th round 18 times, then every 5th round 5 times (or some combination)
This is getting complex. Let me simplify the sleeve:

Option: Reduce sleeve length slightly or adjust cuff sts.

Let's say sleeve decreases: every 6th round 20 times = 40 sts decreased
98 - 40 = 58 sts → then for last section:
Rounds used: 20 × 6 = 120 rounds
Remaining rounds: 133 - 120 = 13 rounds
Decrease 0 more times, work 13 plain rounds, then cuff.
58 sts ÷ 4 = 14.5 — not divisible by 4.

Let me try: decrease every 6th round 22 times, then work even, then cuff with adjustment.
22 × 6 = 132 rounds; remaining = 1 round of plain.
98 - 44 = 54 sts. Then at cuff, decrease 2 sts in first round = 52 sts → 52/4 = 13 ✓

This works! Instructions: Work [dec, 5 plain] × 22 = 132 rounds, work 1 plain round (total 133 rounds before cuff = 37 cm). In first round of cuff, decrease 2 sts evenly to 52 sts, then work k2p2 rib for 7 cm.

Actually, let me simplify even further:
- At separation, sleeve has 98 sts
- Immediately work 4 rounds even to settle
- Then decrease round every 6th round until 56 sts remain
- Work even until sleeve measures 37 cm from underarm
- Cuff: k2p2 rib for 7 cm on 56 sts (56/4 = 14 ✓)

Decreases from 98 to 56: 42 sts = 21 decrease rounds
First 4 rounds even, then 21 × 6 = 126 rounds
Total = 4 + 126 = 130 rounds = 130/3.6 = 36.1 cm ← close to 37 cm ✓ (minor discrepancy OK)

56 ÷ 4 = 14 ✓ for k2p2 rib.

Actually, let me do this even more simply:
- Start with 98 sts
- Decrease 2 sts (one each side) every 6th round 21 times: 98 - 42 = 56 sts
- 21 × 6 = 126 rounds = 35 cm... plus some even rounds to reach 37 cm = work additional 7 rounds even (2 cm)
- Total sleeve to cuff: 133 rounds + let's say work 5 plain rounds after last decrease = 126+5 = 131 rounds, close enough to 37 cm

I'll write it as: "Work 1 round even, then [k1, k2tog, knit to 3 sts before marker, ssk, k1] every 6th round 21 times (56 sts remain). Continue even until sleeve measures 37 cm (14.5 in) from underarm, then work cuff."

**BODY:**

Body: 256 sts, start at left underarm
- 8 underarm CO sts (left) + 120 back + 8 underarm CO sts (right) + 120 front
- Back: stockinette
- Front: basketweave (120 sts = 15 repeats of 8)
- Body length: 38 cm (15") from underarm
- Hem ribbing: 6 cm (2.5")
- Knitting length before ribbing: 32 cm

Rounds: 32 cm × 3.6 = 115.2 → 116 rounds

Then work k2p2 rib for 6 cm = 6 × 3.6 = 21.6 → 22 rounds

For hem ribbing: 256 sts ÷ 4 = 64 ✓ for k2p2 rib ✓

**BASKETWEAVE STITCH PATTERN:**

Classic basketweave over 120 sts (multiple of 8):
Row 1 (RS): *K4, P4; rep from * to end
Rows 2-4: Same as Row 1 (knit the knits, purl the purls as they appear — actually in the basketweave, all right side rows 1-4 work k4p4, and purl rows 2-4 continue the pattern)

Wait, let me be precise. The basketweave in the round:

Since we're working in the round, RS is always facing. The basketweave in the round:

Rounds 1-4: *K4, P4; rep from * to end of front sts (120 sts)
Rounds 5-8: *P4, K4; rep from * to end of front sts
Repeat rounds 1-8 for pattern.

This creates the classic basketweave. ✓

But wait — I need to think about HOW