read better; both are defensible against the table.)
بسم الله الرحمن الرحيم
Assalamu alaikum everyone,
Twenty-four weeks since I last posted, and they were not quiet weeks.
If you held a halal portfolio through 2026 and felt the rules kept changing underneath you, you were not imagining it. Semiconductors fell 16% in a single week in July and recovered all of it. Gold — the thing most of us hold for protection — spent most of the year falling alongside the market instead of against it. Energy moved on policy rather than on demand.
This post is about what that year taught me about building a halal portfolio — not a tip sheet, and not a record of my own trades. One idea, the numbers behind it, and what I am watching next.
What actually broke this year
The hardest thing to unlearn in 2026: a position can be right on the business and still be wrong on the timing, for reasons that have nothing to do with the business.
Oil did not trade on supply and demand this year. It traded on policy.
Semis did not trade on earnings in July. They traded on what people believed about AI spending.
Buy-and-hold assumes the market eventually prices the company. In a year when prices are set by decisions made in capitals, eventually can arrive long after the damage is done.
So I spent the quiet months on one question: where is capital actually rotating, and which sectors are being left behind?
Insurance looks like a loss in a good year
In the framework I use, 40% sits in protection — 30% gold, 10% sukuk. Both are down this year. GLD is −3.4%. SPSK is −1.9%.
On paper that is a drag. It is not a broken thesis. It is what protection costs in a year when nothing goes wrong with the thing you own.
The growth sleeve is up between 56% and 79% over the same stretch. In a year like that, every dollar sitting in gold is a dollar that did not compound — and the bill arrives as a negative number next to GLD.
The bill is the product.
Gold is not there to beat semiconductors. It is there so that if July’s 16% drawdown had kept going instead of reversing, 40% of the portfolio would not have gone with it.
So the question to ask of a safety sleeve at year end is not whether it paid off. It is whether the premium was priced right — and that is a question about how much protection was being bought, not about whether gold had a good year.
The numbers
Price returns through the close of Friday 9 October 2026. Excludes distributions.
| This week | Since 25 Apr | Since 4 Jan | Role | |
|---|---|---|---|---|
| SOXX | −5.0% | +21.3% | +78.6% | Growth · semis |
| SMH | −4.3% | +19.1% | +61.6% | Growth · semis |
| SHOC | −4.1% | +15.6% | +56.5% | Growth · semis |
| XOM | +3.0% | +14.9% | +40.5% | Energy |
| GLD | +1.2% | −11.2% | −3.4% | Safety · gold |
| SPSK | +0.5% | −2.2% | −1.9% | Stable · sukuk |
On allocations. The framework’s published allocation is 60 / 30 / 10 — Growth / Gold / Sukuk — set on 13 January. The April post was headed 60/20/20; that header was stale. The screen’s own regime model currently runs 70/15/15. Where the two differ, the published number is the one I write to.
Three things a year like this does not change
- A pullback is not an entry. Semis have had two meaningful ones since the summer: 16% over seven sessions in July, 7% over three days in August. Both recovered in full. Knowing that afterwards is easy; knowing it on day two is not, which is what a checklist is for.
- A weight that only gets held while it is winning was never a weight. Cutting gold after nine months of lagging is selling the insurance the week before you find out whether you needed it.
- Compliance has an expiration date. A name that screened clean in January is not a name that screens clean today. Ratios drift, filings update, and the screen gets re-run before anything moves.
What I am watching
- Whether this week’s pullback becomes something. Semis gave back 4–5% over the week and still sit roughly 20% above their July low. That is not a test of anything yet — the question is whether it turns into one.
- Whether this week was gold waking up. Look at the table again: in a week semis fell 5%, gold rose 1.2%. That is the first time this year the hedge has behaved like one when it was needed. One week is not a pattern, and I am not reading it as a signal. But it is the first evidence in nine months that the sleeve does what it is held for.
- Sukuk spreads. A stable sleeve that moves 1% in a year is doing its job. One that starts moving more is saying something about liquidity.
One question
When a position hits its stop, do you take it — or do you give it room?
I take it, every time. This year gave me several reasons to regret that and none to change it.
I am less interested in which camp you are in than in what put you there. Was it a rule you read somewhere, or a position you once held far too long?
One note on where this work lives now. I said monthly updates in January and did not keep to it, and part of the reason is that a forum thread is the wrong shape for most of what I wanted to write. The charts, the full screen behind a decision, and proper answers to the questions people send me all need more room than a post like this gives them. So that work now has its own home — a site, and a Sunday email for anyone who would rather have it in their inbox. You are very welcome to both: purefolio.ai. The monthly summary still goes up here, and the next one is the first Saturday of November.
والله أعلم
These updates reflect my personal approach to halal investing and are shared for educational purposes only. Not financial advice.
Amad — Purefolio — Madinah · Zoya community member since 2021