I Built a Trading Strategy at 18 That Backtests at 163% CAGR — and Published It on SSRN

--

It’s called MARS — the Multi-Asset Rotation System. Here’s exactly how it works.

I work in a factory. I’m 18. I start university in September.

In between I built, backtested, and published a quantitative
trading strategy. I named it MARS — the Multi-Asset Rotation
System, also known as the Wardell Rotation.

The 10-year backtest shows 163% CAGR, a Sharpe ratio of 1.78,
and a max drawdown of 32.9%. It runs fully automated on a £5
per month server. I get a Telegram signal once a week. I make
one trade roughly every three months.

This is how it works.

— -

The problem with most retail strategies

Most people who invest do one of three things. They buy and hold
forever and watch their portfolio drop 60% in a crash with no
plan. They try to time the market and fail because nobody can
predict the future consistently. Or they do nothing because it
all feels too complicated.

MARS sits in the middle. It does not try to predict the future.
It reads a handful of signals every Sunday evening and decides
which of four assets is the most appropriate place to be right
now. Then it holds that asset until the signal changes.

That’s it. One decision per month at most.

— -

The four regimes

The core idea of MARS is that markets exist in one of four
states at any given time. Most strategies only recognise two —
risk on or risk off. MARS has four, which gives it an
intermediate buffer that most systems miss entirely.

BULL (score above 0.70) — Buy Bitcoin. When conditions are
genuinely strong, BTC historically returns anywhere from
60% to over 1000% in confirmed bull years. Nothing beats
it when the signals are right.

NEUTRAL (score 0.55 to 0.70) — Buy CSPX, the iShares S&P 500
ETF on the London Stock Exchange. Mixed signals, not clearly
bullish or bearish. Steady compounding while waiting for
clearer direction.

WARNING (score 0.40 to 0.55) — Buy SGLN, the iShares
Physical Gold ETC. Signs of trouble before a bear is
confirmed. Gold provides a hedge with low equity correlation.
This is the regime most strategies miss — the one that
protected capital in late 2021 before the 2022 crash.

BEAR (score below 0.40) — Buy IGLS, the iShares UK Gilts
0–5yr ETF. Bear confirmed. Sovereign bonds preserve capital
while generating a small coupon. Hold until the signal
recovers.

— -

How the signal works

Every Sunday at 8pm a Python bot running on a VPS calculates
a composite score C between 0 and 1.

C = T x 0.50 + N x 0.20 + M x 0.20 + E x 0.10

T is the technical signal — SPY vs its 200-day moving
average, 1-month momentum, and 30-day realised volatility.
This gets 50% of the weight because price action is the most
reliable indicator over time.

N is news sentiment — the bot scans 100 financial headlines
from the past 7 days via NewsAPI and scores them for bullish
vs bearish language. 20% weight.

M is the macro signal — it pulls the US unemployment rate
directly from the Federal Reserve FRED database. Rising
unemployment is a leading indicator of economic stress. 20%
weight.

E is an earnings component defaulting to 0.5–10% weight.

The score determines the regime. The regime determines the
asset. The bot sends me a Telegram message with the result.

— -

The backtest results

I ran this against real documented annual returns for each
asset class from January 2016 to April 2025. That is 112
months of data. The backtest uses real IBKR fee schedules,
realistic bid-ask spreads, and T+1 settlement simulation.
Starting capital was £1,500 with £500 per month deposits
for the first six months — £4,500 total deposited.

CAGR: 163.1% per year
Sharpe ratio: 1.78
Sortino ratio: 3.41
Calmar ratio: 4.95
Profit factor: 7.57x
Monthly win rate: 69.6%
Max drawdown: 32.9%
Total rotations: 43 over 10 years

For context — the S&P 500 averages roughly 10% per year.
Warren Buffett’s lifetime CAGR is approximately 20%. A
Sharpe ratio above 1.0 is considered good by professional
standards. 1.78 is exceptional.

Year by year the strategy correctly navigated the 2018
crypto crash, the 2020 COVID crash, the 2022 bear market,
and the 2025 tariff shock — rotating to gold or gilts
before each major fall.

— -

Why I published it

I published the full methodology on SSRN — the Social
Science Research Network — on 26 April 2026. SSRN Abstract
ID 6653818. Author: Ronnie Wardell.

You cannot patent a trading strategy. But you can establish
priority of authorship through public timestamped
publication. That is what I did. The methodology, the
backtest data, the composite signal formula, and the formal
claims are all in the paper.

The strategy is open. Anyone can read it and use it. The
name Wardell Rotation and the acronym MARS are mine.

— -

How to use it

The strategy is designed for a standard retail cash account
on Interactive Brokers. No derivatives. No margin. No
options. Just four ETFs and Bitcoin, rotating based on a
signal you receive once a week.

If you want to run it manually — read the Sunday signal,
sell your current holding on Monday morning, buy the new
asset on Tuesday after settlement clears.

If you want to run it automated — the bot code is on
GitHub at wardell-rotation.

I am currently running it in paper trading mode on IBKR
with the first live deposit going in this week.

— -

The full paper is available free at SSRN Abstract ID 6653818.

I am documenting the live results publicly. Follow along
at @ronxbt on X.

Not financial advice. Past performance does not guarantee
future results.

_TAGS_
Investing Finance Bitcoin Cryptocurrency Passive Income