01 FIAM × McGill Asset Management Hackathon 2026Draft

Sell, sell, sell.

And buy the right 230. A market-neutral book that earns from picking stocks, not from where the market goes.

TM

Team member 1

Portfolio manager

TM

Team member 2

Quant research

TM

Team member 3

Risk

TM

Team member 4

Execution & data

See the live book →Key results

02

Results

Market-neutral stock selection that beat T-bill + 4%

Before and after costs, Jan 2021 – Aug 2026. Returns come from picking stocks, not from the market.

Information ratio

0.69

0.50 net of costs

CAGR

13.1%

11.4% net · vs 7.5% hurdle

Beta to S&P 500

−0.03

s.e. 0.07

Worst month

−2.9%

−3.0% net · S&P fell 23.9%

Thesis 1Rank, don't predict the market

Each month we rank ~3,000 U.S. stocks by expected return beyond what their market exposure delivers.

Thesis 2Five forecasters, one score

Ridge, LightGBM, a pre-trained neural net and two SEC 8-K filing signals make different mistakes; averaging cancels noise.

Thesis 3Matched on risk

Longs and shorts are paired inside 20 projected-beta groups, so the book carries almost no market risk.

Market sell-offs · monthly backtest, Jan 2021 – Aug 2026

03

Process

Investment process: from signal to execution

Every layer has a rationale grounded in published research. The signal fires at month-end, the book is rebuilt, and orders go out before the close.

Data→

Five forecasters→

Ensemble score→

Neutralize→

Match on risk→

Execute

04

Live P&L

Live P&L and top holdings

The current book, marked to market every 5 seconds during the session.

Connecting468 positions as of 2026-08-20 · refreshed every 5 seconds

--:--:--

Portfolio value

$—

Fetching live quotes…

Waiting for intraday data…

Book, % of NAV S&P 500 (SPY)

vs S&P 500 today

—

SPY —

Winners / priced

—

Long leg · 236

—

Short leg · 232

—

Gross exposure

193.2%

$1.603B long · $1.944B short

Net exposure

−18.6%

Beta-matched, not dollar-matched

Top contributorsP&L · move

Top detractorsP&L · move

Quotes from Yahoo Finance, may be delayed. Value = book NAV ($1.836B) + today's P&L on current positions.

05

Arcana

Arcana in two minutes

The key points from the Arcana US risk model, 1 Jan 2021 – 31 Aug 2026. Use ← → to flip through.

Summary table1 / 11

Risk snapshot at the end of the window

Forward-looking predictions from Arcana's risk model for the book held on 31 Aug 2026. VaR is a percent of gross.

Arcana value	31 Aug 2026
Predicted volatility	3.61% / yr
Daily VaR (95% / 99%)	0.37% / 0.53%
Annual VaR (95%)	5.94%
Predicted beta	−0.022
Stock-specific share of risk	24.1%
Effective independent bets	342
Positions	467 (235 / 232)
Gross	$3,551M
Net, share of gross	−9.1%

Value at risk on $3,551M gross

0.37%

Daily VaR, 95% ≈ $13M

Where predicted risk comes from

Factor 75.9%Stock-specific 24.1%

3.61%

Predicted vol / yr

−0.022

Predicted beta

342

Effective bets

Arcana, out_run_sec2, Arcana US risk model, GMV basis. Portfolio views › Summary table, as of 31 Aug 2026.

Exhibit 1 · Idiosyncratic vs factor2 / 11

82% of the return came from factor exposure

Most of the return is factor exposure, mainly style. The residual — what factors cannot explain — added +5.01%.

Arcana value	Full window
Total return	+27.31% ($563.5M)
Factor return	+22.30% ($479.5M)
Residual return	+5.01% ($84.0M)
Residual IR	0.29
Factor Sharpe	1.14
Residual hit rate	43.1%
Residual win / loss	1.39

Total return +27.31%

82%

18%

Factor return

+22.30%

Residual return

+5.01%

1.14

Factor Sharpe

0.29

Residual IR

1.39

Residual win / loss

Residual is what the Arcana US factors cannot explain.

Arcana, out_run_sec2, Arcana US risk model, historical mode, returns on gross market value. Residual vs. Factor, 1 Jan 2021 – 31 Aug 2026.

Exhibit 2 · Decomposition3 / 11

A short mid-cap tilt earned the most

Style factors explain +23.11%, led by mid cap (+12.25%) and momentum (+6.36%). Components add to the +27.31% total.

Source	Return
Style factors	+23.11%
Mid cap	+12.25%
Momentum	+6.36%
Earnings yield	+4.20%
Size	+3.90%
Volatility	−3.45%
Industry	+2.42%
Beta factors	−3.23%
Residual	+5.01%

Build-up to +27.31%

Style factors

+23.11%

Industry

+2.42%

Beta factors

−3.23%

Residual

+5.01%

Total

+27.31%

Mid cap and size together add +16.2%: the book was short mid caps throughout (loading −0.09 to −0.32).

Arcana, out_run_sec2, Arcana US risk model, historical. Return Decomp › Full detail, 1 Jan 2021 – 31 Aug 2026.

Exhibit 3 · Long vs short spread4 / 11

Both legs add stock-specific return

The long leg carries two-thirds of the residual (+3.4% vs +1.6%). Shorts are hedges with some skill, not the main source of alpha.

Arcana value	Full window
Long P&L	+59.67% ($1,266.5M)
Short P&L	−32.36% (−$703.0M)
Total	+27.31%
Long residual	+3.40%
Short residual	+1.60%
Names that made money	59% L / 38% S

Contribution by leg

Long P&L$1,266.5M

+59.67%

Short P&L−$703.0M

−32.36%

Total$563.5M

+27.31%

59%

Long names that made money

38%

Short names that made money

Shorts lose money in a rising market by design. After removing factors, both legs are positive.

Arcana, out_run_sec2, Arcana US risk model, historical. Name counts from Arcana asset contribution (830 longs, 841 shorts).

Exhibit 4 · Concentration5 / 11

Risk is spread across hundreds of names

With 467 positions and 342 effective bets, no small group of names drives the book's risk.

Arcana value	31 Aug 2026
Top 10, share of total risk	16.1%
Top 10, share of stock-specific risk	13.3%
Top 10, share of skill risk	24.7%
Effective independent bets	342
Long / short	176 / 166
Largest position	0.39% of gross

Top 10 names, share of risk

16.1%

of total risk sits in the top 10

Largest position: 0.39% of gross

467 positions■ 342 effective independent bets (176 L / 166 S)

Arcana, out_run_sec2, Arcana US risk model. Concentration Level (Gross, top 10), book held on 31 Aug 2026.

Exhibit 5 · Earnings P&L6 / 11

Almost all return is earned outside earnings

This is not an event-driven strategy: 98% of P&L comes from outside earnings season.

Arcana value	P&L, $M
Outside earnings season	+559.7
Earnings season (total less outside)	+9.9
Earnings days	−9.0
±5 days around earnings	+18.9
Total (model-agnostic)	+569.5

P&L by calendar, $M

Outside earnings season+559.7M

Earnings season+9.9M

98%

of P&L earned outside earnings season (total $569.5M)

Not an event-driven strategy. Model-agnostic total differs slightly from the risk-model total ($563.5M).

Arcana asset contribution, Earnings Season tab, historical, 1 Jan 2021 – 31 Aug 2026.

Exhibit 6 · Style exposures7 / 11

Long momentum, short mid caps, all the way

Momentum, mid-cap and size loadings never changed sign in 68 months: persistent tilts, not timing.

Loading (z)	31 Aug 26	Range
Momentum	+0.38	+0.12 to +0.55
Mid cap	−0.14	−0.32 to −0.09
Size	−0.08	−0.55 to −0.05
Volatility	+0.14	+0.09 to +0.41
Dividend yield	−0.15	−0.22 to −0.06
Earnings yield	+0.14	0.00 to +0.28
Value	−0.08	−0.14 to +0.07
Growth	+0.06	−0.12 to +0.16

Loading (z), 31 Aug 2026 vs 68-month range

MomentumPERSISTENT●

Mid capPERSISTENT●

SizePERSISTENT●

VolatilityPERSISTENT●

Dividend yieldPERSISTENT●

Earnings yield

Value

Growth

−0.60+0.6

Bar = range over 68 months, dot = latest. Momentum, mid cap and size never changed sign: persistent tilts, not timing. All 11 GICS sectors held; largest sector net is Industrials, −$69M (1.9% of gross).

Arcana exposure decomposition, Style mode, month-end loadings, Arcana US risk model, 1 Jan 2021 – 31 Aug 2026.

Exhibit 7 · Drawdown8 / 11

Worst drawdown −3.7%; the S&P fell 24.5%

Our daily recomputation gives −3.66%, consistent with Arcana's −3.70%.

Value	Full window
Max drawdown (Arcana)	−3.70%
Peak	13 Oct 2022
Trough	4 May 2023
S&P 500 max drawdown	−24.5%
S&P trough	12 Oct 2022

Maximum drawdownCompare S&P 500

−3.70%

HoneX book

−24.50%

S&P 500

Jan 2021Aug 2026

S&P trough 12 Oct 2022
Book peak 13 Oct 2022
Book trough 4 May 2023

The book's worst stretch began the day after the market bottomed, as stocks recovered.

Arcana performance summary (max drawdown); peak, trough and SPY drawdown from Arcana daily total returns, compounded.

Exhibit 8 · Sharpe ratio9 / 11

Sharpe 1.04 overall, 0.29 on stock-specific return

Arcana's Sharpe uses its own risk-free convention; the headline information ratio vs T-bill + 4% is computed separately.

Arcana value	Full window
Sharpe ratio	1.04
Sortino ratio	1.61
Factor Sharpe	1.14
Residual IR	0.29
Residual Sortino	0.47
Realized volatility	4.21%
Factor volatility	3.06%
Residual volatility	2.80%

Annualized, full window

1.04

Sharpe

1.61

Sortino

1.14

Factor Sharpe

0.29

Residual IR

0.47

Residual Sortino

Factor Sharpe above the total tells the same story as exhibit 1: most risk-adjusted return is factor exposure.

Arcana, out_run_sec2, Arcana US risk model, historical, annualized. Summary statistics, 1 Jan 2021 – 31 Aug 2026.

Exhibit 9 · Hit rate10 / 11

Right half the time, with bigger wins

A win/loss ratio of 1.33 means the book only needs to be right 42.9% of days. Drag the hit rate to see the edge.

Arcana value	Full window
Names that made money	49% (1,092 / 2,220)
Daily hit rate	49.5%
Win / loss ratio	1.33
Residual hit rate	43.1%
Residual win / loss	1.39
Selection P&L	+30.6%
Sizing P&L	−3.3%
Residual: selection / sizing	−17.9% / +22.9%

Hit rate × win/loss (1.33)break-even 42.9%

35%actual 49.5%60%

49.5%

Daily hit rate

+0.153

Edge per unit risked

Selection vs sizing P&L

Selection

+30.6%

Sizing

−3.3%

1,092 of 2,220 names made money (49%). On stock-specific return, sizing (+22.9%) did the work while picking alone lost (−17.9%).

Arcana performance summary and asset contribution, historical, 1 Jan 2021 – 31 Aug 2026.

Exhibit 10 · Turnover11 / 11

About 2.2× gross traded per year

Within the 100–400% range the guideline gives for active quant equity.

Arcana value	Full window
Turnover, annualized	2.19×
Traded per year	$5,721M
Average gross	$2,605M
Holding period (approx.)	~115 trading days

Turnover vs the 100–400% guideline band

100%

400%

0%500%

2.20×

Annualized turnover

$5,721M

Traded per year

~115d

Holding period

What if average gross were $2,605M?

$1,500M$4,000M

Holding period approximated as average gross ÷ annual traded value (0.46 years ≈ 115 trading days).

Arcana performance summary, historical, 1 Jan 2021 – 31 Aug 2026. Portfolio Stats › Turnover.

Use ← → keys, swipe, or the numbers to move between slides.

06

SEC data

SEC filings: the data that lifted the result

The biggest change in this version: the same model, now reading SEC filings. Left, before and after; right, our next idea in healthcare.

Before → after

Adding SEC filing signals

Without SECWith SECT-bill + 4%

Information ratio, net

0.15→0.46

Information ratio, gross

0.35→0.63

CAGR, net

8.3%→11.1%

Sharpe ratio

0.73→0.96

Alpha vs S&P 500 / yr

4.8%→7.7%

Max drawdown

−8.1%→−6.7%

Same model, costs and rules; only the SEC filing signals change. Net of costs, Jan 2021 – Aug 2026.

🩺

Next ideaDraft

Healthcare idea (coming soon)

Data and results to be added.

Data: To be added
Signal: To be added
Result: To be added

07

Risk

Risk management process

Concentration, market cap, illiquid positions, drawdown and ratios, plus the tests that show the returns are not luck.

No single name, or small group of names, can sink the book.

444–495

Names held

Mandate 200–500, every day

0.43%

Avg position

Max 1.7% at rebalance

4.4%

Gross in top 10

Average rebalance

342

Effective bets

Arcana, 31 Aug 2026

Near-equal weights (~0.43% of NAV) instead of an optimizer, so no estimate error concentrates the book.
Longs and shorts are spread across 20 risk groups, and no single name may exceed 4.9% of NAV.
Current book: 468 names, largest position 0.84% of NAV.

08

Agentic

Agentic workflow for factor-risk monitoring

An AI agent audits the pipeline and monitors exposures. It flags issues for the team and never produces forecasts, positions or trades.

Risk agent scan

Current book + latest Arcana exposures · flags go to the team, the agent never trades

·PositionsMandate 200–500

·Gross exposureMandate ≤ 200%

·Net exposureMandate ±50%

·Largest positionCap 4.9% · avg 0.43%

·Predicted betaTarget ≈ 0 (Arcana)

·Momentum loading68-mo range +0.12 to +0.55

·Mid-cap loading68-mo range −0.32 to −0.09

·Stock-specific risk shareFactor risk dominates

Press Run scan to check the book.

Quant research development

Agentic workflow

1Team idea
2Agent tests it on dev data
3Written rule + code
4Placebo backtest
5Independent audit

What it caught

Exit-date bug8-K survivorship biasLook-ahead leaksCost model

09

ExecutionDraft

Execution: spread, flow and channel

How the month-end signal becomes orders, what each trade costs, and where it routes.

Month-end data

8-K filings count only if accepted before 15:30 New York time on the last trading day, leaving time to trade before the close.

Broker

Prime broker (TBD)

Algorithm

VWAP / close auction (TBD)

Venues

NYSE · Nasdaq · Cboe (TBD)

Short locates

Prime broker stock loan (TBD)

Bid-offer cost calculator

Our cost model, applied to every net figure (one way, per dollar traded).

Trade size $10M

20bp

One-way cost

Institutional estimate

$20,000

Cost of this trade

0.50%

Borrow / year

If shorted

All groups capped at 300bp. Spreads from the Corwin–Schultz high-low estimate, used for micro and nano caps only.

10

Next stepsDraft

Next steps

Where the strategy goes from here.

New markets

CanadaTSX
Same time zone and similar accounting data: the easiest first step.
EuropeSTOXX 600 exchanges
Deep universe for a second, lowly correlated book.
JapanTSE Prime
Large cross-section and governance-reform tailwinds.

Deployment for pensions

A market-neutral sleeve with beta ≈ 0 diversifies a pension's equity-heavy book.
Near-equal weights and a 230 + 230 structure keep capacity and liquidity transparent.
Every rule is pre-registered and auditable: a single program rebuilds the book from raw data.

For the portfolio manager

Agents as analysts: quote-verified 8-K events reviewed against each month's strongest picks.
Short-squeeze control: a crowding-aware short screen aimed at the 2024–2025 losses.
Stage-two alternative data plugs into the same leakage canary and placebo tests.

11

Q&A

Thank you for your time.

Questions?

Fidelity Investments (FIAM)McGill Desautels Faculty of ManagementCarbon ArcArcana

Thank you to our sponsors