Correlation Is a Weather System
A note on books that look diversified on a watchlist and identical in a risk-on hour — written as opinion, not as a model you should copy.
I think of correlation as weather: sometimes the symbols move like separate towns, and sometimes the whole region is under one front.
Watchlists lie because they list names. Risk does not care about names. A long index future, a long high-beta basket, and a short volatility expression can be one sentence: “I need calm risk appetite to continue.”
SCUTA’s risk articles say this in classroom language. Here I will say it as a preference: I would rather take one clean ticket than three rhyming ones that make me feel busy.
When the front arrives — a policy print, a sudden dollar lurch — you find out whether you had a book or a pile. I would rather find that out on a small combined R.
Questions we hear next
Do I need a quant model?
No. You need the habit of asking what would make every ticket lose together.
Are two index futures always the same bet?
Often they rhyme. Sometimes one leads. The journal should note when you treated them as independent and they were not.
Does this apply to a single-product scalper?
Less as portfolio math, more as session weather: if you stack the same impulse five times, you still have one idea.
Is this advice to hedge?
No. Hedging is a separate craft with its own costs. This is a plea to count.
Funded Rails Are a Product, Not a Personality Test
Commentary on evaluation accounts: they are designed constraints. Treating them as a referendum on your talent is how people press.
What a Quiet Week Is Worth
Opinion: a week of stand-asides and scratches can be the most professional week on the calendar, even if it will never trend on a feed.
The Unfashionable Case for Small Size
Personal view: most students are not under-levered. They are over-entertained. Small size is how the playbook gets a fair sample.