Putting fixed monthly costs in the journal: how many trades does the VPS and the subscriptions need to cover?

[Fairman]

Putting fixed monthly costs in the journal: how many trades does the VPS and the subscriptions need to cover?

Monthly costs for one trader, made up for the example: VPS $25, an indicator subscription $15, a news and data feed $10. Total $50 a month, $600 a year.

How big that is depends on the account. As a share of the balance:
$3,000 account: $50 is 1.7% a month, about 20% a year.
$10,000: 0.5% a month, 6% a year.
$20,000: 0.25% a month, 3% a year.

So the same $50 means two different things. On a $3,000 account the strategy has to earn almost 2% a month just to cover the tools before it makes any money. On $20,000, it barely registers.

In trades: say 88 trades a month (4 a day over 22 days). $50 spread over 88 trades is $0.57 a trade. If R is $30 on the small account, that's 0.019R per trade, which is a fair slice of an edge of 0.1R. On the $20,000 account with R at $200, it's 0.003R.

A row for fixed costs in the monthly journal, with the amounts and a yes or no for whether each one was used that month, is a plain way to see what the running of the setup costs against what it earns.

[FXS HongKong Scalper]

A $3,000 account paying 20% a year in tools needs an exceptional strategy just to break even. A useful exercise is to list which tools actually changed results in the last three months. Many subscriptions get kept out of habit. For small accounts, free alternatives for news and charting, and a VPS only once the strategy is automated, can cut that $50 a lot.

[Fairman]

Annual payments are easy to forget. A $120 yearly subscription feels cheap on the day it's paid and invisible in the other eleven months. Spreading it as $10 a month in the journal puts it where it belongs. The same goes for challenge fees and platform licences. One way is to enter the full amount in the month it's paid and then an equal share in each monthly total, so every month carries its fair part. The sum over a year is the same either way, but the monthly view stops one good month from hiding a cost that turns up in another.

[Fairman]

Which items count? Everything that exists because you trade: VPS, charting subscription, data or news feed, indicators and EAs bought or rented, prop firm fees, courses, a second monitor if you want to be thorough, and the bank charges on withdrawals. Left out of the total are costs of living, and that's a personal call. The test is that without trading, the item wouldn't exist. A sheet with a column for each category and a monthly total makes the comparison with net trading profit direct, and the first time you add everything up the figure is often a surprise to the person doing it.

[Amanhasnoname]

I did this exercise at the start of the year and it was a bit of a shock. Between a VPS I wasn't really using, two indicator rentals and a charting upgrade, I was paying around $70 a month on an account of about $4,000. That was over 1.7% a month before I made a single pip, and my average month was only around 3%.

What I changed was simple. I cancelled everything that hadn't been opened in the past thirty days, moved charting back to the free version, and only kept the VPS because one EA genuinely needs to run through the night. That brought it down to about $22 a month, and my net results looked noticeably better without changing anything about the trading.

I also now log the fixed costs as a separate line in R, like you suggest. Seeing that the tools were eating close to a fifth of my monthly profit was much more persuasive than seeing a dollar amount on a card statement.

[Fairman]

Compare it with the profit, not with the account. Net trading profit for the month, minus fixed costs, is what the activity produced. If a month shows +$180 from trading and $50 in costs, the real figure is +$130. A string of small months can disappear into the cost line. A simple rule: when the fixed costs are more than a quarter of the profit for three months running, it's time to ask whether each one is needed. A VPS used by an EA that runs every day is a different thing from a subscription opened twice a month.

[Fairman]

Review the list every quarter, with the yes or no column. Anything with several no's in a row is a candidate to cancel, and the saving is easy to calculate: $15 a month is $180 a year, which on a $3,000 account is 6% of the balance. Cancelling a subscription that isn't used is among the few ways of increasing returns with no market risk. Some tools can be paused instead of cancelled, so check. If you do cancel, put the date in the journal so that you remember to see whether anything in your process changed.