Being Early Is Being Wrong: The Carry Cost of a Correct Thesis

The most expensive sentence in trading is “I was right, just early.”

It is offered as consolation. It should be read as a confession, because in a market with a clock and a financing rate, early and wrong are the same trade with different feelings attached.

Michael Lewis told the story that made this famous. A handful of people saw the mortgage market for what it was, years before it broke. They were correct in every particular. They also spent those years bleeding premium, explaining themselves to investors who wanted out, watching a position that was right go relentlessly against them, and coming closer to being fired than to being vindicated.

They got paid because they survived the interval. Almost nobody does.

The clock is a cost, and it compounds

Give yourself a thesis worth having. You are certain, and you are correct, and when it resolves the trade returns forty percent gross.

Now add the thing nobody models: the position costs money to hold. Option premium decays. Shorts pay borrow. Futures roll. Margin ties up capital that had somewhere else to be. Call it a carry cost, and express it as a percentage of notional per month.

Every cell in that table is a correct thesis. Not one of them is a mistaken view of the world. The bottom-right corner loses half the capital, and the trader who put it on was right about everything except the calendar.

You will not be there to collect

The table above is optimistic, because it assumes you hold the position. You will not.

While the thesis waits, the mark moves. Carry drags it down steadily; noise pushes it around. Somewhere there is a level of unrealised loss you cannot sit through, whether because of a margin call, an investor, a risk manager, a spouse, or the simple biological fact that a person cannot watch a position bleed for eighteen months without doing something about it.

Model that. Carry of two percent a month, monthly noise of five percent, held for twenty-four months, forced out if the paper loss reaches your tolerance.

With a pain tolerance of thirty-five percent, which is enormous, which almost nobody has, you reach the finish line in fewer than one path in four. The thesis is correct in every single path. In three quarters of them the trade is closed before the market agrees.

This is the mechanism behind the sentence people quote and rarely examine: the market can stay irrational longer than you can stay solvent. It is not a warning about the market’s irrationality. It is a statement about the asymmetry between a view, which is free, and a position, which is not.

The retail translation

You are not shorting subprime. But you are doing this constantly, and the carry is disguised.

The swing trade you have held for three weeks. There is no borrow cost. There is an opportunity cost, and your capital has a hurdle rate whether or not you compute it. The trade that is going nowhere is charging you every other trade you did not take.

The short you keep rolling because gold is obviously overextended. Overextended is not a thesis, it is an opinion about a level with no clock attached. Add a clock and it becomes falsifiable.

The conviction that survives the invalidation level. Your stop was hit, and you re-entered, because you were right and the market was wrong. You have now paid the carry twice, and your entry is worse, and your reason for holding is no longer analysis. It is sunk cost wearing analysis as a costume.

How to write an expiry

It is not a guess about when the market will move. It is a statement about how long you will fund a hypothesis that has produced no evidence.

Anchor it to the thesis’s own mechanism. If the trade is built on a liquidity sweep resolving within the session, its expiry is the session. If it is built on a quarterly earnings cycle, it is the quarter. A thesis that cannot say when it should start working is not a thesis.

Make it a date, not a feeling. “If this hasn’t cleared the range by Friday, I am flat.” Written before entry, when you are still capable of writing it.

Price it. Compute the carry over the expiry window and subtract it from your target before you decide the trade is worth taking. A 2R trade that takes six weeks at real financing cost may be a 1.4R trade, and you may not want it.

Distinguish “not working” from “working slowly.” A thesis beginning to be right leaves evidence: structure shifting, the level holding on retest, the correlation waking up. Absence of evidence, over the window you specified, is the exit.

Conviction is a cost centre

Here is the part that stings.

The trader who is right and early looks identical, from every angle available to him in the moment, to the trader who is simply wrong. Same red position. Same explanation. Same defensible reasoning. Same feeling of clarity about what the market is missing.

The difference is revealed only afterwards, by which time both of them have already made every decision that mattered.

So you cannot use conviction to distinguish them, because both have it. You can only use the clock. And the clock is the one instrument in trading that does not care what you think.

Being early is not a milder form of being right. In a financed position it is the most expensive way to be correct, and the market charges you for the privilege every single month, until you stop.