Quantitative Research · Systematic Trading · Est. 2024

We turn research
into repeatable edge.

Phaedo Technologies is a quantitative research and systematic trading firm. We engineer alpha from first principles — top-down macro to bottom-up signals — and deploy it through fully automated execution and risk control.

DisciplineQuantitative · Systematic
UniverseMulti-asset · 9 markets
BaseVancouver, BC
Scroll
Φ
Our Philosophy · The Second Sailing
I thought I must take refuge in reasoned argument, and seek in them the truth of things.
Socrates · Plato's Phaedo · c. 380 BC

In the Phaedo, Socrates gives up the hope of reading the cause of things directly from the world. To look at reality head-on, he warns, risks blinding the mind — as the eye is blinded staring at an eclipse. He turns instead to what he calls a second sailing: he lays down the strongest hypothesis he can find, holds to whatever agrees with it, and discards whatever does not. The course is slower and more demanding — but it is the safer one, and the only one that reaches the truth.

Twenty-four centuries later, this remains the discipline of quantitative research. Markets cannot be read directly; price is flux and noise, and staring at it only blinds the judgment. So we proceed as Socrates did — by hypothesis. We posit the strongest idea we can defend, test it against the evidence across markets, keep what the data confirms, and reject what it does not. Hypothesis, test, keep or discard. It is the method behind every factor we research and every strategy we trade.

Phaedo Technologies takes its name from that method — the second sailing.

The slower, safer voyage toward what is true.
Our Approach

Top-down to the trade, signal by signal.

We narrow from the macro landscape to a single executable signal through a disciplined, repeatable funnel. Every layer constrains the next — context first, conviction last.

01 / MACRO

Macro Analysis

We start with the regime: liquidity, rates, the dollar, growth and inflation cycles. The macro read sets risk appetite and tells us which assets and which direction deserve attention.

02 / SECTOR

Sector Selection

Within the regime, we rank sectors and asset groups by relative strength, flows, and sensitivity to the prevailing macro drivers — concentrating where the tailwind is strongest.

03 / SECURITY

Company & Instrument Selection

We then isolate the specific instruments — equities, indices, FX, commodities, or crypto — that best express the thesis, screening for liquidity, cleanliness of data, and tradability.

04 / ALPHA

Alpha Discovery & Validation

Finally we engineer the signal. Each candidate alpha is formalized, tested across our universe, and put through three layers of Monte-Carlo validation — noise baseline, bootstrap confidence, and execution perturbation — before any capital is committed.

Selected Alphas

A library built from research, not folklore.

A sample of factors from our research program — formalized, normalized across heterogeneous markets, and validated out-of-sample. Notation is illustrative.

Cross-SectionalMomentum9-Asset Universe

Cross-Asset Momentum Persistence

Ranks each instrument by its trailing risk-adjusted trend and tilts toward the leaders. The core insight: relative strength persists across heterogeneous markets once returns are normalized to a common volatility scale.

We replace fragile cross-sectional ranks (small universe) with a time-series percentile, then confirm with a bootstrap consensus filter — only trading when both systems agree.

Factor · ALPHA-014
signal = ts_rank( return / vol_60 , 252 )
trade  = consensus( ts_system , cs_bootstrap )
Type
Trend
Horizon
Days
Markets
All 9
Mean-ReversionIntraday Range

Normalized Range Reversion

Measures where price closes within its own true range and fades the extremes. Closes pinned to the high or low of an ATR-normalized range tend to revert when not confirmed by volume.

Normalizing by ATR makes the signal comparable across a 10× volatility spread — from EURUSD to ETHUSD — so one asset never dominates the book.

Factor · ALPHA-032
pos    = (close - low) / ATR_20
signal = -1 * zscore( pos )   // fade extremes
Type
Reversion
Horizon
Intraday
Filter
Volume
Volume–PriceDivergence

Volume-Confirmed Divergence

Compares price displacement against the volume that produced it. Moves on thin volume fade; moves on heavy volume extend. The factor scores the agreement between a price delta and its volume delta.

Built on real MT5 tick volume as a liquidity proxy, with a normalization pipeline that keeps the signal stable across instruments and sessions.

Factor · ALPHA-061
signal = sign( Δclose ) *
         rank( Δvolume / volume_60 )
Type
Flow
Horizon
Hours
Input
Tick vol
Gold · XAUUSDSessionReversion

Gold Session Reversion

A gold-specific factor that maps XAUUSD's intraday behaviour by trading session. Overextended moves into the London–New York overlap, unconfirmed by range expansion, show a measurable tendency to mean-revert into the close.

Gold's sensitivity to the dollar and real yields makes its session structure unusually stable — a property we exploit with a normalized, risk-bounded reversion signal.

Factor · ALPHA-XAU
ext    = (close - vwap) / ATR_20
signal = -1 * ext * session_weight
risk   = bounded( |signal| , max_exposure )
Asset
XAUUSD
Horizon
Session
Edge
Reversion
Performance · 2026 YTD

Systematic execution, measured honestly.

Results from one of our automated XAUUSD strategies, January–May 2026. Figures are shown in return and risk terms; capital base is omitted.

Total Return
+501%
Jan – May 2026
Max Drawdown (equity)
−24.9%
Peak-to-trough, open equity
Profit Factor
9.9
Gross profit / gross loss
Win Rate
91.1%
216 / 237 trades
Sharpe (ann.)
7.5
Balance-based estimate
Total Trades
237
Avg hold 6m 40s
Cumulative Return — XAUUSD Strategy% · 2026.01 – 2026.05

Past performance is not indicative of future results. Figures reflect a historical simulation of a single systematic strategy on XAUUSD and include the elevated open-equity drawdown characteristic of grid-based execution. Returns shown are on the strategy's allocated capital.

Capabilities

What we build.

The research and engineering stack behind our own trading — and the foundation for select collaborations.

01

Factor & Signal Research

End-to-end alpha discovery: hypothesis design, formalization, cross-asset normalization, and out-of-sample validation. We license access to selected, validated factor research.

02

Strategy & Model Development

Translating research into production systematic strategies — feature engineering, statistical and ML modeling, portfolio construction, and signal blending.

03

Risk Management Systems

Position sizing, drawdown and exposure controls, Monte-Carlo stress testing, and risk frameworks designed to sit above every strategy.

04

Research Infrastructure

Reusable operator libraries, event-driven backtesting engines, and reproducible Monte-Carlo validation pipelines built for heterogeneous, multi-asset universes.

Contact

Research, collaboration & inquiries.

kylingavin@outlook.com
Founder
Yaowei (Gavin) Yu
Location
Vancouver, British Columbia
Canada

Phaedo Technologies is a quantitative research and systematic trading firm. Information on this site is provided for general informational purposes only and does not constitute investment advice, a recommendation, or an offer or solicitation of any security or investment service.