Paper Trader

Fable — systematic momentum · one review each market day, every decision logged.

Portfolio Value
$104,203.37
as of 2026-10-12
Total Return
+4.20%
+$4,203.37 since inception
Latest Day
—
session P&L
Sessions
30
since 2026-08-24
Cash
$2,604.99
uninvested buffer
Overviewmandate, equity, what it holds
Mandate

A deterministic engine ranks a fixed 43-name universe (mega-caps, sector ETFs, gold) by blended 12-1/6/3-month momentum behind a 200-day trend filter, holds the top 8 at ~12% each with weekly rebalances and enter-8/exit-14 hysteresis, and parks everything uninvested in T-bills. The daily LLM session executes the precomputed plan with veto-only discretion — it may refuse a buy on catastrophic news or broken data, never invent or resize one. The design bet: LLM churn destroys alpha, so judgment was spent once, in the design, and discipline is enforced daily.

How it was decided

Designed end-to-end by Claude Fable 5 on 2026-08-21 under Eric's open grant — any legal strategy, full autonomy, no approval needed — as a strategy experiment and a buddycheck against the discretionary book. The evaluation was pre-committed before the first trade: judge at 60+ sessions on total return plus Sharpe and max drawdown, versus main and the S&P 500.

Equity Curve · from $100,000
Current Holdings · 9
SymbolMarket ValueUnrealized
AMD $15,361 +30.74%
MU $12,866 +9.50%
SMH $12,809 +9.01%
AMAT $12,145 +3.36%
MRK $11,344 -3.46%
XLE $11,340 -0.00%
LLY $11,134 -5.24%
XBI $11,104 -5.50%
BIL $3,496 -0.11%
Riskmeasured after the session, never shown to the agent
Attribution · market vs decisions · 2026-10-12

How much of the return came from simply being exposed to the market, and how much from the agent's own choices. A single-factor split of the inception-to-date return — market = β × benchmark, decisions = whatever is left — measured on the daily close-to-close equity curve, never on these report snapshots.

Total return
+5.28%
inception to date
Market
+0.29%
β × benchmark
Decisions
+4.99%
the residual
Cash drag
−0.02%
2.5% in cash
The decision half is not yet distinguishable from luck. Alpha's t-statistic is 0.60, well short of the ~2 that would make it significant, so +4.99% of "decisions" is the honest arithmetic residual and not evidence of skill. Beta is well measured long before alpha is — that asymmetry is a property of the estimate, not a flaw in the book, and it takes a lot more than 33 sessions to resolve.
The regression behind the split
EstimateValueReading
Beta0.36 t = 1.00 · not yet distinguishable from zero
R²3.1% the benchmark explains little of this book's day-to-day
Alpha (annualised)+39.59% t = 0.60 · NOT significant
BenchmarkS&P 500 33 aligned sessions · 2026-08-13 → 2026-10-09
Today alone. The book moved −2.33% against a benchmark +0.59% — +0.21% of that was exposure and −2.54% was position selection. One day of a two-term decomposition is noise; it is here because the daily reading is what accumulates into the figures above.
FRTB Capital · standardised approach · 2026-10-12

What a bank would have to hold in capital against this exact book, under the Basel market-risk rules (BCBS d457 (MAR20-MAR23, MAR32-MAR33)). Recomputed every session from live positions; nothing here is visible to the agent while it trades.

Capital requirement
$47,462
SbM + DRC + RRAO
Of equity
45.5%
capital / portfolio value
Market-risk RWA
$593,279
capital × 12.5
Overstated. 4 name(s) (BIL, SMH, XBI, XLE) could not be classified from reference data and fell to the "other sector" bucket at a 70% risk weight — the regulation's own fail-closed rule. The number above is safe-direction wrong.
Components, buckets, and why DRC is the interesting line
ComponentChargeShare
SbM delta · high correlations bind $32,223 67.9%
SbM vega · no optionality in the book $0 0.0%
SbM curvature · linear payoff $0 0.0%
DRC — jump to default · LGD 100% $15,240 32.1%
RRAO · no exotic underlyings $0 0.0%
Total capital requirement$47,462100%
Equity buckets · MAR21.72
BucketRisk weightNamesNet valueKb
11 70% BIL, SMH, XBI, XLE $38,749 $27,124
8 50% AMAT, AMD, MU $40,372 $14,889
5 30% LLY, MRK $22,478 $5,463
Why the second-biggest line is the interesting one. The DRC is jump-to-default: the issuer disappears overnight. Equity has no recovery, so Basel sets loss-given-default at 100% and charges an unrated name 15% of its market value — 32% of the requirement here, and nothing about hedging, forecasting or diversification reduces it. That the high correlation scenario binds is itself a readout: a long-only book is dearest when correlations rise, a hedged one when they fall.
Value at Risk · historical simulation · 2026-10-12

The loss this book exceeds on the worst few days in a hundred, measured by replaying 499 days of its own constituents' returns through today's positions. Recomputed every session; nothing here is visible to the agent while it trades.

VaR 99% · 1 day
$4,851
worst 1 day in 100
Of equity
4.66%
loss / portfolio value
ES 97.5%
$5,033
average of the tail beyond
VaR 99% · 10 day
$12,714
regulatory horizon
Quantiles, component VaR, the backtest, and what VaR will not tell you
ConfidenceHistoricalOf equity ParametricGapTail pts
95% $2,853 2.74% $3,113 −8.4% 25
99% $4,851 4.66% $4,403 +10.2% 5
ES 97.5% $5,0334.83% the average of the tail, not its edge
Why the 10-day figure is measured, not scaled. The textbook shortcut multiplies the one-day VaR by √10, which assumes daily returns are independent. Here that gives $15,339 against a measured $12,714 — 20.6% too high. Overlapping windows cost nothing to compute and need no such assumption.
Component VaR · at 99% · sums to the total
NameMarket valueWeightComponentShare
AMD $15,361 15.1% $1,156 26.3%
MU $12,866 12.7% $1,138 25.8%
AMAT $12,138 11.9% $786 17.9%
SMH $12,809 12.6% $661 15.0%
XBI $11,104 10.9% $268 6.1%
LLY $11,134 11.0% $177 4.0%
MRK $11,344 11.2% $114 2.6%
XLE $11,338 11.2% $103 2.3%
BIL hedge $3,496 3.4% −$0 −0.0%
These are allocations, not a ranking. Each name's number is the dollars of VaR it is responsible for, and they sum exactly to the total — which is what lets one go negative: a position that offsets the rest of the book reduces total risk, and no ranking by position size would ever show it.
The part that makes it a measurement. Over 249 out-of-sample days the 99% threshold was breached 2 time(s) against 2.5 expected — Kupiec p = 0.747, verdict consistent with the model, Basel traffic light green. The traffic light only punishes too many breaches; Kupiec is the test that also catches a model being too conservative, which is why both are shown. Each day's threshold is fitted on the prior 250 days only, so it is never tested on data it has already seen.
Two different questions. Everything above shocks today's positions with market history. The account's own equity curve — 42 sessions, 19.5% annualised volatility — says what it has actually done: 95% VaR 2.33% · 99% VaR 2.73%. They diverge whenever the agent has been trading.
What VaR will not tell you. It names a threshold and says nothing about how far past it a bad day goes — make the tail twice as deep without moving the cutoff and VaR does not move at all. That blind spot is why Basel replaced it with Expected Shortfall for capital, and why ES is quoted beside it above.
Counterparty Risk · SA-CCR · 2026-10-12

Market risk is symmetric — the price moves, you win or you lose. Counterparty exposure is one-sided: the market value clipped at zero, so it exists only when a trade has moved in your favour and the other side still owes you something. A fully-paid share bought on an exchange has none of that. You paid cash, it cleared, it settled, and nobody owes you anything — so the charge below is zero because the trade is over, not because nothing was measured.

SA-CCR EAD
$0
α × (RC + PFE) · 0 derivative(s)
CVA capital
$0
spread risk, not default
Settlement in flight
$11,340
T+1
Custodial concentration
100%
one broker, everything
The derivation: SA-CCR terms, what was excluded, settlement, leverage and concentration
SA-CCR termValue
Derivative trades in the netting set 0
Replacement cost · max(V − C, 0) $0
Add-on aggregate · supervisory factors $0
PFE multiplier · floored at 5% 1.0000
PFE · multiplier × add-on $0
α · CRE52.1 1.4
Exposure at default$0
CVA capital · CRE55, BA-CVA reduced $0
Why · what never reaches SA-CCR
KindPositionsMarket valueReason
cash equity long 9 $101,572 fully-paid long share: bought for cash, centrally cleared and settled — no open obligation, so no counterparty exposure
Settlement is the one that is genuinely live. US equities settle T+1, so a fill from this session is a commitment against a leg that has not been delivered yet: $11,340 in flight, $11,445 past due, capital charge $0. Basel charges a delivery-versus-payment trade only once it is five business days overdue (CRE70), so in-flight exposure is disclosed at a zero charge rather than hidden at zero exposure.
Leverage · capability vs utilisation
ProvisionedMargin debitShort valueGrossNet
4× · shorting on · options L3 $0 $0 0.97× 0.97×
Fully funded. This is what makes the zeros above readable as a fact about today's positions rather than about the account's permissions: the book is provisioned for 4× margin, shorting and options, and is currently using 0.0% of the first and 0.0% of the second. The first short or written option moves the exposure above off zero on its own.
The counterparty that does exist. Alpaca (paper) holds $2,605 of cash and $101,572 of securities — 100% of the book, by construction. Not an SA-CCR figure and not a capital charge, but a counterparty report that omitted the only counterparty there is would be worse than no report at all.
What this is really measuring. Alpaca simulates the fills, so there is no real counterparty behind these positions. What is measured is the SHAPE of the book — the counterparty profile it would carry if the trades were real.
Sessionsevery decision, in the agent's own words
Latest Review · 2026-10-12
Trades
ActionSymbolAmount
BUY XLE $11,340
Agent Reasoning
Session — Weekly rebalance; 26/43 names pass trend filter. Actions — BOUGHT XLE $11,340.30 (rank 6, score +0.345, energy entering via 12-1m +51.8%/6m +17.3%/3m +16.8%). News check clean, no veto flags. Book (post-trade, approx weights on ~$104.2k equity): - AMD 14.7%, MU 12.4%, SMH 12.3%, MRK 10.9%, AMAT 11.6%, LLY 10.7%, XBI 10.7% - XLE ~10.9% (new) - BIL/cash: ~2.5% Watching — TXN (#7), XOM (#8), CVX (#9) close behind XLE for next rebalance; CAT (#10) and AAPL (#12) further back. No held names near exit threshold this week — AMAT, MRK, XBI, LLY all retain positive scores and remain above their 200-day trend lines per the signal board. No vetoes. Market backdrop benign (VIX 15.4, SPX flat, drawdown 4.7% vs 15% breaker) — no risk-desk interference triggered.
History · 8 most recent of 29 earlier sessions
2026-10-09 $105,058.32 — 1 trade
ActionSymbolAmount
CLOSE CAT entire position
Session — Daily check (exits only); 24/43 names pass trend filter. Actions — CLOSED CAT (full position, ~$11,454 mkt value): dropped off eligibility (200dma break / negative 12-1m), rank now "None." Proceeds (~11.5k) roll to BIL per mandate — no replacement buy in plan. Book (post-trade, approx.) — AMD 15.0%, MU 12.8%, SMH 12.4%, AMAT 11.8%, MRK 10.6%, LLY 10.4%, XBI 10.3%, BIL now ~14.2% of equity (cash sweep pending settlement). No new risk positions added today. Watching — XLE, TXN, XOM, CVX, XBI (rank 6-10) are next-in-line candidates on the signal board if next week's rebalance opens a slot; XBI sitting on the weaker end of current holdings (-7.6% unrealized) — worth a trend-filter check at next rebalance. Market backdrop calm: VIX 15.2, SPX +0.24% today, breaker unarmed-risk (drawdown 3.8% vs 15% threshold). No BUYs were on today's plan, so no vetoes to report.
2026-10-08 $106,210.56 — 0 trades

No trades this session — held all positions.

Session — Daily check (exits only); 25/43 names pass trend filter. Trade plan empty — no rebalance triggers today. Actions — None. All eight current holdings (MU, AMD, AMAT, SMH, MRK, CAT, XBI, LLY) remain on the eligible board with positive momentum scores — no trend-break exits required. No vetoes needed since no BUYs were planned. Book — Equity $106,230. Holdings: AMD 15.3%, MU 13.1%, SMH 12.5%, AMAT 11.7%, CAT 11.1%, MRK 10.3%, LLY 10.3%, XBI 10.1%, BIL 3.3%, cash 2.4%. Notable movers: AMD +38.5% unrealized, MU +18.7%, SMH +12.8%; laggards XBI -9.1%, LLY -6.8%, MRK -6.6% — none breaking trend yet. Watching — TXN (#6) and the energy cluster (XLE #8, XOM #9, CVX #10) are building momentum just outside current holdings — candidates if they overtake a held name at next weekly rebalance. AAPL (#13) and NVDA (#14) re-entering the eligible board but still well back in rank. Market: SPX -0.33%, NASDAQ -0.55% on the day, VIX 15.5 (calm, well below defensive/freeze thresholds). Drawdown from HWM 2.8%, breaker armed at 15% — no action triggered. Holding steady.
2026-10-07 $106,826.79 — 0 trades

No trades this session — held all positions.

Session — Daily check (exits only); 25/43 names pass trend filter. Trade plan empty — no rebalance triggered, no trend breaks in current book. Actions — None. All 8 held positions (MU, AMD, AMAT, SMH, MRK, CAT, XBI, LLY) remain in the top-14 ranked set with positive scores, so no forced exits. No vetoes needed since there were no planned BUYs. Book — MU 12.5%, AMD 15.4%, SMH 12.5%, AMAT 11.9%, CAT 11.3%, LLY 10.4%, MRK 10.4%, XBI 10.2%, BIL 3.3%, cash 2.3%. Equity $106,912 (+6.9% since inception). Unrealized gains concentrated in AMD (+$4.7k), SMH (+$1.6k), MU (+$1.6k); drawdowns in XBI (-$856), MRK (-$682), LLY (-$643) — all within normal rotation range, no trend-filter violations. Watching — TXN (#7) and XLE/XOM/CVX (energy cluster, #9-11) are building score momentum just outside the current book; NVDA (#12) re-entering strength. These are candidates for next weekly rebalance if ranks hold. Market backdrop: VIX 15.8 (up 5.4% today, still well below defensive/freeze thresholds), equities soft today (S&P -0.56%) but up on the week — no regime change, no risk-desk constraints triggered.
2026-10-06 $108,646.67 — 0 trades

No trades this session — held all positions.

All positions remain above their 200-day trend filters per today's eligibility board (all current holdings — MU, AMD, AMAT, SMH, MRK, CAT, XBI, LLY — are in the 27/43 eligible list). Trade plan is empty, so no action required. Session — Daily check (exits only); eligibility breadth 27/43. Actions — None. Trade plan was empty; no trend-break exits triggered. No BUYs to veto-check. Book — Top holdings by weight: AMD 15.0%, MU 12.7%, SMH 12.6%, AMAT 12.0%, CAT 11.5%, XBI 10.3%, LLY 10.1%, MRK 10.1%. BIL 3.2%, cash 2.3%. Equity $108,689 (+8.7% since inception, drawdown 0.6% from HWM). Watching — TXN (+0.382, rank 6) and XLE (+0.283, rank 9) are the nearest unheld names knocking on the door for next rebalance if any current holding's score decays. NVDA (rank 11) continues to strengthen on 3m momentum but not yet rank-qualified. Market backdrop calm (VIX 15.4, regime well below defensive threshold), no overrides warranted even if it were.
2026-10-05 $107,510.85 — 0 trades

No trades this session — held all positions.

Session — Weekly rebalance; eligibility breadth 24/43. Signal engine returned an empty trade plan — no closes, no buys. Holdings and scores remain stable versus last rebalance (MU, AMD, AMAT, SMH, MRK, XBI, CAT, LLY all still rank in top 12 with positive scores above the trend filter). Actions — None. No vetoes needed since no BUYs were planned. Book (equity $107,589.57): - MU 12.8% | AMD 15.1% | SMH 12.6% | AMAT 12.2% | MRK 10.2% | XBI 10.4% | CAT 11.1% | LLY 10.1% - BIL 3.2%, cash 2.3% All eight current holdings remain the top-ranked eligible names; weights are near target and within the 20% position cap, so hysteresis keeps the book unchanged this week. Watching — TXN (#7, +0.354) and the energy complex (XLE #9, XOM #10, CVX #12) are the nearest challengers outside the current book; a continued move in oil/energy trend could put XLE/XOM into contention for LLY or CAT at next rebalance if their 3m momentum keeps improving. NVDA (#13) and AAPL (#14) are further back but climbing — watching for trend-filter confirmation. CAT's -12% 3m reading and LLY's -6.2% 3m are the weakest legs in the current book; no action today since score/rank still support holding, but these are the most likely rotation candidates if momentum continues to decay. Market backdrop: VIX 16.0 (up 4.6% today, still well below defensive/freeze thresholds), S&P +0.17%, Nasdaq +0.69%, Dow -0.55%. Drawdown from HWM 1.5%, breaker not triggered. No risk-desk constraints bound today given zero order flow.
2026-10-02 $109,231.51 — 0 trades

No trades this session — held all positions.

All positions remain above trend/eligibility (all 8 holdings — MU, AMD, AMAT, SMH, MRK, XBI, CAT, LLY — appear in the current top-14 eligible board), and the trade plan is empty. No trend-break exits are triggered. No action taken. Session — Daily check (exits only); eligibility breadth 25/43. Trade plan empty — no rebalance today. Actions — None. No trend-break exits triggered; all 8 current holdings remain on the eligible/signal board. No vetoes needed (no BUYs planned). Book — Equity $109,227. Holdings: AMD 15.0%, MU 13.0%, SMH 12.5%, AMAT 12.2%, CAT 11.4%, MRK 10.2%, XBI 10.3%, LLY 9.9%, BIL 3.2%, cash 2.3%. Broad-based gains led by AMD (+39.3%), MU (+20.9%), SMH (+16.4%), AMAT (+13.9%); LLY, MRK, XBI modestly underwater but intact. Watching — TXN (#6, +0.376) and XOM/XLE/CVX (energy cluster, #9-11) are climbing the board and could pressure CAT/LLY/XBI/MRK for slots at next weekly rebalance given their weaker relative scores. NVDA (#13) re-entering eligibility is also worth noting as a potential future candidate. Market backdrop constructive (SPX +1.08%, VIX 15.6, no regime stress).
2026-10-01 $106,474.43 — 0 trades

No trades this session — held all positions.

Daily check complete — no trade plan today, positions reviewed and all intact, no trend-break exits triggered. Session — Daily check (exits only); eligibility breadth 25/43. Trade plan was empty — no action required. Actions — None. No exits or entries scheduled today; no vetoes needed since no BUYs were planned. Book — Equity $106,528. Top holdings: AMD 14.7% (+33.3% unrl.), MU 12.8% (+15.8%), SMH 12.4% (+12.2%), AMAT 12.1% (+9.9%), MRK 10.6% (-4.3%), XBI 10.7% (-2.8%), CAT 11.0% (-0.7%), LLY 10.2% (-7.9%). BIL 3.3% + cash 2.3% (~5.6% dry powder). Watching — TXN (#7, +0.331) and XOM/XLE (#10/#11, energy showing strength on 3m) are closest unheld names knocking on the door for next weekly rebalance. LLY is the weakest current holding (-7.9%) — worth watching for a trend-filter break if weakness persists, though no action is warranted today. Market backdrop benign: VIX 16.5, regime well below defensive/freeze thresholds, drawdown from HWM only 0.5%.
2026-09-30 $107,046.95 — 0 trades

No trades this session — held all positions.

Session — Daily check (exits only); eligibility breadth 26/43. Trade plan was empty — no trend-break exits triggered. Actions — None. Plan called for no trades today; all 8 held names (MU, AMD, AMAT, MRK, SMH, XBI, LLY, CAT) remain above their trend filters per the signal board, so no risk-management exits were required. No vetoes needed since no BUYs were planned. Book — Equity $107,094.52. Holdings roughly balanced: AMD 14.5%, MU 13.0%, SMH 12.3%, AMAT 11.8%, CAT 11.1%, MRK 10.7%, LLY 10.4%, XBI 10.6%. BIL 3.3%, cash $2,500 (2.3%). Market: SPX +0.27%, NASDAQ +0.53%, VIX 15.9 (calm, well below defensive/freeze thresholds). Watching — TXN (#9), XLE (#10), XOM (#11) sit just below current holdings on the score board and are candidates to rotate in at next weekly rebalance if any held name's score deteriorates further. LLY, MRK, XBI are the weakest-performing current holdings (small unrealized losses) but remain trend-qualified — no action until rebalance signal.
Full session log · every session →