A²PT · Strategy Learning Hub
Strategies, unpacked.
156 strategies. Learn the idea through a story, see the mechanics, then test the payoff against real observations.
Start with Covered call156strategies
one at a time
- 133Factors & multi-factor investing
Retail order flow signals
Retail order flow may contain information, but can also reflect attention, crowded trades and short-term price pressure; returns after extreme buying may differ from ordinary net-buying signals. The tool triggers on net flow as a share of average daily volume (ADV), then lets the user choose follow or fade and enter a return scenario.
- 134Global macro
Trading against
Central-bank FX intervention may smooth or reverse currency moves, but markets can anticipate it and direction, credibility and duration vary. Some research links technical-rule returns to intervention periods; this does not mean every intervention is tradable. This simulation maps a manually entered intervention direction to a long or short position and applies a hypothetical FX move.
- 135Trend & time-series momentum
Round-number order clustering
FX research finds stop-loss and take-profit orders cluster near salient round numbers, potentially slowing prices near a level and accelerating them after a break. The pattern need not generalise across markets or levels. This tool uses distance, state, price direction and an assumed next move to compare fading an approach with following a break.
- 136Machine learning
Markov regime-switching trading
Markov-switching models treat returns or volatility as arising from latent regimes with different statistical properties, changing according to transition probabilities. Estimation requires choices about state count, distributions and parameters; short samples, parameter drift and label switching can destabilise results. This teaching calculator performs a one-step probability update only and does not estimate a model.
- 137Factors & multi-factor investing
Alternative-data revenue nowcasting
Card transactions, retail scanner data, web traffic or app usage can provide operating clues before earnings, but coverage, representativeness, privacy, vendor revisions and sample selection affect a nowcast. This tool maps a manually entered activity change through a sensitivity to revenue growth, then applies an explicitly assumed stock response.
- 138Event-driven & special situations
Lazy prices: changes in annual-report text
Lazy Prices compares changes in annual-report text and finds that new content in some risk-factor sections is associated with subsequent returns. Similarity measures can mistake formatting, boilerplate and irrelevant edits for signals; text measurement must also avoid using restated versions. This simulation accepts change percentages directly and does not download or compare filings.
- 139Hedging & risk management
Retirement withdrawal / decumulation rules
Retirement withdrawals balance depletion risk, inflation and spending stability. Fixed real withdrawals stabilise consumption but raise the withdrawal rate after market losses; a fixed percentage of balance reduces depletion risk but makes income fluctuate; guardrails adjust spending as withdrawal rates change. This simulation applies a user-entered annual return path and does not estimate success probabilities.
- 140Market making & liquidity
Order anticipation / back-running
Large institutional orders may be split over time, and temporary price impact and liquidity demand affect execution costs. Research studies high-frequency trading around large orders, but trading ahead on confidential order information or manipulating markets can violate rules. This page models hypothetical price-impact economics only; it gives no method for detecting, front-running or back-running orders.
- 141Market making & liquidity
AMM liquidity provision
A constant-product AMM trades automatically as reserve ratios change. Price moves alter the pool composition, creating impermanent loss versus simply holding; fees may offset some of it. Outcomes depend on the path, fee schedule, concentrated liquidity, arbitrage and contract risk. This page models only a simplified, equal-weight, full-range pool.
- 142Event-driven & special situations
Index inclusion / reconstitution effect
Benchmark changes require tracking funds to adjust holdings around announcements and effective dates, potentially creating predictable demand. Index-provider selection, anticipation, arbitrage capital, tracking error and market size all shape price effects; some studies find the effect has weakened. This simulation uses manual demand and impact assumptions.
- 143Hedging & risk management
Deep hedging
Deep hedging uses neural networks or other function approximators to learn hedge policies while accounting for transaction costs and risk preferences. Training requires suitable historical or simulated paths, a risk objective and out-of-sample validation. This page trains no model; it applies a user-entered policy hedge and a delta benchmark to one-period linear-plus-gamma P&L.
- 144Arbitrage & relative value
Latency arbitrage
When venues update at different speeds, stale quotes risk being selected by faster traders. Latency races involve costly infrastructure, failed fills, adverse selection and market-design concerns; trading against stale quotes may also be restricted by venue rules and fair-dealing requirements. This scenario only nets a manually entered hypothetical edge against costs; it gives no speed or quote-capture methods.