LP Analyses
—Cumulative Performance
P&L Δ — Daily Portfolio P&L & Top / Bottom Contributors
Total
—
High (single day)
—
Low (single day)
—
Avg
—
Win days
—
Range
—
Top Contributors
Bottom Contributors
Alpha & Beta return — XBI
Alpha return
Total
—
High (cum)
—
Low (cum)
—
Period XBI
—
Top α (beat XBI)
Bottom α (lagged XBI)
Beta return
Total
—
High (cum)
—
Low (cum)
—
Avg β (weighted)
—
Top β contributors
Bottom β contributors
Performance Summary
Monthly Returns — Portfolio vs XBI
Rolling Sharpe (rf=0)
Rolling Sortino (rf=0)
Rolling Daily Return Skew
Rolling Excess Kurtosis
Intra-Portfolio Correlation (26-week rolling)
Gross & Net Exposure Over Time
β-Adj Exposure (XBI) ?β-Adjusted Exposure = Σ (position weight × per-stock XBI beta).
Each name's contribution to total market exposure scales by its sensitivity to XBI — a 10% position in a β=1.5 name contributes 15% of β-adj exposure, while a 10% position in a β=0.5 name contributes 5%.
For a long-only biotech book with median β≈1.15, expect β-adj gross around 100–130%.
Each name's contribution to total market exposure scales by its sensitivity to XBI — a 10% position in a β=1.5 name contributes 15% of β-adj exposure, while a 10% position in a β=0.5 name contributes 5%.
For a long-only biotech book with median β≈1.15, expect β-adj gross around 100–130%.
—
β-Adj High ?Peak β-adjusted gross over the selected date range. Indicates the maximum market exposure the book ran during the window.
—
Position Count ?Count of held instruments on the most recent day. Excludes flat-weight (0%) names.
For a concentrated biotech book, typical range is 20–30. Below 15 = highly concentrated; above 40 = diversified / index-like.
For a concentrated biotech book, typical range is 20–30. Below 15 = highly concentrated; above 40 = diversified / index-like.
—
Period Turnover ?Period Turnover = Σ over each day in range of Σ |weightt − weightt−1| across all tickers.
Captures cumulative rebalance activity. Each unit of "1%" means 1 percentage point of NAV churned between names that day; sum across all days gives total period activity.
Note: a single rebalance that swaps a 5% position for another counts as 10% turnover (5% out + 5% in). For a low-frequency catalyst-driven book, expect 10–50% over a 6-month window. High three-digit values mean frequent rotation.
Captures cumulative rebalance activity. Each unit of "1%" means 1 percentage point of NAV churned between names that day; sum across all days gives total period activity.
Note: a single rebalance that swaps a 5% position for another counts as 10% turnover (5% out + 5% in). For a low-frequency catalyst-driven book, expect 10–50% over a 6-month window. High three-digit values mean frequent rotation.
—
Beta-Adjusted Exposure (XBI) — daily ?Daily line of Σ (wi,t × βi) across the held book, expressed as a % of NAV.
Dotted reference line at 100% marks "unit β exposure". Values above 100% mean the book has more market sensitivity than 100% long XBI; below means less.
Drift between rebalance dates is driven by individual names moving differently from the average.
Dotted reference line at 100% marks "unit β exposure". Values above 100% mean the book has more market sensitivity than 100% long XBI; below means less.
Drift between rebalance dates is driven by individual names moving differently from the average.
Per-name β contribution to gross exposure (latest day) ?For the most recent day in the data, each held ticker's weight × β contribution to β-adj gross. Sorted descending.
Tells you where your market exposure is concentrated — a 10% RVMD position at β=1.15 contributes 11.5% of β-adj gross, while a 10% UTHR position at β=0.40 only contributes 4%.
Tells you where your market exposure is concentrated — a 10% RVMD position at β=1.15 contributes 11.5% of β-adj gross, while a 10% UTHR position at β=0.40 only contributes 4%.
Turnover Δ — daily |Δ weight| ?Daily total Σ |Δwi,t| across every held ticker. Captures actual trading activity each day.
Most days will be near 0 — that's just natural drift from price moves (since drifted weights re-normalize, small daily Δ's appear).
Spikes mark rebalance dates where positions were actively added, trimmed, or replaced. The taller the spike, the larger the trade.
For a 25-name book, a 10% spike day means roughly 10 percentage points of NAV moved between names.
Most days will be near 0 — that's just natural drift from price moves (since drifted weights re-normalize, small daily Δ's appear).
Spikes mark rebalance dates where positions were actively added, trimmed, or replaced. The taller the spike, the larger the trade.
For a 25-name book, a 10% spike day means roughly 10 percentage points of NAV moved between names.
Top per-name turnover contributors (range total) ?Sum of |Δw| per ticker over the selected date range. Tells you which names were traded most actively.
High values typically mean either (a) a position was added then trimmed/sold within the window, or (b) the name was rebalanced up/down multiple times. Sustained tight-band drift won't make this list.
High values typically mean either (a) a position was added then trimmed/sold within the window, or (b) the name was rebalanced up/down multiple times. Sustained tight-band drift won't make this list.
Expected Volatility ?Expected Vol = Σ wi × σi, where σi is each ticker's 60-day rolling annualized stddev of daily returns.
This is a naive aggregation — it sums per-stock vols weighted by position size, ignoring correlations. For a highly-correlated biotech book, this systematically understates true portfolio vol (correlations push it higher).
Compare to Realized Vol to see how good the forecast is.
This is a naive aggregation — it sums per-stock vols weighted by position size, ignoring correlations. For a highly-correlated biotech book, this systematically understates true portfolio vol (correlations push it higher).
Compare to Realized Vol to see how good the forecast is.
End
—
High
—
Low
—
Avg
—
Per-name expected vol contribution (latest day) ?Each held ticker's w × σ contribution to portfolio Expected Vol. Sorted descending. Highlights which positions are driving market-risk exposure.
Expected Beta — SPY ?Expected β SPY = Σ wi × βSPY,i, using precomputed 400-day regression betas from
Different from the XBI β on the Exposure tab — SPY β measures sensitivity to the broad market (S&P 500), not biotech sector. For an LP risk discussion, this is the headline market-exposure number.
Biotech often has SPY β below 1 because its idiosyncratic clinical/regulatory drivers dilute broad-market sensitivity.
stock_macro_betas.Different from the XBI β on the Exposure tab — SPY β measures sensitivity to the broad market (S&P 500), not biotech sector. For an LP risk discussion, this is the headline market-exposure number.
Biotech often has SPY β below 1 because its idiosyncratic clinical/regulatory drivers dilute broad-market sensitivity.
End
—
High
—
Low
—
Avg
—
Per-name SPY β contribution (latest day) ?Each held ticker's w × βSPY. Tells you which positions are driving broad-market exposure.
Realized Volatility ?Realized Vol = 60-day rolling stddev of daily portfolio returns × √252, annualized %.
This is the vol the book actually delivered, not predicted. Reflects the true second moment of P&L distribution.
Compare to Expected Vol — gap = correlation effect. Larger gap = more correlated book than the naive sum suggests.
This is the vol the book actually delivered, not predicted. Reflects the true second moment of P&L distribution.
Compare to Expected Vol — gap = correlation effect. Larger gap = more correlated book than the naive sum suggests.
End
—
High
—
Low
—
Avg
—
Realized vol is a book-level metric — no per-name decomposition without a full covariance matrix.
Value at Risk — 95% / 1 day ?Parametric VaR95,1d = 1.645 × σdaily, where σdaily is the 60-day rolling stddev of daily portfolio returns (in %).
Interpretation: "On 95% of days we expect to lose less than X%." A VaR of 2.67% means: on a typical day, we'd expect losses worse than 2.67% only ~5% of the time (about once a month).
Limitations: assumes normal distribution. Real fat-tail biotech losses (clinical fails, FDA setbacks) can blow through parametric VaR. Use Expected Shortfall (CVaR) or scenario stress for tail context.
Interpretation: "On 95% of days we expect to lose less than X%." A VaR of 2.67% means: on a typical day, we'd expect losses worse than 2.67% only ~5% of the time (about once a month).
Limitations: assumes normal distribution. Real fat-tail biotech losses (clinical fails, FDA setbacks) can blow through parametric VaR. Use Expected Shortfall (CVaR) or scenario stress for tail context.
End
—
High
—
Low
—
Avg
—
Per-name VaR contribution (latest day, naive) ?Each held ticker's w × σ × 1.645 / √252 — the naive marginal VaR if positions were uncorrelated. Sums to portfolio VaR only under zero-correlation assumption; real number is lower due to diversification.
Attribution computed on the 52 trade cycles from the trade log (entry → exit, weighted by entry sizing).
Number of Positions
Total P&L Attribution (sizing-weighted return)
Average Sizing (% of NAV)
Average Return per Position
Historical Exposure — stacked over time (drifted weights × selected dimension)
Daily Rollup — Batting Average & Slugging Ratio (instrument-level, unweighted)
Batting Average (% of positions up each day)
End
—
Avg
—
High
—
Low
—
Slugging Ratio (avg winner / |avg loser|, daily)
End
—
Median
—
High
—
Low
—
Return Distribution (all trades)
Sizing vs Return
Position Returns Waterfall (all trades, sorted)
Profit Factor Formula (closed trades only)
Winners vs Losers
Top 5 / Bottom 5 (sizing-weighted return)
P&L Distribution Waterfall (grouped)
Holding Period — Exited Positions
Holding Period — Open Positions
Return Capture Analysis — % of available return captured per position
Sizing vs Selection P&L Decomposition ?Selection P&L = what you would have earned if every position were equal-weighted (just picking the right names). Sizing P&L = what you earned on top of that from sizing decisions (over/under-weighting). Total = Selection + Sizing. If Sizing > 0, your sizing helped. If Sizing < 0, equal-weighting would have been better.
Top 5 Sizing Decisions That Helped
Top 5 Sizing Decisions That Hurt
Δ Sizing vs Return (each position) ?Each dot is one position. X-axis = how much we over- or under-weighted vs equal weight. Y-axis = realized return. Top-right quadrant = overweighted a winner (good sizing). Top-left = underweighted a winner (missed alpha). Bottom-right = overweighted a loser (bad sizing). Bottom-left = underweighted a loser (saved P&L).
Scenario Stress P&L — current portfolio applied to historical drawdowns ?Apply the historical index return from each named drawdown period (XBI return over that window) to today's portfolio, using realized beta from the inception-to-date period. This is a quick "what would today's book have done" stress test — not a full re-pricing.
Worst 5 Drawdown Contributors (realized)
Best 5 Drawdown Cushions (realized)
Upcoming Catalyst Calendar (open positions only)
Trading Activity by Rebalance — buys/sells + dollar turnover
Standout Calls on the Upside
Missed Opportunities
Trade Log (click column to sort)
Position View — select a ticker
Avg Buy Price
—
Avg Sell Price
—
Avg Instrument Price
—
Current Price
—
Min Price
—
Max Price
—
Avg Exposure
—
P&L Contribution
—
Instrument Price & Trades ?Daily close price with translucent green bubbles for buys and translucent red bubbles for sells. Bubble size proportional to trade magnitude (Δw beyond what price drift explains).
Return Performance — Cumulative P&L & α-Unit (vs XBI) ?P&L = cumulative additive contribution of this position to portfolio NAV (%).
α-Unit (vs XBI) = cumulative excess return after subtracting w × r_XBI. Positive α-Unit means this position beat what passive XBI exposure of the same size would have delivered.
α-Unit (vs XBI) = cumulative excess return after subtracting w × r_XBI. Positive α-Unit means this position beat what passive XBI exposure of the same size would have delivered.
Delta Exposure ?Daily drifted weight of this position as % of NAV. Shows when it was added, scaled up/down, and exited.