The published Algomyth engine backtest averages $18,814 a year net on a mid-size account at the Medium sizing level. That is not the number that reaches your bank. Several more costs sit between a good month on the screen and money you can actually spend, and most performance figures you will see elsewhere quietly pretend they do not exist.
So here is the full stack, line by line, with the honest note attached to each: whether it is already inside that $18,814 figure or still has to come off it. None of these costs are hidden or unusual. They are just rarely added up in one place.
Most cost mistakes begin by starting from a gross figure, which is why there is not one here to start from. The published $18,814 a year already has commissions and slippage subtracted, trade by trade, at $1.90 per contract round turn plus a tick. Everything else in this article comes off that figure too, so treat $18,814 as the ceiling for an average year, not the take-home. What each cost does to it is the whole question, worked out honestly in how much a trading bot actually makes.
Costs fall into two piles: the ones already inside the net figure (commissions and slippage) and the ones still to come off it (data, infrastructure, evaluation fees, the license and taxes). Confusing the two is how people overestimate what they keep. Start from net, then subtract the second pile.
Every round turn costs a commission, roughly $1.90 per contract, plus about a tick of slippage, the difference between the price you expected and the price you got. Across the 1,164 trades of the backtest report, that is charged inside every trade before the report prints anything. On the micro Nasdaq it works out to about $2.40 a trade, against a net edge near $75 a trade.
Two dollars and change a trade sounds trivial because this system trades once a day. It is not trivial for a system that trades twenty times a day, which is the quiet reason high-frequency retail systems die: costs scale with the number of trades, not with profit. The full breakdown is in slippage and commissions. For the Algomyth systems it is a solved cost, already in the published figure. It is listed here only so you know it is not waiting to surprise you later.
You need NinjaTrader 8 and a real-time data feed for US index futures. How much that costs depends entirely on how you access it. Many prop firms bundle the platform license and the data into a single monthly platform fee, in which case your marginal cost here is close to zero. If you pay it yourself, a non-professional real-time CME feed is a low double-digit monthly subscription, and NinjaTrader 8 can be run on a free license tier with a supported account or leased separately.
The number to write down is therefore a range, not a certainty: anywhere from about $0 to $50 a month depending on your prop firm and how you license the platform. The one thing you should not do is pay a professional data rate by accident, so confirm you are classified as a non-professional. For an individual trader it is a large difference.
The machine has to be on and connected at 9:30 AM ET. If you leave a computer you already own switched on, the cost is a few dollars of electricity and a one-time uninterruptible power supply. If you rent a virtual private server so the platform runs in a data center that never sleeps, that is a low-double-digit to roughly $40 monthly cost. Neither buys you speed, which a once-a-day system does not need; both buy uptime. When each one makes sense is laid out in do you need a VPS for automated trading. Budget somewhere between $0 and $40 a month here, and start at the low end.
This is the line item that separates an honest cost estimate from a brochure. If you trade funded capital, you pay for an evaluation before you trade a prop firm's money, and funded accounts get closed by their own drawdown rules, including accounts that were making money. When one closes, you buy another evaluation.
A 50k evaluation costs about $100. How many you get through in a year depends on your firm's drawdown rule and on when you start, so there is no published figure for it and we will not invent one; what the report does say is that the worst drawdown of the period, $5,273, is deeper than a typical 50k evaluation allows. Budget for more than one, and remember it is money paid up front, in cash, whether or not the account ever pays you back. In a flat stretch you pay it and receive nothing. Typical evaluation costs run about $100 for a 50k, $250 for a 100k and $400 for a 250k.
Change the account size, the number of contracts, the commission rate and the fees, and see what the same backtest looks like on your numbers instead of the defaults.
The system itself has a price, listed on pricing. Whether you pay once or on a subscription changes how it lands on a monthly budget, and the trade-off between the two models is covered in subscription versus license. Amortized over a year of use it is a modest monthly figure next to the evaluation fees, but it is a real cost and it belongs in the total. Note what it does not entitle you to: a result. It is the price of the tool, not a share of any profit.
Profit is taxed, and futures get their own treatment in the United States. Gains on US index futures generally fall under Section 1256, which uses a blended 60/40 long-term and short-term rate and marks open positions to market at year end. That is often more favorable than the rate on ordinary short-term gains, but it is still a real subtraction from every profit figure on this site, and it is not a monthly bill: it is a share of what is left after all the costs above. The mechanics are in futures trading taxes and Section 1256, and the specifics are between you, a tax professional and the IRS.
Here is the whole stack in one place, for an average active month, starting from the net figure the backtest actually produced.
| Cost | Already in the net figure? | Rough monthly amount |
|---|---|---|
| Commissions and slippage | Yes, already deducted | ≈ $2.40 a trade |
| Platform and market data | No | $0 to $50 |
| VPS or electricity | No | $0 to $40 |
| Evaluation fees | No | About $100 each, as often as an account fails |
| The license (amortized) | No | See pricing |
| Taxes (Section 1256) | No | A share of profit |
Add the second pile together and, in an active month, the recurring out-of-pocket costs beyond commissions land somewhere around $50 to $90 for data and the machine, before evaluation fees, before the license and before tax. Against an average year of $18,814 that is a small slice. The point is not that the costs are large. It is that they are real, up front, and yours, and two of them (evaluation fees and taxes) are the ones people forget precisely because they do not arrive as a tidy monthly invoice.
Put concrete numbers on it. Start an average year at $18,814 net of commissions, then subtract roughly $360 for data and $240 for a VPS, plus whatever you spend replacing failed evaluations, and you are under $18,000 before the license and before tax. Whatever the license adds, and whatever Section 1256 takes, comes off that. It is still a healthy result in the simulation, and it is smaller than the headline suggests.
Commissions come out of every trade. Evaluation fees come out of your pocket before a single trade pays you back.
The flat months are where this matters. In a month the system makes little or nothing, the costs do not pause. You still pay for data, for the machine, and for any evaluation you had to replace. That is why the sensible way to size the whole operation is against a bad stretch, not an average one, and why the question of how much starting capital you actually need is worth its own article: how much money do you need.
Hypothetical performance. The Algomyth figures here come from a backtest over historical data, not from a live account. Simulated results are prepared with hindsight, carry no financial risk, and cannot fully reflect real execution, slippage or liquidity. Past performance, real or simulated, does not guarantee future results.
For a once-a-day system like this one, the recurring costs beyond commissions typically land around $50 to $90 a month for platform and data plus a VPS or electricity, with evaluation fees, the license and taxes on top. Commissions and slippage, about $2.40 a trade on the micro Nasdaq, are already inside the published figure. Your exact total depends on whether your prop firm bundles data, whether you rent a VPS, and how often accounts are consumed.
Yes. The published $87,359 is a net figure, with commissions of $1.90 per contract round turn and a tick of slippage already subtracted on every one of its 1,164 trades. There is no gross figure published alongside it, on purpose. That is why this article treats commissions as a solved cost rather than something waiting to reduce the number further. Every other cost here, from data to taxes, still comes off that figure.
No. A VPS is optional, and for a system that trades once a day it buys uptime rather than speed. If you already own a computer with reliable power and internet, leaving it switched on with sleep disabled and a wired connection does the same job for the cost of electricity and a one-time battery backup. A VPS becomes worth its low monthly cost when you travel, have unreliable power, or run several accounts where one missed morning outweighs a year of hosting.
Because the evaluation is real money you pay up front to a prop firm before you ever trade its capital, and it is not refunded if the account later fails a drawdown rule. Even a profitable system has accounts closed by trailing drawdown limits, and each replacement is another fee. The published worst drawdown, $5,273, is deeper than a typical 50k evaluation allows, so evaluation fees belong in any honest cost total.
Gains on US index futures generally fall under Section 1256 of the tax code, which applies a blended 60/40 long-term and short-term rate and marks open positions to market at year end. For many traders that treatment is more favorable than the ordinary short-term rate, but it is still a genuine cost that reduces every net figure. Tax treatment depends on your personal situation, so confirm the details with a qualified professional rather than relying on a general description.
Everything we write comes back to one idea: verify, then decide. The live terminal exists for exactly that.
Unlock Algomyth Live