The Business on One Page

The card

Module 13 finished with a card of four columns and a deletion test. This one finishes with five rows and no test at all, because the rows are not competing measurements of one thing. They are five separate ways the business can stop, and they are printed in the order in which they bind.

Take rows three and four to the year, because the year is where a household lives. The net earning of 99 a month is 1,191 a year, which is 1.19 per cent of the account. The withdrawal of 500 a month is 6,000 a year, which is 6.00 per cent. The business is short by 4,809 a year before tax, before the fixed costs lesson 80 itemised, and before a single thing has gone wrong.

The fourth column is the one this module was written to fill in. Every row above it is a measurement and every row is fine on its own: 4.875 trades a month is a normal pace for a daily rule, 0.1230 is a normal round trip, and 500 a month is a modest wage. Nothing here is a mistake. The business fails because the five are multiplied together rather than considered one at a time, and the multiplication is what no report performs.

Reading the order

Start at the top and stop at the first row that binds. The pace never binds by itself; it only sets how often everything else happens. The cost binds when it is a large fraction of the edge, and at 79.7 per cent it has already bound before the reader reaches the third row.

That is the sentence this whole module exists for. The account arithmetic in lessons 97 and 99 is arresting, and it is downstream: it is the arithmetic of a business whose edge was already reduced by four fifths on the previous page. A trader who fixes the withdrawal, or raises the capital, or extends the horizon, has fixed rows three, four and five while row two is still taking the money.

And row two is the one with a known repair. Lesson 11 measured the same round trip running from a fraction of a basis point to hundreds depending on the instrument. Lesson 98’s third exit was to hold longer, so that R grows while the 0.1230 stays fixed. Both of those change row two directly, and both of them make rows three and four look completely different without touching either.

What the repair is worth

Take the same business and change one thing: hold each trade long enough that one R is 3.0886 a share rather than 1.5443, which is two bars of movement rather than one. Nothing else moves. The cost is still 0.1230 a round trip and the edge is still a tenth of an R, so the cost falls to 0.0398 of an R and the net edge rises from 0.0204 to 0.0602.

Doubling the holding period triples the net earning and cuts the wait by a factor of nearly nine, because the wait moves with the square. It is still 37 years, which is still not a business, and that is the honest reading: the repair that helps most does not get this system to viability, it gets it from impossible to merely infeasible.

The reason is row one, and row one is the one row this course cannot repair. At 4.875 trades a month there are 58.5 chances a year for the edge to express itself, and no amount of improving each chance compensates for how few of them there are. Lesson 98’s middle column is where a business lives: 45 trades a month buys three winning months in four, and 45 a month is a different rule family on a different timeframe, not this one tuned.

The card does not say do not trade. It says this configuration — a daily crossover on one instrument, five trades a month, a hundred thousand, a wage taken monthly — is not the configuration. Every row is a dial, and the module’s finding is which dial to turn first.

So write your own five rows before your next month, in this order, and stop at the first one that binds.

What this does not settle

That the five rows are the whole business. They are the five this course can measure. Lesson 78 has tax, lesson 79 has the alternative uses of the same person’s time, and lesson 80 has the fixed costs of the operation itself, none of which appear on this card and all of which subtract. The card is a floor on the problem rather than a description of it.

That the two-bar repair is available. Doubling the holding period is written here as a free change and it is not: a rule that holds twice as long is a different rule with a different edge, and lesson 74 already showed this family’s whole correlation structure reversing between the first fourteen moves and the last fourteen. The table above holds the edge fixed at a tenth of an R while changing the thing that generates it, which is exactly the kind of assumption module 13 spent five pages refusing to make. It is here because the direction is right even though the figure is not.

That the pace cannot be raised on this instrument. It can, by running more of lesson 63’s 253 rules at once, which is what lesson 76 pointed at: all 253 together leave only one of twenty-eight days quiet. What that does not fix is lesson 71’s 0.9472 of correlation, so the trade count rises far faster than the number of independent bets, and lesson 91 measured the result: the whole grid earns 4.16 against a single rule’s 4.10.

That any of this is advice. It is a card with five rows on it computed from figures this course measured, and the reader’s own five rows will differ in every one. The transferable object is the order of the rows, which is the only part of this page that does not depend on the numbers in it.

And the concession that costs most: a hundred lessons have measured what a rule does, what it costs, what a book of rules does to itself and what a business built on one can support, and not one of them has measured whether a particular person should do this. That is not modesty. It is the boundary of what arithmetic can reach, and the reason every page in this course ends with what it does not settle rather than with what to do.