Arcane
Subscribe

Strategies · Library

Eight trades hedge funds run

Each card is one family of trade, from the paper that documented it: the rule, how it fails, the numbers a backtest of it must report and the limits a plan built from it obeys. Alpha files every open Call under one card. Any plan built from a card is on paper, with a stop and the limits below. A person approves or rejects it, and Alpha never places an order.

All 172 strategies from 151 Trading Strategies, by asset class and family.

No card shows a score yet. A card earns one only when paper plans built from it have closed and been scored, the way Calls are.

Hypothetical backtest; not investment advice; paper trading only.Past results, real or simulated, do not predict future ones. ARCANE places no live orders from this page.

The eight cards

From the 60 open Calls that resolve soonest

  1. 01 Trend

    Hold each market in the direction of its own return over the past year, and size every position to the same expected volatility.

    Files a Call when A Call asks whether a price or index will be above or below a level by a date.

    Breaks when A trend turns sharply, as after a crash, before the signal flips.

    Holding
    21 to 252 days
    Stop
    2× the expected move
    Risk used
    100% of your budget
    Largest position
    20% of equity
    Daily stop
    2% loss
    Pause after
    10% drawdown

    Exit The twelve-month return changes sign, or the stop is hit.

    Moskowitz, Ooi and Pedersen (2012), Time series momentum, Journal of Financial Economics 104(2)

    9 open Calls filed here

  2. 02 Pairs and statistical arbitrage

    Find two securities whose prices moved together for a year. When the gap between them opens past two standard deviations, buy the laggard and sell the leader.

    Files a Call when A Call compares two assets: one outperforming, or a spread or ratio between them.

    Breaks when The relationship itself breaks, through a merger, a default or a change in the business.

    Holding
    5 to 126 days
    Stop
    1.5× the expected move
    Risk used
    50% of your budget
    Largest position
    10% of equity
    Daily stop
    2% loss
    Pause after
    8% drawdown

    Exit The gap closes, six months pass, or the stop is hit. Each leg is drafted on its own with half the risk.

    Gatev, Goetzmann and Rouwenhorst (2006), Pairs Trading: Performance of a Relative-Value Arbitrage Rule, Review of Financial Studies 19(3)

    0 open Calls filed here

  3. 03 Carry

    Own what pays more than it costs to finance (the return if prices stand still) and sell what pays less, across currencies, bonds, commodities and stock indexes.

    Files a Call when A Call turns on yields, rate gaps between countries, the shape of the yield curve or futures roll.

    Breaks when Markets flee risk all at once and high-yielding positions fall together.

    Holding
    21 to 252 days
    Stop
    2× the expected move
    Risk used
    75% of your budget
    Largest position
    15% of equity
    Daily stop
    2% loss
    Pause after
    10% drawdown

    Exit The carry changes sign, or the stop is hit.

    Koijen, Moskowitz, Pedersen and Vrugt (2018), Carry, Journal of Financial Economics 127(2)

    7 open Calls filed here

  4. 04 Event-driven

    US stocks have tended to rise in the 24 hours before scheduled Fed decisions, whatever the decision. ARCANE’s rule goes further than the paper: position before a dated event only when Alpha’s chance differs from the market’s by a stated margin, and close once the outcome is public.

    Files a Call when A Call resolves on a dated decision, vote, ruling, release or deal. Every Call no other card claims lands here.

    Breaks when The price jumps through the stop when the news lands; a stop limits the plan, not the gap.

    Holding
    1 to 30 days
    Stop
    1.5× the expected move
    Risk used
    50% of your budget
    Largest position
    10% of equity
    Daily stop
    1.5% loss
    Pause after
    8% drawdown

    Exit The Call resolves, or the stop is hit.

    Lucca and Moench (2015), The Pre-FOMC Announcement Drift, Journal of Finance 70(1)

    43 open Calls filed here

  5. 05 Volatility

    Hold less after a month of high volatility and more after a calm one. The paper found this raised return per unit of risk, because volatile spells are not paid for with higher returns.

    Files a Call when A Call is about volatility itself: the VIX, implied or realised volatility, or the size of a move rather than its direction.

    Breaks when Volatility jumps overnight, before the position can be cut.

    Holding
    5 to 63 days
    Stop
    2.5× the expected move
    Risk used
    50% of your budget
    Largest position
    10% of equity
    Daily stop
    1.5% loss
    Pause after
    8% drawdown

    Exit Volatility returns to its average, or the stop is hit.

    Moreira and Muir (2017), Volatility-Managed Portfolios, Journal of Finance 72(4)

    1 open Call filed here

  6. 06 Factor

    Own companies that are small, cheap, profitable and invest conservatively, against their opposites, and rebalance once a year.

    Files a Call when A Call turns on earnings, guidance, valuations or a style of stock such as value, quality or small companies.

    Breaks when A factor can lag the market for years at a stretch.

    Holding
    63 to 365 days
    Stop
    2× the expected move
    Risk used
    75% of your budget
    Largest position
    15% of equity
    Daily stop
    2% loss
    Pause after
    12% drawdown

    Exit The yearly rebalance, or the stop.

    Fama and French (2015), A five-factor asset pricing model, Journal of Financial Economics 116(1)

    0 open Calls filed here

  7. 07 Mean reversion

    Buy last week’s biggest losers and sell last week’s biggest winners, betting that short, sharp moves partly undo themselves.

    Files a Call when A Call asks whether a price will fall back, rebound or retrace to a level it just left.

    Breaks when The move was news, not noise, and keeps going; trading costs also eat a short-horizon edge.

    Holding
    1 to 10 days
    Stop
    1× the expected move
    Risk used
    50% of your budget
    Largest position
    10% of equity
    Daily stop
    1.5% loss
    Pause after
    6% drawdown

    Exit One to two weeks pass, the move retraces, or the stop is hit.

    Lehmann (1990), Fads, Martingales, and Market Efficiency, Quarterly Journal of Economics 105(1)

    0 open Calls filed here

  8. 08 Liquidity

    Thinly traded assets pay a premium for being hard to sell. Hold them for it in calm markets; when trading thins across the market, cut size and move to the most liquid way to hold the view.

    Files a Call when A Call turns on funding, flows, reserves, repo, trading volumes or bid-ask spreads.

    Breaks when Liquidity vanishes everywhere at once and the premium turns into a forced sale.

    Holding
    21 to 252 days
    Stop
    2× the expected move
    Risk used
    50% of your budget
    Largest position
    10% of equity
    Daily stop
    1.5% loss
    Pause after
    8% drawdown

    Exit Market-wide liquidity recovers, or the stop is hit.

    Amihud (2002), Illiquidity and stock returns: cross-section and time-series effects, Journal of Financial Markets 5(1)

    0 open Calls filed here

Kakushadze and Serur (2018), read in full

Every strategy in 151 Trading Strategies

Each one in our words, with its section cited and the data it needs, filed under the card above that trades the same way. 48 can be built as a symphony; 23 run on data ARCANE already has.

Two sections that describe crimes (§17.2, §17.6) are not listed.

  1. Options58
  2. Stocks21
  3. ETFs8
  4. Fixed income15
  5. Indexes5
  6. Volatility7
  7. Currencies6
  8. Commodities6
  9. Futures7
  10. Structured assets6
  11. Convertibles2
  12. Tax arbitrage3
  13. Weather, energy, inflation4
  14. Distressed assets7
  15. Real estate8
  16. Cash3
  17. Crypto2
  18. Global macro4

02Options58

  1. §2.2Covered callOwn the shares and sell a call above today's price, banking the premium. Suits a flat to mildly up view; upside is capped at the strike, downside is the stock's fall less the premium.CarrySymphony proxy
  2. §2.3Covered putShort the shares and sell a put below the price, collecting the premium. Earns income on a bearish-to-flat view; the gain on a fall is capped at the put strike.Carry
  3. §2.4Protective putBuy shares and a put at or below today's price as insurance. If the stock drops the put pays, so the loss is fixed; if it rises you keep the gain minus the premium paid.Hedges and structures
  4. §2.5Protective callShort the shares and buy a call at or above today's price as insurance against a squeeze. The loss on a rally is capped at the call strike plus the premium paid.Hedges and structures
  5. §2.6Bull call spreadBuy a call near the current price and sell a higher-strike call, same expiry. Costs less than a plain call, gains while the stock rises to the upper strike, then flattens.Trend
  6. §2.7Bull put spreadSell a put at a higher strike and buy a lower-strike put, same expiry, for a net credit. You keep the credit if the stock stays above the upper strike.Carry
  7. §2.8Bear call spreadSell a call at a lower strike and buy a higher-strike call, same expiry, for a net credit. You keep the credit if the stock stays below the lower strike.Carry
  8. §2.9Bear put spreadBuy a put near the current price and sell a lower-strike put, same expiry. Costs less than a plain put, profits as the stock falls to the lower strike, then flattens.Trend
  9. §2.10Long synthetic forwardBuy an at-the-money call and sell an at-the-money put at the same strike and expiry. The pair behaves like holding the stock or a forward to that date, for little or no net cash.TrendSymphony proxy
  10. §2.11Short synthetic forwardBuy an at-the-money put and sell an at-the-money call at the same strike and expiry. Behaves like a short in the stock or a forward sale, with little net cash outlay.Trend
  11. §2.12Long comboBuy an out-of-the-money call and sell an out-of-the-money put, same expiry. The sold put pays for the call, so entry cost is small, but a big fall hurts.TrendSymphony proxy
  12. §2.13Short comboBuy an out-of-the-money put and sell an out-of-the-money call, same expiry. The call premium covers the put, so entry is cheap, but a big rally is costly.Trend
  13. §2.14Bull call ladderA call spread plus a second sold call at an even higher strike, bringing in extra credit. Gains in a moderate rise, but a sharp surge past the top strike starts losing money.Trend
  14. §2.15Bull put ladderSell a put near the money, buy a put at a lower strike, and buy another still lower. Usually the outcome of a bull put spread gone wrong: it now gains if the stock falls hard.Volatility
  15. §2.16Bear call ladderSell a call near the money, buy a call at a higher strike, and buy another higher still. It starts as a bearish credit spread and turns into a bet on a large rally.Volatility
  16. §2.17Bear put ladderA put spread plus a second sold put at a still lower strike, bringing in extra credit. Gains in a moderate fall, but a crash past the lowest strike stops the profit and starts a loss.Trend
  17. §2.18Calendar call spreadSell a short-dated call and buy a longer-dated call at the same strike. Pays most if the stock sits right at the strike when the short one expires, since the near leg melts faster.Volatility
  18. §2.19Calendar put spreadSell a short-dated put and buy a longer-dated put at the same strike. Profit is greatest if the stock sits at the strike when the near put expires.Volatility
  19. §2.20Diagonal call spreadHold a deep in-the-money, longer-dated call that moves almost like the stock, and keep selling short-dated out-of-the-money calls against it for income. A cheaper stand-in for a covered call.Carry
  20. §2.21Diagonal put spreadHold a deep in-the-money, longer-dated put that tracks a falling stock and sell short-dated out-of-the-money puts against it to earn income.Carry
  21. §2.22Long straddleBuy a call and a put at the same strike and expiry. Pays if the stock moves far in either direction; loses the whole premium if it barely moves.Volatility
  22. §2.23Long strangleBuy a call above and a put below today's price. Cheaper than a straddle, but the stock has to travel farther before the position makes money.Volatility
  23. §2.24Long gutsBuy a call below the price and a put above it, so both start with built-in value. Pays on a large move either way, costs more than a straddle, but the worst loss is smaller than the premium.Volatility
  24. §2.25Short straddleSell a call and a put at the same strike and expiry, pocketing both premiums. Wins if the stock ends near the strike; a large move either way produces large losses.Volatility
  25. §2.26Short strangleSell a call above and a put below the price. The winning range is wider than a short straddle's, but the credit is smaller.Volatility
  26. §2.27Short gutsSell a call below the price and a put above it, collecting a large credit because both options have built-in value. Profit is capped and risk higher than a short straddle.Volatility
  27. §2.28Long call synthetic straddleShort the shares and buy two at-the-money calls at one strike. The short stock plus one call acts as a put, so the whole position is a long straddle built a different way.Volatility
  28. §2.29Long put synthetic straddleBuy the shares and buy two at-the-money puts at one strike. The stock plus one put is like a call, so the whole structure is a long straddle.Volatility
  29. §2.30Short call synthetic straddleBuy the shares and sell two at-the-money calls at one strike. The stock plus one short call is like a sold put, so the whole structure is a short straddle.Volatility
  30. §2.31Short put synthetic straddleShort the shares and sell two at-the-money puts at one strike. The short stock plus one sold put acts as a sold call, so the whole structure is a short straddle.Volatility
  31. §2.32Covered short straddleOwn the shares, sell a call and also sell a put at the same strike and expiry. Bigger premium than a covered call, but you take on more downside: effectively two shares' worth of fall below the strike.CarrySymphony proxy
  32. §2.33Covered short strangleOwn the shares, sell a call above the price and also sell a put below it with the same expiry. Earns more than a plain covered call with a cushion before the put is touched.CarrySymphony proxy
  33. §2.34StrapBuy two at-the-money calls and one at-the-money put with the same strike and expiry. It is a long-volatility bet with a bullish tilt: a big move pays, and a rise pays twice as fast as a fall.Volatility
  34. §2.35StripBuy one at-the-money call and two at-the-money puts at one strike and expiry. A long-volatility bet with a bearish tilt: a big move in either direction pays, and a fall pays twice as fast.Volatility
  35. §2.36Call ratio backspreadSell a smaller number of calls near the money and buy a larger number of calls further out of the money (often two bought for one sold). It profits from a sharp rally and may cost little or even pay a credit.VolatilitySymphony proxy
  36. §2.37Put ratio backspreadSell a few puts near the money and buy more puts further out of the money. It pays on a sharp drop, with a modest loss if the price rises and a deeper loss if it sinks only a little.Volatility
  37. §2.38Ratio call spreadBuy one call below the market and sell two or three calls nearer the market, collecting a net credit if possible. It earns from a flat to gently softer market and carries open-ended loss on a big rally.Hedges and structures
  38. §2.39Ratio put spreadBuy one put above the market and sell two or three puts nearer the market for a net credit. It earns from a flat to gently higher market and has heavy loss if the price crashes.Hedges and structuresSymphony proxy
  39. §2.40Long call butterflyBuy one call at a low strike and one at a high strike and sell two calls at the middle strike, all equally spaced. A bet the price ends near the middle strike, with loss limited to the net premium.Volatility
  40. §2.40.1Modified call butterflyA call butterfly where the outer strikes are not evenly spaced, with the upper wing closer to the middle than the lower one. It keeps the neutral view but leans slightly bullish.Volatility
  41. §2.41Long put butterflyBuy one put at a low strike and one at a high strike and sell two puts at the middle strike, equally spaced. Same profile as the call version: a bet on a finish near the middle, with loss limited to the premium.Volatility
  42. §2.41.1Modified put butterflyA put butterfly with uneven strike spacing so the top wing is tight and the bottom wing wide. It keeps the neutral view with a mild bullish bias and should ideally be done for a credit.Volatility
  43. §2.42Short call butterflySell calls at a low and a high strike and buy two calls at the middle strike, equally spaced, for a net credit. A bet the price will move away from the middle, with a small reward and limited risk.Volatility
  44. §2.43Short put butterflySell puts at a high and a low strike and buy two puts at the middle, equally spaced, for a net credit. It profits from a move away from the middle, with small reward and limited loss.Volatility
  45. §2.44“Long” iron butterflySell an at-the-money put and call at the same strike, and buy a put below and a call above as protection, equally spaced, for a credit. Earns most if the price finishes at the middle strike.Hedges and structures
  46. §2.45“Short” iron butterflyBuy an at-the-money put and call at one strike and sell a lower put and a higher call, equally spaced, for a net debit. It profits if the price moves away from the middle, with defined loss.Volatility
  47. §2.46Long call condorBuy calls at the lowest and highest strike and sell calls at the two middle strikes, all evenly spaced, for a small debit. It profits if the price finishes in a band between the middle strikes.Volatility
  48. §2.47Long put condorThe put version of a long condor: buy puts at the lowest and highest strike, sell puts at the two middle strikes, evenly spaced, for a small debit. Gains if the price ends between the middle strikes.Volatility
  49. §2.48Short call condorSell calls at the lowest and highest strikes and buy calls at the two middle strikes, evenly spaced, for a small credit. A bet the price will end outside the range, with a small reward.Volatility
  50. §2.49Short put condorSell puts at the lowest and highest strikes and buy puts at the two middle strikes, evenly spaced, for a small credit. Wins if the price lands outside the whole span; loss is limited.Volatility
  51. §2.50Long iron condorSell an out-of-the-money put and call and buy a further-out put and call as protection, for a net credit. It earns the credit if the price stays between the two short strikes.Hedges and structures
  52. §2.51Short iron condorBuy an out-of-the-money put and call near the market and sell a further-out put and call to cut the cost, for a net debit. It profits from a large move in either direction.Volatility
  53. §2.52Long boxBuy a call and sell a put at one strike and sell a call and buy a put at another, so the payoff at expiry is fixed. Its value is the strike gap, so profit depends on paying less than the gap.Hedges and structures
  54. §2.53CollarOwn the underlying, buy a put below the market for protection, and sell a call above it to pay for the put. Losses and gains are both bounded.Hedges and structuresSymphony proxy
  55. §2.54Bullish short seagull spreadSell a put below the market, buy a call near the market, and sell a higher call, usually for zero cost. It gains if the price rises, up to a cap, and loses if the price falls.Hedges and structuresSymphony proxy
  56. §2.55Bearish long seagull spreadBuy a put below the market, sell a call near it, and buy a higher call to cap the loss on a rally, usually at zero cost. It gains if the price falls.Hedges and structures
  57. §2.56Bearish short seagull spreadSell a put far below, buy a put near the market, and sell a call above, usually for zero cost. It gains on a moderate fall but carries unlimited loss on a rally.Hedges and structures
  58. §2.57Bullish long seagull spreadBuy a put far below, sell a put near the market, and buy a call above, usually for zero cost. It gains on a rise without a cap and loses a bounded amount on a fall.Hedges and structuresSymphony proxy

03Stocks21

  1. §3.1Price-momentumRank stocks by their return over a past formation window, usually about a year with the latest month skipped. Buy the top group (and short the bottom group in a market-neutral version). Winners tend to keep winning for some months.TrendSymphony proxy
  2. §3.2Earnings-momentumRank stocks by how far the latest quarterly earnings per share beat the figure from a year earlier, scaled by the variability of that gap. Buy the biggest positive surprises and short the biggest negative ones, expecting earnings news to be absorbed slowly.Factor
  3. §3.3ValueRank stocks by book value relative to price. Buy the cheapest decile and short the most expensive decile, on the idea that cheap companies are mispriced and drift back toward fair value over months.Factor
  4. §3.4Low-volatility anomalyStocks that were calm in the past have tended to earn at least as much as the jumpy ones, which cuts against the idea that more risk pays more. Buy the lowest-volatility decile, short the highest.FactorSymphony proxy
  5. §3.5Implied volatilityStocks whose call implied volatility rose over the past month have tended to do better, and those whose put implied volatility rose have tended to do worse. Go long the first group and short the second.Volatility
  6. §3.6Multifactor portfolioCombine several stock factors such as value and momentum, which often move against each other, so the blend is steadier than either. Either split money between single-factor portfolios or average each stock's ranks across the factors.FactorSymphony proxy
  7. §3.7Residual momentumLike price momentum, but each stock's returns are first cleaned of the part explained by market, size and value factors. Ranking on what is left aims to reward stock-specific trends and avoid riding broad market swings.TrendSymphony proxy
  8. §3.8Pairs tradingPick two stocks that usually move together. When one gets ahead of the other, short the one that ran and buy the one that lagged, expecting the gap to close. Dollar amounts are balanced so market moves cancel.Pairs and relative valueSymphony proxy
  9. §3.9Mean-reversion – single clusterTake a group of stocks that move together, such as one industry. Short those that rose more than the group average and buy those that fell more, sizing each bet by how far it sits from the average, expecting the group to pull back together.Mean reversionSymphony proxy
  10. §3.9.1Mean-reversion – multiple clustersRepeat the single-group fade across several groups, such as sectors, each demeaned against its own average. A regression on group membership gives the same answer in one step, and groups can be funded evenly.Mean reversionSymphony proxy
  11. §3.10Mean-reversion – weighted regressionA wider form of the group mean-reversion trade. Regress stock returns on risk factors (industry groups and other characteristics), weighting calmer stocks more, and bet against the leftover. This makes positions neutral to every factor, not only one group.Mean reversionSymphony proxy
  12. §3.11Single moving averageHold a stock while its price is above a moving average and exit, or go short, when it falls below. The average is a trend filter: price staying above it signals an uptrend. Works one stock at a time.TrendSymphony
  13. §3.12Two moving averagesHold while a short moving average is above a longer one, exit or go short when it falls below. An optional stop closes a long if the price drops a few percent in a day, without waiting for the cross.TrendSymphony
  14. §3.13Three moving averagesRequire three moving averages of increasing length to be stacked in order before entering, to filter false signals. Long when short is above medium is above long; exit when the short falls under the medium.TrendSymphony
  15. §3.14Support and resistanceCompute a pivot level from yesterday's high, low and close, with a resistance above and a support below it. Go long above the pivot and exit near resistance; go short below the pivot and exit near support.Mean reversion
  16. §3.15ChannelTrack the highest and lowest prices over a window as a channel. Buy when price touches the floor and sell or short at the ceiling, expecting a bounce. A break through the edge may instead mark a new trend.Mean reversionSymphony proxy
  17. §3.16Event-driven – M&AAfter a takeover is announced, the target usually trades a bit below the deal price. Buy the target (and short the buyer in a stock deal) to collect the gap if the deal closes, accepting a large loss if it fails.Event-driven
  18. §3.17Machine learning – single-stock KNNFor one stock, describe each past day by features such as moving averages of price and volume. Find the past days most similar to today and average what happened next to forecast the coming days. Trade if the forecast clears a threshold.Hedges and structures
  19. §3.18Statistical arbitrage – optimizationGiven expected returns for many stocks and a covariance model, choose weights that maximize return per unit of risk, optionally forced to net to zero dollars. It is a sizing method wrapped around some other signal.Pairs and relative value
  20. §3.19Market-makingPost buy orders at the bid and sell orders at the ask to earn the spread. It struggles when informed traders trade against you, so makers add short-term direction signals and rely on speed to stay first in the queue.Liquidity
  21. §3.20Alpha combosBlend a very large number of weak trading signals into one portfolio, weighting each by its own volatility and removing overlap between them with a regression. Each is too faint to trade alone after costs; combined they can be worthwhile.Hedges and structures

04ETFs8

  1. §4.1Sector momentum rotationHold the sector ETFs that did best over the past 6 to 12 months and sell the laggards, rebalancing every month or few. Recent winners tend to keep winning for a while, and ETFs avoid buying many single stocks.TrendSymphony proxy
  2. §4.1.1Sector momentum rotation with MA filterSame sector rotation, but a top-ranked ETF is only bought if its price is above its own long moving average (and shorted only if below). The filter skips winners that have already started to roll over.TrendSymphony proxy
  3. §4.1.2Dual-momentum sector rotationPick the best sector ETFs by past return, but only while the broad market ETF is in an uptrend. When the market trend is down, move everything into an asset that does not move with stocks, such as gold or Treasuries.TrendSymphony
  4. §4.2Alpha rotationRank ETFs by their alpha, the part of their return left over after a regression on market, size and value factors, instead of raw return. The idea is that skill-like excess return persists a bit.FactorSymphony proxy
  5. §4.3R-squaredRank ETFs by how much of their return the factor regression explains (low fit means more independent behaviour) and by alpha, then favour low-fit, high-alpha ETFs. Low fit is read as a sign of active selection.Factor
  6. §4.4Mean-reversionRank ETFs by where yesterday's close sat inside its high-low range. Closes near the top of the range are treated as rich and near the bottom as cheap, betting the next day partly reverses.Mean reversionSymphony proxy
  7. §4.5Leveraged ETFs (LETFs)Leveraged and inverse-leveraged ETFs reset daily, which tends to wear down their value over time in choppy markets. The trade shorts both a bull and a bear fund on the same index and parks the cash in Treasuries.Hedges and structures
  8. §4.6Multi-asset trend followingHold a small set of ETFs across stocks, bonds, gold and more, keeping only those with positive past return (and optionally above a long moving average). Weight by return, return per unit of volatility, or return per variance.TrendSymphony proxy

05Fixed income15

  1. §5.2BulletsHold bonds of one maturity only, chosen from the view on rates: longer if rates are expected to fall, shorter if they are expected to rise. Buying over time spreads the entry rate.CarrySymphony proxy
  2. §5.3BarbellsSplit money between short and long maturity bonds. The long end earns more yield, the short end can be reinvested if rates rise, and the mix has more convexity than a middle bond with the same average sensitivity.CarrySymphony
  3. §5.4LaddersSpread money evenly across bonds that mature at regular steps, say one to ten years. As the shortest matures, buy a new longest. This keeps duration steady and smooths reinvestment.CarrySymphony
  4. §5.5Bond immunizationBuild a bond portfolio whose duration equals the time until a known future payment, so a parallel move in rates leaves the funding result about unchanged. With three bonds convexity can be matched too.Hedges and structures
  5. §5.6Dollar-duration-neutral butterflyCombine a long position in a short and a long maturity with a short position in a middle maturity, sized so cost is zero and the rate sensitivity cancels. It profits from a change in the shape of the curve.Pairs and relative value
  6. §5.7Fifty-fifty butterflyA butterfly where each wing carries half the middle's rate sensitivity. It stays roughly neutral to small steepening or flattening, but is not zero cost.Pairs and relative value
  7. §5.8Regression-weighted butterflyA butterfly where the wing sizes use a historical ratio of how much the body-to-short-wing spread moves compared with the long-wing spread, because short rates tend to be more volatile.Pairs and relative value
  8. §5.8.1Maturity-weighted butterflySame as the regression-weighted butterfly, but the spread ratio comes from the gaps between the three maturities rather than a regression.Pairs and relative value
  9. §5.9Low-risk factorFavour bonds with high credit quality and short maturity, because lower-risk bonds have tended to earn more per unit of risk than riskier ones.FactorSymphony proxy
  10. §5.10Value factorEstimate a fair credit spread for each bond from its rating and maturity, then buy bonds whose actual spread is widest above that fair value, assuming they are cheap.Factor
  11. §5.11Carry factorEstimate each bond's expected gain as its yield plus the price gain from sliding down the curve as it ages, then buy the highest-carry and sell the lowest-carry bonds.Carry
  12. §5.12Rolling down the yield curveBuy medium or long bonds where the curve is steepest and hold while they slide toward shorter maturity and higher prices, then sell and repeat.CarrySymphony proxy
  13. §5.13Yield curve spread (flatteners & steepeners)Bet on the gap between short and long yields. Expect rates to fall and the curve to steepen: buy the front, sell the back. Expect rates to rise and the curve to flatten: the reverse. Matching dollar durations avoids small parallel shifts.Carry
  14. §5.14CDS basis arbitrageWhen a bond's yield spread is higher than the price of default insurance on it, buy the bond and the insurance for a locked profit.Pairs and relative value
  15. §5.15Swap-spread arbitrageA dollar-neutral trade between an interest rate swap and a Treasury of the same maturity. Profit depends on how the floating rate compares with the repo rate, effectively a bet on bank funding stress.Pairs and relative value

06Indexes5

  1. §6.2Cash-and-carry arbitrageCompare an index future with its fair price from the spot level, dividends and interest. If the future is rich, sell it and buy the basket; if cheap, do the reverse; close when they converge.Pairs and relative value
  2. §6.3Dispersion trading in equity indexesIndex options often price higher volatility than the stocks inside would justify. The trade buys straddles on the stocks and sells a straddle on the index, sized to match payoffs.Volatility
  3. §6.3.1Dispersion trading – subset portfolioA cheaper version of dispersion that only buys straddles on a subset of stocks with options and low specific risk, using a smoother correlation model.Volatility
  4. §6.4Intraday arbitrage between index ETFsTwo ETFs on the same index (like SPY and IVV) sometimes get slightly out of line for moments. Buy the cheaper and sell the dearer when the gap beats costs; unwind when it closes.Liquidity
  5. §6.5Index volatility targeting with risk-free assetHold an index and cash so that portfolio volatility stays near a target: the index weight is the target divided by recent index volatility. Rebalance on schedule or when weight shifts enough.Hedges and structuresSymphony proxy

07Volatility7

  1. §7.2VIX futures basis tradingCompare the front VIX future with the VIX index. When the curve slopes up the future tends to drift down, and when it slopes down it tends to drift up, so the trade fades the basis.Carry
  2. §7.3Volatility carry with two ETNsShort a short-dated volatility note and buy a medium-dated one in a ratio from past data. The short-dated note tends to bleed value because the curve usually slopes up; the long leg softens spikes.Carry
  3. §7.3.1Hedging short VXX with VIX futuresSame short on the front-heavy note, but the hedge is a basket of medium-dated VIX futures instead of the other note, with weights chosen to follow the note closely.Pairs and relative value
  4. §7.4Volatility risk premiumOptions tend to be priced for more movement than occurs, so sell index straddles when implied volatility sits above the realised figure. It pays in quiet markets and loses in sell-offs.Volatility
  5. §7.4.1Volatility risk premium with gamma hedgingKeep the sold straddle direction-neutral by trading the underlying as it moves, so the return comes from time decay rather than a view on volatility level. The hedging cost eats the premium if the market trends.Volatility
  6. §7.5Volatility skew - long risk reversalPuts far from the money usually cost more than matching calls. Buy the call and sell the put to capture the skew. It is really a bullish bet that loses if the index falls enough.Volatility
  7. §7.6Volatility trading with variance swapsA variance swap pays on the difference between realised and agreed variance, so no continual delta hedging is needed. Long profits from higher realised volatility. Can also be used for a dispersion trade.Volatility

08Currencies6

  1. §8.1Moving averages with HP filterCurrency series are noisy, so first strip out high-frequency wiggles with a smoothing filter, then compare a short and a long average of the smooth trend. Short above long is a buy signal.TrendSymphony proxy
  2. §8.2Carry tradeCurrencies with high interest rates have tended to hold or gain value rather than fall as theory says. Buy forwards on high-yield currencies and sell them on low-yield ones; hedged versions are risk-free but rarely pay.Carry
  3. §8.2.1High-minus-low carryRank currencies by forward discount, buy the top group and sell the bottom group to get a dollar-neutral portfolio. Group size depends on how many currencies are used.Carry
  4. §8.3Dollar carry tradeAverage the forward discount across a set of foreign currencies. If positive, go long all of them against the dollar; if negative, short them. It tracks the health of the US economy.Carry
  5. §8.4Momentum & carry comboMix a currency momentum rule with a carry rule, weighting each to minimise variance of the combined return using past return correlation.TrendSymphony proxy
  6. §8.5FX triangular arbitrageTrade around a loop of three currencies and profit if the product of rates exceeds one after bid/ask spreads. Gaps close in moments so speed is essential.Pairs and relative value

09Commodities6

  1. §9.1Roll yieldsWhen the curve slopes down (backwardation) long futures tend to earn positive roll; when up, short. Rank commodities by front-over-second price ratio, buy the high and sell the low.Carry
  2. §9.2Trading based on hedging pressureWeekly regulator reports show how long hedgers and speculators are in each commodity. Split by speculators, then pick by hedgers; buy one end, sell the other, hold about six months.Liquidity
  3. §9.3Portfolio diversification with commoditiesHold a fixed slice in commodities alongside stocks to cut risk, or tilt the slice up and down with policy rates.Hedges and structuresSymphony
  4. §9.4ValueDefine value as the price from five years ago over the current price. Buy the cheapest third, sell the dearest third, monthly.FactorSymphony proxy
  5. §9.5Skewness premiumCompute the skew of past returns per commodity, buy the lowest-skew group, sell the highest.Factor
  6. §9.6Trading with pricing modelsModel the log price as mean-reverting, fit it to history, compare model futures prices with market ones; buy cheap, sell rich. Beware of in-sample fits that fail out of sample.Mean reversion

10Futures7

  1. §10.1Hedging risk with futuresA firm that will need or sell a commodity later takes the matching futures position now to lock the price.Hedges and structures
  2. §10.1.1Cross-hedgingUse a related futures contract when the exact one is not traded. Leaves basis risk and mismatch risk; hedge ratio from regression.Hedges and structures
  3. §10.1.2Interest rate risk hedgingUse interest rate futures to protect a bond holding, sizing by a conversion factor or by duration.Hedges and structures
  4. §10.2Calendar spreadBuy the near contract and sell the deferred (or opposite) based on expected supply and demand tightness.Pairs and relative value
  5. §10.3Contrarian trading (mean-reversion)Compare each future with the equal-weighted average over the last week, buy the underperformers and sell the outperformers, optionally scaled by volatility.Mean reversionSymphony proxy
  6. §10.3.1Contrarian trading - market activityAdd filters: take contracts with the largest volume rise and the weakest open interest rise, then apply the reversal rule to them.Mean reversion
  7. §10.4Trend following (momentum)Go long contracts with positive past returns and short those with negative, scaling each by inverse volatility. Variants demean against the market.TrendSymphony proxy

11Structured assets6

  1. §11.2Carry, equity tranche – index hedgingBuy the riskiest slice of a credit-loss pool and sell the matching credit index to cancel most of the market exposure. The tranche pays more premium than the index costs, so the difference is carry. The risk left over is losses concentrated in the riskiest slice.Carry
  2. §11.3Carry, senior/mezzanine – index hedgingSell protection-style exposure on a higher-quality tranche and hedge by buying the credit index. The index pays more premium than the short tranche costs, so the gap is carry. It is roughly the opposite of the equity tranche trade.Carry
  3. §11.4Carry – tranche hedgingBuy a low-quality tranche and hedge it by selling a higher-quality tranche of the same pool. The riskier slice pays more premium, so the net is positive carry while market-wide credit moves roughly cancel.Carry
  4. §11.5Carry – CDS hedgingBuy a low-quality tranche and hedge with a sold single-name credit default swap that costs less than the tranche pays. The net premium is the carry, but a single name hedges the pool only loosely.Carry
  5. §11.6CDOs – curve tradesTrade the gap between short and long maturity tranches. A flattener sells short-term and buys long-term protection when the curve is expected to flatten; a steepener does the opposite. It can be sized to be neutral on notional, risk duration or carry.Carry
  6. §11.7Mortgage-backed security (MBS) tradingBuy pass-through mortgage bonds and hedge their rate sensitivity with interest rate swaps. The main leftover risk is prepayment: when rates fall, homeowners refinance and the bond's price gains are capped. Hedge sizes come from a prepayment model or from fitting history.Carry

12Convertibles2

  1. §12.1Convertible arbitrageBuy a convertible bond, which tends to be priced cheaply when issued, and short a number of the issuer's shares equal to the option's sensitivity times the conversion ratio. The hedge is revised daily. Profit comes from the cheapness and from volatility in the stock.Hedges and structures
  2. §12.2Convertible option-adjusted spreadBuy the convertible bond of an issuer with the higher option-adjusted spread and sell another of the same issuer with the lower, expecting the two spreads to converge. The spread is found by splitting price into a plain bond plus a call option and shifting the interest curve until the model matches the market price.Pairs and relative value

13Tax arbitrage3

  1. §13.1Municipal bond tax arbitrageBorrow money and buy tax-exempt municipal bonds. If the borrowing cost after the tax deduction is lower than the bond yield, the difference is profit. It only works for firms whose tax rules allow both the exemption and the interest deduction.Carry
  2. §13.2Cross-border tax arbitrageSome countries give domestic shareholders a tax credit on dividends, and the share price drops by more than the dividend on the ex-dividend date. A foreign holder can sell before the dividend and buy back after, or lend the stock to a domestic holder and get the credit value shared back.Event-driven
  3. §13.2.1Cross-border tax arbitrage with optionsA foreign holder sells the stock before a dividend and writes a deep in-the-money put. If tax credits push put prices up, the put is exercised after the dividend and the holder keeps the dividend plus credit value, matching the stock-loan version.Event-driven

14Weather, energy, inflation4

  1. §14.1Inflation hedging – inflation swapsBuy an inflation swap to receive actual inflation and pay a fixed rate, or sell one for the reverse. The fixed rate is usually the yield gap between Treasury and inflation-protected bonds. Profit comes when realised inflation beats what was priced in. Variants pay once at the end or every year.Hedges and structuresSymphony proxy
  2. §14.2TIPS-Treasury arbitrageSell a Treasury bond and buy a replicating package: an inflation-protected bond, inflation swaps that turn its indexed coupons into fixed ones, and small zero-coupon bond positions to match the rest. The package replicates the Treasury's payments for less than the Treasury costs, leaving an upfront gain.Pairs and relative value
  3. §14.3Weather risk – demand hedgingBusinesses whose sales depend on temperature hedge with futures or options on heating or cooling degree days, which measure how far daily average temperature is below or above 65 F. The hedge size comes from how strongly weather-affected demand moves with the index.Hedges and structures
  4. §14.4Energy – spark spreadThe spark spread is the power price minus the gas cost to generate it, using the plant's heat rate. Short electricity futures and long gas futures to hedge a gas-fired plant or bet on its margin. Contract sizes force the hedge ratio to be rounded to a small whole-number ratio.Pairs and relative value

15Distressed assets7

  1. §15.1Buying and holding distressed debtBuy the debt of companies in or near default at steep discounts and hold, mostly spread across industries and seniority levels. Few holdings pay off but the winners recover a lot, helped by market overreaction around default or bankruptcy dates.Event-driven
  2. §15.2Active distressed investingBuy distressed assets in order to gain some control over the company's direction, either through Chapter 11 or an out-of-court deal with creditors. Profit comes from increasing the value of the reorganised firm.Event-driven
  3. §15.2.1Planning a reorganizationA large debt holder files its own reorganisation plan in court to win a seat in management and lift the firm's value. Plans from big creditors tend to be more competitive.Event-driven
  4. §15.2.2Buying outstanding debtBuy a distressed company's debt at a discount, expecting that part of it converts into equity after reorganisation and gives the holder a degree of control.Event-driven
  5. §15.2.3Loan-to-ownLend on a secured basis to a distressed company that is not yet bankrupt. Either it recovers and the equity gains, or it files and the loan converts into equity with control rights.Event-driven
  6. §15.3Distress risk puzzleFirms more likely to go bankrupt have tended to earn lower, not higher, returns. Buy the safest firms and sell the riskiest, ranked by a modelled bankruptcy probability, with monthly or yearly rebalance.FactorSymphony proxy
  7. §15.3.1Distress risk puzzle – risk managementThe healthy-minus-distressed spread loses badly when markets rebound after a crash. Scale the position by target volatility divided by volatility realised over the prior year to limit those losses.Factor

16Real estate8

  1. §16.2Mixed-asset diversification with real estateAdd real estate to a stock and bond portfolio because its correlation with them is low, even in crises and especially over long horizons. The size of the allocation depends on risk preference and horizon.Hedges and structuresSymphony
  2. §16.3Intra-asset diversification within real estateSpread real estate holdings over area, property type, size and region instead of concentrating in one place. Standard portfolio-construction methods set the weights.Hedges and structuresSymphony proxy
  3. §16.3.1Property type diversificationHold different kinds of property, such as apartments, offices, industrial buildings and shopping centres. Studies suggest this lowers non-market risk even after costs.Hedges and structuresSymphony proxy
  4. §16.3.2Economic diversificationDiversify property by regions grouped by their main industry, jobs and income levels. It can reduce non-market risk and costs.Hedges and structures
  5. §16.3.3Property type and geographic diversificationCombine two ways of spreading risk, for example four property types across four US regions, giving sixteen groups to hold.Hedges and structures
  6. §16.4Real estate momentum – regional approachProperty markets in US metro areas show momentum: areas that have done well recently tend to keep doing well, and weak ones keep lagging. REITs or housing index futures can be used for a zero-cost version.Trend
  7. §16.5Inflation hedging with real estateReal estate returns have tended to follow inflation, with commercial property adjusting faster than some other types. Hold property to protect against rising prices.Hedges and structuresSymphony proxy
  8. §16.6Fix-and-flipBuy a run-down property well below market, renovate it and sell for more than the purchase and renovation cost combined.Hedges and structures

17Cash3

  1. §17.3Liquidity managementDecide how much cash to keep so unexpected needs can be met without selling other assets at a cost. For companies, too little cash can mean lost opportunities or distress.LiquiditySymphony
  2. §17.4Repurchase agreement (REPO)Lend cash and receive securities as collateral, with an agreement to reverse the trade later (or borrow cash the same way). Terms run from a day to about six months.Carry
  3. §17.5PawnbrokingMake a secured cash loan against an item such as jewellery or electronics, at a fraction of its appraised value. If not repaid, keep or sell the item.Carry

18Crypto2

  1. §18.2Artificial neural network (ANN)Train a neural network on technical indicators of Bitcoin returns, such as moving averages, volatility and RSI, to predict which return band comes next. Buy when the top band is predicted and sell when the bottom band is.Trend
  2. §18.3Sentiment analysis – naïve Bayes BernoulliCollect tweets with chosen keywords, mark each word as present or absent, and use a naive Bayes classifier to predict whether Bitcoin goes up or down. Trade on the predicted class.Hedges and structures

19Global macro4

  1. §19.2Fundamental macro momentumMarkets underreact to changes in growth, trade, monetary policy and risk mood. Rank countries on these four trends, buy the top group and sell the bottom group, combining asset classes with equal weights.Trend
  2. §19.3Global macro inflation hedgeWhen headline inflation runs well above core inflation, commodity prices are pushing it. Allocate to commodities by the size of that gap, from nothing up to the full portfolio.Hedges and structures
  3. §19.4Global fixed-income strategyRank government bonds of different countries on growth, inflation, sovereign risk, real rates, value, momentum and term spread. Buy the best group and sell the worst, often through country bond ETFs.FactorSymphony proxy
  4. §19.5Trading on economic announcementsHold stocks on days of key announcements such as Federal Reserve meetings and Treasuries on other days.Event-driven

What every backtest reports

Never the Sharpe ratio alone
  1. 01

    Sharpe

    Yearly return above cash, divided by yearly volatility.

  2. 02

    Sortino

    The same, counting only the falls.

  3. 03

    Calmar

    Yearly return over the worst fall from a peak.

  4. 04

    Max drawdown

    The deepest fall from a peak to a trough.

  5. 05

    Hit rate

    The share of days that made money.

  6. 06

    Turnover

    How many times a year the whole portfolio is traded.

The overfitting guard

The same bars the builder uses
  1. 01

    Deflated Sharpe confidence at least 95%

    The chance the true Sharpe ratio is above zero, after raising the bar for every version of the idea that was tried.

  2. 02

    Overfitting odds at most 20%

    How often the best version on one half of history lands in the bottom half of the other.

  3. 03

    Still positive with costs doubled

    A result that only survives cheap trading is not a result.