Skip to content
Kestrel

Automated trading · from $0.50

Start automated trading with $0.50

Kestrel is a trading robot. Connect your trading account and it trades it for you — on its own, day and night, while you sleep. Or put in as little as $0.50 and let it trade alongside everyone else’s. You pay 5% of the profit, and nothing at all when it loses.

Every figure here traces to 2,745 recorded trades — losing years and drawdowns included.

Where the pooled account stands today

Auto-updating

Read live from our own books and updated while this page is open. Nothing here is a placeholder or a counter that climbs on its own — when it says nobody has deposited yet, nobody has.

Active users
2
4 signed up in total
Total investment
$50.23
put in by members, confirmed
Total profit
+$0.00
made for members, +0.0% on what went in
This year
+0.0%
measured since 2026-09-07, not annualised
Running for
Not set
start date not set in admin
80.5%
of simulated years ended in profit
which means about one in five did not
+30.9%
median one-year result
after the 5% performance fee
24%
median drawdown along the way
41% in the worst 5% of runs
2,745
trades behind these figures
2021-09-08 to 2026-09-03

How it works

On the managed plan your money never leaves your own brokerage account. We place trades in it; we cannot withdraw from it.

Connect your own account

You open an account with your broker, in your name. Kestrel connects to place and manage trades. Withdrawal rights stay with you, and you can disconnect at any time.

Set your own limits

Risk per trade, maximum open positions, daily trade cap, and a pause switch. The limits are enforced on our side, not suggested in a document.

The system trades the rules

The same rules, every signal, with no discretion and no overrides. Entries, stops and targets are fixed before a position opens.

A bank, a professional trader, or this

A bank pays you almost nothing, but it is close to impossible to lose money in one. A professional trader might make you more, but you rarely see their real numbers, and most charge you whether they win or lose. Kestrel is built to fix that second part specifically: a fixed rule with no exceptions, every result published including the losing years, and a fee only when it actually makes you money. It is not built to fix the first part — this still carries real risk, same as any trading does.

Can you actually lose money

A bank

Almost never

A human pro trader

Yes — often more than you expected

Kestrel

Yes — median drawdown 24%, published up front

What a typical year looks like

A bank

Very little, but steady

A human pro trader

No way to know until it has already happened

Kestrel

+30.9% historically — never guaranteed, and about 1 year in 5 finished down

What you pay

A bank

Usually nothing

A human pro trader

Often a fee whether they win or lose, plus a cut of profit

Kestrel

Nothing — unless it makes money, then 5% of the profit only

Do you see what really happened

A bank

Just your balance

A human pro trader

Usually a few good months, not the full history

Kestrel

Every trade since 2021-09-08, wins and losses both

Can a bad mood change a decision

A bank

No decisions being made

A human pro trader

Yes — fear and fatigue are part of being human

Kestrel

No — the same fixed rule, every time

Works nights and weekends

A bank

Not applicable

A human pro trader

No — a person has to sleep

Kestrel

Yes — markets are open, it is watching

Minimum to start

A bank

Usually nothing

A human pro trader

Often thousands

Kestrel

$0.50

The Kestrel figures are simulated results from 2,745 recorded trades, 2021-09-08 to 2026-09-03 — not a guarantee, and not the same as a live track record. See the full methodology for exactly how they were produced.

What we will not tell you

Most platforms in this category publish a rising line and a large percentage. We think that is the least useful thing you could be shown, so here is what that line leaves out.

Read the full methodology

We will not promise a monthly profit.

At this system's measured consistency, roughly four to five months in a year finish below where they started. That is what the arithmetic gives; anyone claiming otherwise is describing something other than trading.

We will not hide the drawdown.

The median run gives back 24% from a peak before recovering, and the worst five percent of runs give back 41%. If that number would make you close the account, this is not the right product.

We will not call a backtest a track record.

These figures come from simulation over 2,745 recorded trades between 2021-09-08 and 2026-09-03. Live trading is recent and its sample is still too small to confirm or contradict them.

We will not round a statistic in our favour.

The edge measures 0.0611 R per trade at a t-statistic of 2.27. The instruments were chosen from a wider search, which inflates that statistic — so treat it as suggestive, not settled.

We will not pretend $0.50 is an investment.

The minimum is that low so you can watch the system run with your own money before committing anything you would miss, and because a platform that only works above a certain balance has decided who is allowed in. Sending it costs you about a dollar in network fees — start small to test, not to earn.

Year by year

Simulated at 1.00% risk per trade, net of the 5% performance fee. Partial years at each end of the sample are shown as they are, not annualised.

YearTradesWin rateResult
20219332%-1.4%
202238133%-15.2%
202367238%+39.0%
202462838%+25.9%
202559839%+47.9%
202637342%+76.7%

Try it on a demo account first. It costs nothing and proves more than any page can.

The demo runs on a demo brokerage account with simulated money, stops automatically after 12 hours, and places at most two trades a day. Watch it work before it touches anything real.