I Was Terrified of Margin. Now I Govern It.

Governing margin instead of fearing it does not mean the fear goes away. It means a rule outranks it. Two Fridays taught me that, months apart, on the same book. The first time, the number screamed at me to sell and the rule said wait — and waiting was right. The second time, nothing felt urgent at all, and the rule said act now — immediately, no waiting period. What changed between those two Fridays was not my nerve. It was that I stopped asking my gut for the instruction.

TL;DR

The instinct and the instruction are not the same signal, and they do not move together. On one Friday my gut demanded I sell and the two-week persistence rule told me to wait — correctly, because the very next reading cleared the zone. Months later my gut said nothing at all while the same rule told me to act immediately, because the zone I had drifted into removes the waiting period entirely. Governing margin instead of fearing it is not becoming calmer. It is building a rule that does not ask how you feel.

What does it actually mean to govern margin instead of fearing it?

Four years ago a 10.95% week in my collateral would have ended with me at my laptop at midnight, selling something I had not planned to sell, to make a number I did not fully understand go back under a line I had picked out of instinct. I know this because it happened. I am not describing a hypothetical reader. I am describing a Friday in my own account, on a book carrying a $600,000 margin loan against securities that had just fallen from $2,100,000 to $1,870,000 — a reading of 32.09%, into what I now call the Harvest zone.

The thing that changed is not that I stopped feeling the fear. I still feel it. What changed is that the fear no longer gets a vote. There is a rule, and the rule was written down before the bad week happened, and on the bad week the rule — not the feeling — decided what I did. That is the whole distance between the old identity and the new one: my edge used to be believing I was disciplined; now my edge is a written, enforced process that does not care whether I am.

Margin sits in one of four zones — Clear below 30%, Harvest from 30% up to but not including 35%, Freeze from 35% up to but not including 40%, Forced at 40% and above — and the full zone system, with the instruction attached to each one, is where I keep the complete reference. This piece is not that reference. It is the story of the two Fridays that taught me the difference between reacting to a number and being governed by one — because the zones alone did not change my behavior. Watching the rule override my gut, twice, in opposite directions, is what did.

The fear never left. It just stopped being the one making the decision.

The week the number screamed and the rule said wait

The Friday before, the book read 28.57% — Clear, unremarkable, the kind of reading I used to barely register. Then the market had an ordinary bad week. Nothing broke; nothing was in the news that survived the weekend. But my securities fell to $1,870,000 against the same $600,000 loan, and the reading came back 32.09% — a first breach into Harvest.

Every instinct I had said the same thing: get out from under this, now, before it gets worse. That instinct is not stupid — it is the same one this blog's margin-dread piece treats as a real signal worth naming, not a character flaw. It is also, on its own, unreliable: it cannot tell the difference between a week that is the start of something and a week that is just a week — and neither, reliably, can I, in the moment I am feeling it.

So I did not act. Not because I felt calm — I did not — but because the rule I had written down said a breach has to show up in two consecutive weekly readings before it drives anything. One bad Friday is noise; two is a condition. I read again the following Friday. The book had recovered to $2,050,000, and the reading came back 29.27% — Clear again. The breach had not persisted. There was nothing to act on, and there never had been.

Friday	Securities	Reading	Zone	What I felt	What the rule said
Before	$2,100,000	28.57%	Clear	Nothing — didn't check closely	No instruction
The bad week	$1,870,000	32.09%	Harvest	Sell now, get under 30%	First breach — wait, read again
A week later	$2,050,000	29.27%	Clear	Relief, in hindsight	Breach did not persist — no action was ever needed

Notice what the rule did not say, even at the worst of it. Harvest is not an instruction to liquidate; it is a narrower one — trim, take some gains, bring the book back toward target — and even that narrower instruction only fires on the second confirming reading. Had I sold on the bad week, on my own authority, I would have crystallized a loss at the single worst Friday of the whole episode, for a governance reason that turned out not to exist. The two-week rule did not protect the portfolio from the market. It protected the portfolio from me.

The week that felt like nothing, and the rule said act now

The second Friday that mattered did not feel like anything at all, which is exactly why it is the more dangerous of the two. There was no single bad week to point to. The book had simply drifted, quietly, over a run of unremarkable Fridays, the way a slow leak empties a tire without ever announcing itself as a puncture. When I finally sat down and read the number properly, my securities stood at $1,450,000 against the same $600,000 loan — a reading of 41.38%. Forced.

My gut had nothing to say. That is the honest, uncomfortable part of this story: no single week had felt urgent enough to check closely, so by the time I checked, urgency was no longer optional. The instinct that had screamed at me over a one-week 10.95% drop in the first episode said almost nothing about a slower, larger drift into a genuinely worse zone. Fear is a good alarm for a sudden threat and a poor one for a gradual drift: each small step feels close enough to the last one to not register as new information, and the sum of many unremarkable steps can still add up to something remarkable.

The rule did not care that the week felt calm. Forced is the one exception to the two-week persistence rule that governed the entire first episode: a reading at 40% or above is acted on in the week it is first read, with no waiting period at all. I did not get to argue that it had crept up gradually and deserved a second look before I responded. The zone the number was in, not the mood of the week, decided what happened next.

The first episode taught me the rule can hold me back from a panic. The second taught me it can push me forward past a calm I had no business feeling.

It is worth being honest about how much room the rule still leaves. Even sitting inside Forced at 41.38%, the book was not close to a broker-initiated liquidation: generally, a margin account holding margin stocks must keep equity at or above 25% of the long securities' value, and from the Forced reading above the collateral would need to fall a further 44.83% before it reached that floor. Many brokers set their own house requirement above that regulatory minimum, especially on a concentrated book, so the real cushion can be smaller than the FINRA number alone suggests.

Either way, FINRA is explicit that a firm is not required to warn you before its own line is crossed: "A firm isn't required to notify you if your account equity drops below the minimum maintenance equity," and "Firms don't have to let you choose which securities or assets are sold to meet a margin call" (FINRA, "Know What Triggers a Margin Call," accessed 6 August 2026). Forced does not act immediately because the broker is about to act. It acts immediately because the entire reason to set a threshold well inside the regulatory one is so that the decision stays mine, made on my own schedule, instead of arriving later as someone else's.

That headroom is not a reason to relax the cadence, and I want to be precise about what this system does and doesn't catch. It reads once a week, on Friday. A move that happens on a Wednesday does not wait for me to notice it, and a broker's own liquidation authority does not run on my calendar either. The two-week rule and the weekly reading govern an ordinary drift measured in weeks; they are not a substitute for checking the account the same day a genuinely sharp, fast move happens. If your collateral falls hard in a single session, that is a reason to read the number early, not to wait for Friday.

A Forced reading does one more thing, quieter than any of this. This system also keeps a weekly Governance Score — one number, out of 100, built from how well the whole operation is run: the margin zone, how far allocation has drifted, whether the weekly reading happens on schedule, and whether there's a written policy at all. A Forced reading caps that whole score at 34, no matter how the other three components are doing, which lands it in the lowest band the framework has, Exposed / At the Edge. Nothing about that number forces a trade the way the zone instruction does. It just keeps score, whether or not anything else is making you look.

The Instinct Ledger

Here is the artifact these two Fridays turned into. It is deliberately small: one row, once a week, for as long as you hold a margin balance. It does not ask you to compute anything new — only to write down, side by side, what your gut is saying and what the rule is saying, before you know whether they will agree.

Fill it in for this week's book. Use your own numbers, not the ones above.

This Friday's date	What my gut says, in one line	This week's zone	Second consecutive reading in this zone or worse?	What the rule instructs	Did I follow it?
{{fill in}}	{{fill in}}	{{Clear / Harvest / Freeze / Forced}}	{{Y / N — N/A if Forced}}	{{wait / act, and what "act" means in that zone}}	{{Y / N}}
…	…	…	…	…	…

Two things become visible fast if you keep this honestly. The first is how rarely the gut column and the rule column say the same thing — mine agreed on maybe one week in five. The second, and the one that actually changes behavior, is the "did I follow it" column: the first few weeks are the test of whether you have a rule or just a document. My own ledger has one row, in the second episode above, where the honest answer to that last column was briefly no. I want to be straight about that rather than present four years of perfect compliance — the ledger's value is that it makes the gap between the instruction and the action visible to you, on paper, the week it happens, rather than letting it stay a private thing you were more disciplined about in memory than you were in fact.

One limit worth stating plainly: this ledger is only useful once you already know which zone you are reading. It says nothing about whether your zone boundaries or your denominator are the right ones — that is a governance question this piece deliberately does not reopen. The full mechanics of when a breach actually drives an action — the two-week rule, the Forced exception, what each zone's instruction actually is — live on the margin safety reference.

Frequently asked questions

What is the two-week persistence rule, and why doesn't it apply to Forced?

It is the rule that a margin breach has to appear in two consecutive weekly readings before it drives an action, so a single ordinary bad week — the kind markets have often — does not trigger a trade. The one exception is the Forced zone, 40% utilization and above, which is acted on in the week it is first read, with no waiting period. The reasoning is that Forced is the existential zone: waiting a week there doubles the exposure window on the one reading where that is least affordable, so the rule that protects you from overreacting everywhere else is deliberately absent there.

Isn't waiting when the market is falling dangerous?

It can be, which is exactly why the rule does not say "always wait." Inside Harvest (30% up to 35%) and Freeze (35% up to 40%), a single week's move is genuinely ambiguous — it could be noise or the start of a trend — so the rule asks for a second confirming reading before it acts. The moment a reading reaches Forced, 40% and above, that wait is removed entirely and the instruction is to act the same week. The rule changes its own patience depending on which zone the number is in; it does not treat every falling week the same way.

How do I know if I'm reacting to fear or to a real signal?

You mostly can't tell in the moment, which is the entire point of writing the rule down before the bad week arrives rather than during it. On one Friday my own book fell 10.95% in a week and my instinct screamed to sell — the two-week rule said wait, and the very next reading proved the fear had been louder than the actual risk. Months later, on the same book, a slow multi-week drift crossed into a worse zone without ever feeling urgent — my instinct said nothing, and the rule said act immediately, because the calm had been quieter than the actual risk. A rule fixed in advance is what catches both directions, because it does not consult the feeling either time.

What if my reading is close to a zone boundary — does it count as a breach?

Zone boundaries need an explicit inclusivity convention or a reading that lands exactly on the line belongs to two zones at once. The convention this piece uses throughout: Clear is below 30%, Harvest runs from 30% up to but not including 35%, Freeze runs from 35% up to but not including 40%, and Forced is 40% and above. A reading of exactly 30.00% is Harvest, not Clear; a reading of exactly 40.00% is Forced, not Freeze.

Do I need to feel calm before I can trust myself to govern my portfolio?

No — and waiting to feel calm is a trap. A slow, quiet drift into a genuinely worse zone can feel completely unremarkable right up until the reading itself says otherwise, which means calm is sometimes exactly the wrong feeling to be trusting. Governing a portfolio does not require managing your emotional state first; it requires a rule that produces the same instruction whether you are anxious, calm, or somewhere in between, and then actually following it — which is what a simple weekly log of "what I feel" versus "what the rule says" is built to make visible over time.

Incomestead recommends. You decide.