Part 10
Directional Strategies
On this page
10.1 Discretion and price action
Part 9 was a set of rules. Each strategy there had a defined entry, a defined exit, a sizing rule and a backtest behind it, so a trader could run it the same way every week and judge it on hundreds of past trades. Directional trading works differently. A view that a market will rise or fall over the next few weeks has no fixed rulebook and no clean backtest, and two traders reading the same chart will often place different trades. The decisions about whether to trade, where to enter, where to be wrong and when to leave all sit with the trader.
That makes this part far more discretionary, and it changes what the platform's readings are for. The Regime, positioning, skew and momentum narrow the field: out of hundreds of markets they point to the handful where trend and crowd line up, or where they have stretched to an extreme. They also give the context for a view: whether it runs with a mature move, against a crowded one, or into a market whose inputs disagree. None of them says where to buy. Price action supplies that part: the level to enter at, the place a stop goes, and the point at which the idea is invalidated.
So every trade in this part has three layers. The data says which markets are worth a look and which way the evidence leans. The chart says where to engage and what proves the view wrong. The trader decides, sizes and manages it. The rest of this lesson is the technical baseline for the second layer. Part 8 covered the theory behind these concepts in more depth; what follows is the working version, aimed at building a trade.
10.1.1 Thinking about discretion honestly
The outcome of any single discretionary trade is close to a coin flip. A clean setup at a good level can fail the moment you enter, and the trade you nearly skipped can run straight to target. What makes discretion workable is repetition: a setup defined tightly enough that you can count it, taken enough times for its expected value to show. Part 11 covers how many trades that takes and how to size while you find out. The practical point here is that a setup has to be specific. "Long at support" is too loose to journal. "Long at the first retest of a weekly demand zone, stop below the zone, in a market with a Regime above zero" is a setup you can count, and a journal of setups like it slowly turns into a rulebook.
10.1.2 Supply, demand and urgency
Every trade has a buyer and a seller, so "more buyers than sellers" never happens. What changes is urgency. When buyers are willing to pay higher and higher prices to get filled, price rises. When sellers accept lower and lower prices, it falls. Price stops when the urgent side runs out or resting orders on the other side absorb it. Markets spend most of their time in balance, trading around a price both sides accept, and comparatively little time trending. When price leaves balance, it tends to travel to the next area where a lot of volume traded before. Balance, and the moves between areas of balance, sits under every concept below.
10.1.3 Market structure: trend or range
Structure is the sequence of swing highs and swing lows. An uptrend makes higher highs and higher lows. A downtrend makes lower highs and lower lows. Anything else is a range. Reading which of the three you are in comes first, because it decides which setups make sense: in a trend you buy pullbacks or sell rallies with it, and in a range you trade the edges and expect breakouts to fail more often.
The level that matters most in a trend is the swing low that produced the latest high (or, in a downtrend, the swing high that produced the latest low). A close through it is a structure shift.
10.1.4 Support, resistance and the level flip
Support and resistance are prices where the market reacted before. You find them by looking left: prior highs and lows where price turned. They work because traders see the same levels and place orders around them, so each level holds a pool of resting buy or sell orders.
Two habits make levels more useful. First, once a level breaks, it is often tested from the other side: broken resistance becomes support and broken support becomes resistance. Second, a level weakens with each test, because every touch fills some of the resting orders that defend it. The first retest is the best one, and a level that has been hit five times is closer to breaking than to holding.
There are two ways to trade a level. The conservative way waits for price to reach it and show a reaction, then enters. The aggressive way rests a limit order at the level in advance. The limit order gets a better price and a tighter stop, and it also gets filled on the trades where the level breaks outright or price swings deep through it before turning. Blind limit orders belong on first tests of a level only.
10.1.5 Failed breaks and follow-through
When price breaks a level, breakout traders buy with market orders and shorts sitting at the level get stopped out, which is more market buying. A large seller who needs size uses exactly that burst of buying to fill sell orders passively. If the breakout runs out of buyers, there is no follow-through, everyone who bought the break is underwater, and their exits push price back through the level. These failed breaks (also called swing failures or excess) happen in every market and on every timeframe.
Follow-through is the test that separates a real break from a thin one. A break that keeps going, closes beyond the level and builds new trading above it is accepted. A break that pokes through, stalls and closes back inside is rejected. Judge it on the timeframe that produced the level: a weekly level is accepted or rejected on weekly closes, and a five-minute wick through it says little.
10.1.6 Efficient and inefficient moves
How price travels says who was behind it. An efficient move is two-sided: overlapping candles, pullbacks, a grind where both buyers and sellers get filled at most prices. An inefficient move is one-sided: large candles stacked with little overlap, where someone needed to get filled urgently and took every offer in the way.
Inefficient moves tend to get revisited. The trader behind an urgent move rarely fills full size, so the area often gets defended on the way back, and participants who missed the move want in at those prices.
10.1.7 Supply and demand zones
A demand zone is the consolidation before an inefficient move up that breaks structure, meaning the move closes above the previous swing high. A supply zone is the mirror: the consolidation before an inefficient move down that breaks structure. The structure break is the filter. A base followed by a weak move that fails to clear the prior swing is just a pause.
The consolidation is where a large participant was filling orders before overwhelming the other side, so some of that interest is often still there on a return, and traders caught on the wrong side of the base want out at breakeven. The zone gives an entry at its edge and invalidation on its far side, and like a level, its first touch is the best one.
Strong moves often skip the full retracement. When a move carries real urgency, everyone waits for the pullback into the original base, and price often holds the body of the inefficient move as support instead, then makes its next leg. A trader who only bids the base misses those trades, and one who chases the middle of the leg gets the worst location on the chart. Marking both the base and the inefficient leg gives two areas to plan around.
10.1.8 Building the trade
With those concepts in place, a trade comes together in the same order every time.
Timeframes. Use a higher timeframe for location and a lower one for the trigger. For trades held days to weeks, mark levels and zones on the weekly chart and execute and manage on the daily. Popular timeframes (monthly, weekly, daily, four-hour) earn their place because everyone watches them, which is what makes their levels react. Manage the trade on the timeframe you entered on. A position taken off a daily zone and then watched on the five-minute chart gets cut on noise.
Entry. Either rest a limit order at the zone or level, or set a price alert there and decide when it fires. The limit order suits first tests of clean zones and markets you cannot watch. The alert suits levels where you want to see the reaction first: a rejection candle, a failed break, a close back inside.
Stop. The stop goes beyond the level or zone, at the point where the idea is wrong: below a demand zone, above a supply zone, past the swing that defines the structure. Placing it exactly at the obvious high or low puts it where everyone else's stop sits. A volatility-based stop (a multiple of the market's average range or realized volatility) avoids that cluster and makes risk comparable across markets. Some traders combine the two: a hard stop set by volatility that always executes, and a soft stop at the technical level that triggers an exit only on a close through it.
Target and management. Set the first target at the next obvious level or the next area of liquidity: a prior swing, an untested zone, an old balance area. Higher-timeframe trades tend to run further than intraday ones, so cutting them at the first sign of profit gives up the moves that pay for the losers. Move the stop for technical reasons only: when the market makes a new swing that holds, the stop can go behind it. A stop moved to entry because the trade is in profit is an arbitrary number with no link to the chart.
Context check. Before placing the order, set the trade beside the data from the rest of this part. A long at a weekly demand zone in a market whose Regime has turned firmly negative, with positioning crowded long and skew bidding puts, is a different trade from the same chart in a market where the readings lean the same way. The chart supplies the location; the context decides whether the location deserves capital and how much.
10.1.9 From the chart to the platform
That is the technical baseline: structure to tell trend from range, levels and zones to locate a trade, failed breaks and follow-through to judge a move, and a plan for entry, stop and target before the order goes in. The chart answers where. The next lessons answer which markets. They walk through where the candidates come from on the site: the Regime, starting from the S&P 500, then the equities, futures and crypto screeners and lenses, each with worked examples of a reading turning into a trade.
End of lesson 10.1Discretion and price action
10.2 The Regime: reading extremes and divergences, starting with SPX
Every directional trade is taken into some set of conditions: a trend running or fading, an options market calm or bidding protection, a crowd leaning one way or spread out. The Regime puts those conditions into one number per market, from -100 to +100. This lesson builds the SPX version, gives the inputs for the other asset classes, and teaches the two readings that carry most of its value in every market: the extremes, and the divergences with price.
10.2.1 What the Regime is
The Regime is a description of conditions. Its sign says which way the evidence leans, positive when the inputs lean bullish and negative when they lean bearish. Its size says how much of the evidence agrees and how firmly. A reading near zero is a mixed market, and a mixed market can still travel a long way. The Overview marks plus and minus 70 as the extreme zone: a reading out there means nearly every input points the same way, and points there hard. The platform publishes the Regime as context, with no performance record attached; your view, your chart and your risk rules make the decision.
Each asset class reads the inputs its market publishes:
| Market | What the Regime reads |
|---|---|
| SPX | Breadth, the shape of the VIX futures curve, credit spreads, and the Momentum indicator |
| Equities | Momentum, outperformance against the market, options skew, and dark pool short volume |
| Futures | Momentum, options skew, curve carry, and trader positioning from the COT report, with open interest |
| Crypto | Momentum, funding, options skew where options are listed (BTC and ETH), and open interest |
The single-market versions are built in the lessons that follow: equities, futures and crypto each get their own. SPX comes first because it is the backdrop for every equity trade and because its inputs are the easiest to picture.
10.2.2 How the SPX Regime is built
The SPX Regime reads the equity market from four angles, each one a lesson from earlier in the course.
Breadth. The share of Russell 3000 stocks trading above their short-term moving average, lightly smoothed. A market where most stocks are rising has broad participation; a market held up by a handful of names has thin participation, whatever the index does. Part 6 covers breadth in its market internals lesson.
The VIX futures curve. The gap between spot VIX and the front future, plus the gap between the front and second futures, lightly smoothed. In calm markets the curve slopes upward and both gaps are negative. When short-dated fear runs above longer-dated fear the curve inverts into backwardation, and the input turns bearish. Part 3 covers the curve in the VIX lesson.
Credit. The spread of high-yield bonds over investment-grade bonds, measured as its distance from its own slow moving average. Widening credit, where lenders ask more to hold risky debt, reads bearish; tightening reads bullish.
Momentum. The platform's Momentum indicator on SPX, from the momentum lesson in Part 7, which already lives on a -100 to +100 scale.
The three macro inputs are on different units (a percentage of stocks, VIX points, basis points of spread), so each is standardized first. Each one becomes a z-score against its own trailing window: how far today's reading sits from its recent average, in standard deviations. The VIX curve and credit inputs have their signs flipped, so that stress always reads negative and calm always reads positive.
Each z-score then passes through a squashing function that maps any value into the range -100 to +100. It is close to linear near zero and flattens toward the ends, in the same spirit as a tanh curve: a reading one standard deviation from normal lands at half the scale, and each further standard deviation adds less. A single wild input therefore cannot drag the whole composite to an extreme by itself.
The four inputs are then averaged into one composite, which needs at least three of them present on the day, and the composite gets a short moving average of a few sessions so that one noisy day does not flip the reading.
Two consequences follow from the construction, and both matter for how you read it. First, because each input is squashed before averaging, a composite beyond -70 requires several inputs to be stretched far from normal at the same time: breadth washed out, the VIX curve inverted, credit widening and the trend down, together. That combination is rare, and it is what a market near capitulation looks like. Second, because the inputs are measured against their own recent history, the Regime describes how unusual today is relative to its own recent history. A long, calm uptrend becomes the normal state, and the bullish inputs stop scoring high once calm is the baseline, which helps explain why SPX readings at the top end are rarer than at the bottom.
The SPX page carries each witness on its own tab: VIX term structure, credit and breadth, each with its own regime chart and z-score, plus the oversold panel on the breadth tab. When the Regime moves, those tabs show which of them moved it. Part 6 walks through that page in the SPX dashboard lesson.
10.2.3 Every Regime is read at its extremes and its divergences
Most of the time the Regime sits somewhere between the 70 lines, leaning one way or the other, and that middle reading says little you could not see from the chart. Two situations deserve attention in every market, SPX, a single stock, a futures contract or a perpetual.
Extremes. A reading past plus or minus 70 means the inputs agree strongly. A positive extreme is a move that is established and widely held. A negative extreme is a market where trend, positioning and stress all point down at once. Extremes describe the present: a strong trend can hold an extreme reading for weeks, so the extreme itself is a reason to look closely, and the trigger comes from price.
Divergences. A divergence is price and the Regime telling different stories. Price makes a new high while the Regime makes a lower high, or slides back toward zero and below it. Price makes a new low while the Regime holds above its previous low. The inputs under the Regime (breadth, credit, positioning, skew, funding) often change before the index does, because they measure the participants and the risk appetite behind the move. When price keeps going and the support underneath it thins out, the move is running on less.
Read the number in this order: the level, then the direction of travel, then which input carries it. A reading that has sat beyond 70 for weeks describes a mature move; one that has just crossed zero describes a market whose inputs recently changed sides. On SPX the tabs on the SPX page show which input is doing the work; on single markets the Lens shows each input beside the total.
10.2.4 SPX: low extremes and a bearish divergence
The low extremes line up with the lows in price. Through 2022 the Regime dropped to the -70 line repeatedly as the index ground lower. In October 2023 it reached the -70 zone as SPX pulled back to its 200-day average, right before the rally that ran into 2024. In the spring 2025 selloff it broke through -70, the deepest reading on the chart, at the sharpest drop in price on the chart. The dip in early 2026 took the reading to roughly -55 as price fell back to the 200-day average. A reading that deep needs several inputs stretched to the bearish side together, and each of these sat near a low in price, though in 2022 the low took several visits to the zone before it held.
The top end looks different. In the five years shown the Regime peaks around +60 to +65 and stays below +70, even through the strongest stretches of the uptrend. On SPX the upper extreme is rare, so tops show up as divergences.
The divergence is the right edge of the chart. Price made its highs just above 7,800 in late 2026 and held up there. The Regime, which had peaked around +60 earlier in the advance, rolled over and slid below zero to about -20 while the index sat near its highs. Breadth, the VIX curve, credit and momentum were deteriorating underneath a price that had barely moved.
A divergence is a warning; the timing comes from price. The daily chart supplied it.
The breakout to 7,800 came in one sharp move from the 7,510 area, and from there the index ranged for weeks under a top near 7,820. It poked above that line once, failed to hold it, and then printed lower highs, near 7,770 and then 7,740. The last two sessions on the chart broke down toward 7,540, just above the 7,510 line that marked the old breakout level and the bottom of the range. Below that sit the next support near 7,235 and the 200-day average near 7,310.
Put together, that is how the setup could have been read for a short: a Regime falling while price sat at its highs, then price failing to make a new high and turning lower from the range top. The divergence provided the context, the failed high provided the trigger, and the support lines gave the levels to manage against: the 7,510 area as the first test, with the 7,235 support and the 200-day average below it. The stop belongs above the failed high, which is the level that would say the read was wrong. The price action lesson at the start of this part covers the failed breakout and the support flip in detail.
10.2.5 Low extremes as bottom signals, and the systematic book
The low extremes are the most useful reading the SPX Regime gives. When breadth, volatility and credit all wash out at the same time, the selling has become indiscriminate, and indiscriminate selling tends to cluster near lows. On the chart above, the turns that held came after the Regime had sat at or near -70 and price had started back up.
The same idea is built into the systematic equities portfolio. Its SPX Tactical L/S sleeve engages its long side only after the market reaches an oversold extreme and then reclaims its trend, aiming to capture the sharp rebounds that follow capitulation. That oversold extreme comes from the same breadth washout that feeds the Regime's breadth input, and the oversold panel on the SPX page's breadth tab marks those washouts and reclaims. The book's Equity Momentum sleeve uses the same oversold reclaim as its cue to come back into stocks early after a selloff. The short side of SPX Tactical works the other way round: it engages only after an overbought top breaks down, a close relative of the failed-high read above.
The sequence in both legs matches the discretionary read: an extreme or a divergence first, then price confirming before any position is taken. A low extreme on its own says the market is washed out; it gives no guarantee that the washout is over. In a genuine bear market, such as 2022 on the chart, the reading can visit -70 several times on the way down. The reclaim is what turns the condition into a trade.
10.2.6 The Regime vs 1M Move chart
The Regime vs 1M Move chart sets each market's Regime against what price did over the past month, one dot per market. It is on the homepage tabs and near the top of each Global page, and it is the fastest way to find extremes and divergences across a whole board.
The horizontal axis is the Regime. The vertical axis is the month's return in sigmas: the move divided by the move the market's own volatility implied for a month, so a bond future and a small-cap perpetual can share an axis. Equities and futures scale by implied volatility and crypto by realized volatility, so keep the three charts apart. Dashed lines mark a Regime of 70 either way and 1 and 2 sigma months.
| Quadrant | Regime | 1M move | What it describes |
|---|---|---|---|
| Uptrend confirmed (top right) | Positive | Up | Price is moving with a bullish Regime |
| Pullback in uptrend (bottom right) | Positive | Down | Price has fallen over the month against a bullish Regime |
| Bounce in downtrend (top left) | Negative | Up | Price has risen over the month against a bearish Regime |
| Downtrend confirmed (bottom left) | Negative | Down | Price is moving with a bearish Regime |
Regime vs 1M move
Scatter of each market by its Regime reading against the past month return measured in sigmas.
The futures board on 2026-09-25 showed the range. Corn sat at +97 with a +0.4 sigma month, every input bullish and price grinding higher inside a normal range. Lean hogs read -69 with a -2.7 sigma month, a downtrend in the far corner. Live cattle read -72 while price rose 1.2 sigma, a bounce against a bearish Regime.
The outer corners are the extremes, where a new position joins a move that is already widely held. The two divergence quadrants are where price and the Regime disagree, and those are the dots to open: check which input carries the reading and find the level on the chart that would settle which story is playing out.
10.2.7 Using it in a trade
Write the view down first, then check the Regime. When they agree, the open questions are timing and room left in the move. When they disagree, the plan names the reason (a positioning extreme expected to unwind, a level that broke, an event the inputs cannot see) and the trade starts smaller. An extreme in your direction after a large month suits defined risk or reduced size; a divergence trade waits for a price trigger; a reading near zero gives no directional context, and on a name with listed options the non-directional strategies in Part 9 are worth a look first. While a trade is open, a change in the Regime goes into the journal, and the exit belongs to the stop.
10.2.8 The mistakes people make with it
Trading a crossing of zero or 70 as a signal, when price decides the entry. Calling a reversal off an extreme, when a low extreme in a bear market can repeat several times before the low is in. Shorting a divergence before price fails, when a market can grind higher on a fading Regime for weeks. Reading the total without its inputs, since two markets at the same number can rest on different evidence. Comparing sigma moves across asset classes, since crypto scales by realized volatility and equities and futures by implied.
End of lesson 10.2The Regime: reading extremes and divergences, starting with SPX
10.3 Directional equities
Equities carry the most data per name on the platform: momentum, skew, dark pool, realized and implied volatility, across several hundred liquid names. The equity Regime and the Directional screener narrow that universe to a short list of active markets with interesting positioning. From there, the chart and your own judgment make the decision. Instruments, strikes and sizing are in the Execution lesson.
10.3.1 How the equity Regime is built
The equity Regime reads four inputs, each turned into a score on the same bounded scale before they are combined.
Momentum. The platform's momentum indicator from Part 7, run on the stock's own price. It is already a bounded score, so it enters as it is. This input says which way the trend runs and how firmly.
Momentum relative to the market. The stock's return over a trailing window minus the return of the S&P 500 ETF over the same window. That gap is divided by how much the gap normally moves day to day, so a 10 percent lead means more in a quiet utility than in a volatile chip name. This input asks whether the name is leading the market or riding it. (The S&P 500 ETF itself runs on the other three inputs.)
Skew. The 25-delta, 30-day skew, in the platform's call-minus-put convention, so a negative reading means puts are bid. Each day's skew is compared with its own readings over a trailing window and turned into a z-score: how many standard deviations today sits from that name's normal. Skew adds a read from a different crowd: price shows what has happened, and skew shows what options buyers are paying up to own next. A call wing bid well above its usual level is someone paying for upside, information the price chart cannot show.
Dark pool. The off-exchange short volume ratio, z-scored against its own trailing window the same way. Part 1 covered why this reads contrarian to its name: most short-marked off-exchange volume is dealers selling to fill incoming buy orders, so the platform reads a high ratio as buying pressure. Dark pool adds a footprint of flow. It comes from how trades print, independent of both the price and the options market.
Each z-score is passed through a squashing curve that maps it onto -100 to +100 and flattens near the ends, so one wild day can push a score toward the edge without running it off the scale. Skew and dark pool are then smoothed over a few days to take out single-session noise. The four scores are blended into one, smoothed again, and finally rescaled against the reading's own typical size over a trailing window, so an ordinary day sits well inside the band and readings past plus or minus 70 stay rare. The result is capped at plus or minus 100.
All four inputs are required. A name without a usable options market or without dark pool data has no Regime, and a gap in one input leaves the reading blank for that day.
The sign follows the evidence: positive when the trend, the relative strength, the options market and the flow lean bullish, negative when they lean bearish. A reading past 70 means all four agree firmly. The Lens shows each input beside the total, which is where you check what carries a reading.
10.3.2 The Directional screener
The Directional screener is the equity Regime and its inputs as a sortable board, grouped by sector, with each sector headed by its sector fund's 1-week and 1-month return. A liquidity slider sets the minimum average options volume, and a Stocks/ETFs toggle switches the universe.
The extreme filters are toggles. Each one keeps only the rows at an extreme on that reading, and several active toggles combine, so a row has to pass all of them:
| Filter | Keeps names where |
|---|---|
| Momentum | the momentum score is at or beyond plus or minus 70 |
| Skew | the skew z-score is at or beyond plus or minus 2 |
| Dark pool | the dark pool z-score is at or beyond plus or minus 2 |
| Vol exp | 5-day realized vol is at 1.7 times the 60-day or more |
| Contracting | 5-day realized vol is at 0.6 times the 60-day or less |
| Regime | the Regime is at or beyond plus or minus 70 |
Use it to find where something is happening. A Regime extreme says all four inputs agree firmly on a name. A skew or dark pool extreme says one crowd has leaned unusually hard. What the screener leaves open is what to do about any of it, and that is where a systematic approach to single stocks runs out of road. Single names gap on earnings, on guidance and on headlines that no positioning input sees coming, and the same extreme reading means different things on a stock coming off a base and on one six months into a run. The screener hands you candidates. The chart, the levels and your discretion decide which of them become trades.
10.3.2.1 Skew and dark pool on their own
The single-input filters are the sharper end of the screener. A skew extreme gives you a fork: ride the demand when you think the move has more to give, or fade it when you think it is exhaustion. The chart settles which.
Dark pool works the same way from a different source. An extreme z-score says the hidden flow in a name is unusually one-sided, which is context for which side of the name to be on.
10.3.3 Stocks drift up
One fact shapes how to read every equity extreme. Stocks carry a positive drift: research finds the equity market rises over time, so the long side has the wind behind it. That makes the two ends of the Regime scale unequal.
A negative extreme on a strong name, one with a rising long-term trend and a history of leading its sector, is more often a pullback that has run its course than the start of a collapse. Trend, relative strength, skew and flow all lean bearish at once at the moments when sellers are most exhausted, so these readings are a good place to look for early candidates for the next momentum leg. The trade still waits for the chart: a level holding, a reclaim, a base.
A positive extreme is a weaker fade. Strong names can sit past +70 for weeks while the drift and the trend carry them, and shorting into that fights both. A positive extreme is a reason to stop adding, tighten up, or wait for a pullback. Shorts belong to names where the chart has already broken and the Regime confirms it.
10.3.4 PLTR: Regime and price at extremes together
At the lows, momentum was pinned near -70 and the Regime sat at its floor while price sat on the 120 band. Every input leaned bearish, and the stock was at a level that had held for months. On a name that had led the market for two years, that is the drift argument in practice: a candidate for the next leg, waiting for the chart. The confirmation was messy. Price broke below the band in late June, which would have stopped out an early buyer, and then reclaimed it in July. The reclaim of the band was the trigger, and the reading turning up from -70 was the context behind it.
At the highs, momentum reached +70 and the Regime hit its cap as price returned to the prior high near 207. Trend and positioning were fully lined up, into resistance: the place to tighten or take profits on a long from the lows. Price did turn lower from the band, and a fade there was still the lower-quality trade for the reasons above. The next long setup is the next level price comes back to.
10.3.5 GME: the extreme case
Momentum and the Regime both deep in the extreme zone, at a level the market had respected for two years, is the extreme case. The readings described the conditions correctly: every input leaned hard bearish. The opportunity came from those conditions meeting a level, and the support band defined where the trade was wrong. The rally that followed took price from about 18 into the high 20s in roughly two months. A name like GME moves far more than a typical stock in both directions, so its size and stop follow its volatility (Execution lesson).
10.3.6 SPOT: an extreme that starts a watchlist
SPOT is the trend-following case. The Regime reaching its cap as price leaves a range that held all year means trend, relative strength, options and flow all turned bullish at once. The breakout candle is the worst-placed entry, since the gap is where the move is most stretched against its own volatility. The same workflow comes back in the futures and crypto lessons:
- The Regime extreme puts the name on the watchlist.
- The chart marks where you want to be filled: the demand zone around 530-540, where the breakout consolidated before it went.
- A resting limit order at the zone, or a price alert that tells you when price gets there so you can read the candles before committing.
- The stop goes beyond the zone, where the breakout would be invalidated.
SPOT came back toward the zone and continued to around 590. Had it kept running without a pullback, the limit order would have stayed unfilled, which is an acceptable cost. A missed trade costs nothing; a chased one sets the stop far from the entry.
10.3.7 Volatility expansion and contraction
The Vol exp and Contracting filters answer a different question from the directional ones. They compare a name's realized volatility over the past week with its realized volatility over roughly the past three months, and they say nothing about direction.
Vol exp (5-day realized vol at 1.7 times the 60-day or more) finds names that woke up. Something is moving them: a breakout, a breakdown, news, a gap. For a directional trader this is a list of moves already under way. Pair it with the Regime: an expansion on a name whose Regime is building toward an extreme is a move with the positioning behind it, and an expansion on a name with a neutral Regime deserves more doubt, since the move has no positioning behind it yet.
Contracting (5-day realized vol at 0.6 times the 60-day or less) finds names that went quiet. A stock coiling into a tight range after a trend often resolves with a larger move, and the chart tells you where the edges of that range are. Contraction also matters for the instrument you pick. Implied volatility tends to follow realized volatility down, so a coiled name is often the cheaper moment to buy options for the breakout, before the move reprices them.
Both filters combine with the others. Contracting plus a Regime extreme finds names where positioning lines up while price sits still, coiled with the evidence behind it. Vol exp plus Skew finds names moving hard while one options wing is bid.
End of lesson 10.3Directional equities
10.4 Directional Futures trading
Futures sit at the discretionary end of the book. The trade comes from reading positioning and judging where in a move a market stands, and the platform's job is to put that positioning on one screen. Every futures market also has its options data on the platform, so the volatility risk premium, directional options and calendar strategies from Part 9 apply here too. This lesson covers the directional, positioning-driven way to trade them.
10.4.1 Why plain trend following got harder
For decades the standard directional futures strategy was trend following: buy what has risen, sell what has fallen, across dozens of markets, and let a few large trends pay for many small losses. Research finds that this strategy has delivered much weaker returns over roughly the last two decades than in the decades before: lower risk-adjusted returns, long flat stretches, and smaller payoffs in the equity selloffs it used to profit from. Studies link the decline to more capital running similar signals, an industry shift toward slower models as assets grew, and markets that reverse more often around policy decisions.
The practical lesson is that a trend signal on its own is a crowded, thin edge. What still separates one entry from another is where the crowd stands inside the trend. A trend that the speculators have not yet joined has fuel left. A trend that every speculator already holds, with the options market paying up in the same direction, has mostly spent it. The futures Regime is built to measure exactly that.
10.4.2 The indicators, one by one
Each reading below has its own lesson earlier in the course. Here is what each one tells a directional trader and where it appears.
Momentum. The platform's momentum reading from Part 7, a volatility-scaled trend score. It says which way the trend runs and how firmly. Past plus or minus 10 the platform treats the market as trending, inside that band as chop. It is the only input that is pure price.
25-delta skew z-score. Skew is the 25-delta call's implied volatility minus the 25-delta put's, so a negative number means puts are bid. The z-score places today's skew against its own trailing year. A reading past plus 2 says the options market is paying an unusual premium for upside, past minus 2 for downside. It is a read on what the options crowd is willing to pay for, and it moves daily.
Curve and carry. The Contango reading is the annualized slope from the front contract to the one about a year out, with its own z-score. Backwardation (deferred contracts cheaper than the front) pays a long holder as contracts roll toward spot, and usually signals a tight physical market. Contango pays the short. Carry means different things in different markets, as Part 4 showed: in the grains and energies it is a statement about physical supply, in financial futures it is mostly interest rates.
COT. The weekly Commitments of Traders report splits open interest by trader group. Commercials are the producers and users who hedge; they tend to buy weakness and sell strength. Large speculators are the funds that mostly follow trends. The platform's COT Index scales each group's net position from 0 to 100 against its own trailing 52 weeks: 100 is the most net long that group has been all year, 0 the most net short, with 90 and 10 marked as extremes. The Positioning Index goes one step further and combines every trader category into one 0 to 100 reading that leans with the commercials and against the speculators, so a high reading means commercials are long and speculators short. Both update once a week, three business days after the Tuesday the report describes.
Open interest. The number of open contracts. It has no direction of its own, so it works as a measure of how much money stands behind the other readings. A positioning extreme with open interest elevated is a bigger, more fragile crowd than the same extreme in a thin market.
Valuation. The contract's price divided by a benchmark (the dollar index for currencies and gold, the 30-year bond for equity indices, gold for most commodities), as a z-score against its own year. Price ratios trend, so a reading past 2 is a state the market sits in for a while, and it gets a light weight.
Seasonality. The average detrended path through the calendar over 15 years, with the average forward return for the next 30, 60 and 90 days. It earns weight only where there is a physical cause: planting and harvest in the grains, demand cycles in the energies. Everywhere else a seasonal average is mostly the residue of a few years that happened to share a month.
10.4.3 Where the readings live
The Lens plots every market on one standard-deviation scale. Its Regime view shows each market's Regime as a ringed dot with its inputs beside it (momentum, skew, carry, speculator COT, open interest strength, and valuation for context), so you can see in one row which input is stretching a market and whether the inputs agree.
The screener is the same information as a sortable table: the three COT groups with their weekly changes, valuation, the 30-day seasonal, skew and its z-score, the curve z-score, momentum and the Regime. Its four extreme filters (COT, valuation, seasonality, skew) combine, so switching on two of them leaves only markets stretched on both.
10.4.4 How the futures Regime is built
The futures Regime is one daily reading per market from -100 to +100, built to answer one question: how far trend and positioning lean the same way, measured against that market's own history. It has four inputs and a scaler.
- Momentum, the daily trend score described above, used as it is.
- Skew. The 30-day 25-delta skew is ranked against its own trailing year and passed through a squashing curve that maps it to a bounded score of plus or minus 100. The curve is close to linear for ordinary readings and flattens in the tails, so one panicked session cannot dominate the total. A short average then smooths day-to-day quote noise.
- Carry, for commodities only (metals, energies, grains, meats, softs). The annualized slope from the front contract to one roughly a year out is sign-flipped so that backwardation reads bullish, then ranked and squashed the same way as skew. Equity indices, bonds, currencies and the CME crypto contracts leave carry out, because their curves mostly reflect interest rates.
- Speculator positioning. This input reads the speculative groups in the COT report (managed money, leveraged funds, asset managers and other reportables). For each group it checks several angles against the trailing year: the size of the long book, the size of the short book, how many traders sit on each side, and the net position, both as its place in the year's range and as a z-score. Each angle that sits at the top or bottom of its range counts as one stretched vote, long or short. The votes are weighted by how much of the open interest the group's net position represents and by how many of its traders crowd onto one side, then combined across groups. A positive score means speculators are stretched long. The score is dated to the day the report is published, so the Regime never sees a report before the market did.
The inputs are averaged with equal standing. That average is then multiplied by an open interest factor: when open interest over the past few weeks is elevated against its own history the factor sits near full strength, and when it is low the reading is damped. The factor can shrink the reading and never flips its sign.
Last, the result is smoothed over a few sessions and divided by its own typical size over the trailing year, then capped at plus or minus 100. That final step is why the scale means the same thing in every market: a reading of 70 says trend and positioning lean together far harder than they usually do in this particular market, whether it is a quiet bond future or a volatile soft.
Two points follow. The Regime's COT input leans with the speculators, so it rises as the crowd piles in; the Positioning Index leans against them. Read the Regime for how crowded the trend is and the Positioning Index for contrarian pressure. And because the COT input moves once a week, a step in the Regime on a Friday is usually the new report.
10.4.5 At the extremes: mostly mean reversion
When the Regime reaches plus or minus 70, momentum is strong, speculators are stretched in the same direction, the options market is usually paying up that way, and in commodities the curve often agrees. Almost everyone who wants the trade already has it. For most markets most of the time, the default at that point is to look for a fade at a higher-timeframe level: a weekly supply or demand zone, a prior swing, a level that held on the weekly chart. The Regime says the crowd is stretched; the level says where it is likely to run out of buyers or sellers. Neither is a trigger alone. Trends can stay at an extreme for months, so the fade needs the level and a price reaction there.
10.4.5.1 Sugar into weekly supply
Sugar spent most of 2025 and the first half of 2026 grinding lower, bottoming in the 13s and 14s. From midsummer it rallied hard, and the rally ran straight into an old weekly supply zone.
The platform's data shows each input carrying the move. By mid-October momentum, speculator COT and carry all sat near the top of their scales and skew was bid toward calls, and the Regime printed 99. Price then failed under the zone and the reading started to come off.
10.4.5.2 Corn: a thin push with no follow-through
Corn is the same pattern on a daily chart, with an extra clue in how the extreme was reached.
A move that reaches an extreme on light participation is the weakest kind of extreme. Nobody new arrived to keep it going, so once the level held the crowd had only one way to go. The platform's corn Regime read 99 in mid-September while price sat under 544, and had dropped below minus 25 by late October.
10.4.5.3 Soybean meal: an over and under at the extreme
Soybean meal shows the mean-reversion trade with a market structure break as the trigger.
The sequence is the template for a fade at an extreme. The Regime said the crowd was fully long. The break below the swing low said the trend had stopped making higher lows. The failed push back over the 368 line, followed by a close back under it, gave the entry with a stop just above the high of that push. Each piece alone was weak. Together they turned a stretched market into a trade with a defined risk.
10.4.5.4 Cocoa: the same idea from the short side
Cocoa in early June was a mean-reversion trade at a positioning extreme in the other direction. After a long decline, large speculators held one of the largest net shorts of their multi-year range and had started to cover.
The Regime marked the washout long before there was anything to buy. It sat at the low extreme in late January and February, and the trade only came together in June, when speculators started covering and price reclaimed its range. That gap is normal. An extreme tells you which markets to keep on the list, and the setup on the chart can take days, weeks or months to arrive. The work is patience: track the futures board, keep the stretched markets on a watchlist, and wait for price to confirm before committing.
10.4.6 Trend following inside this approach
The trend trades still exist; they come at a different moment. In-trend entries often appear when the Regime falls from an extreme back toward its midpoint while the trend on the higher timeframe holds. The crowd resets: some speculators take profit, skew cools, short-term momentum dips. Then price pulls back into a higher-timeframe level, and the trend resumes from there with fresh room to run.
10.4.6.1 Dollar index: July extreme, August reset, weekly demand
The platform's data shows the reset in detail. In July every input leaned long and the Regime was pinned at 100: a crowded trend, the wrong place to start a new long. Through August speculator positioning came off its extreme, skew briefly flipped toward puts and momentum turned down, and the Regime dropped to about 10 while price fell into the 98s. The weekly trend was intact, the crowd had thinned, and price was sitting in demand. That combination is the trend-following entry: a stop under the zone and a target back at the highs and beyond.
10.4.6.2 Euro: an early short
The euro in mid-June is the early version of the same idea on the short side: positioning that had barely started to lean.
10.4.7 Risk, sizing and capital
Risk per trade runs about 1 to 4 percent of the account, scaled to conviction: a level, a clear Regime read and agreeing inputs earn the top of the range, a thinner read the bottom. Fades at an extreme sit lower in the range than in-trend entries after a reset, because the crowd can stay stretched longer than the account can wait.
The stop is an ATR trail: one and a half times the daily ATR, plotted from the daily close and ratcheted behind the trade. The exit needs a daily close through the line; an intraday wick through it does not count. Size by dollar risk to that stop, never by margin, since a contract's multiplier makes margin a poor guide to exposure. A full basket across indices, bonds, currencies, metals, energies and grains at real size to the stop needs more capital than the options strategies, so a smaller account runs fewer markets and accepts less diversification. Order types, contract choice and the micro contracts are covered in the Execution lesson.
End of lesson 10.4Directional Futures trading
10.5 Directional crypto
Crypto runs the same game as futures: momentum watched alongside positioning, and a judgment about which part of the trend a market is in. The difference is the speed of the positioning data. In futures the COT report arrives once a week; in crypto, funding, open interest and liquidations update every day, so the crowd's leverage is visible almost as it builds.
The approach here trades the majors and the established large caps. The platform carries data on a long tail of smaller coins, and those can be traded, but thin liquidity, fast reversals and venue risk make them far more work to watch. Everything in this lesson gets stricter the further down the market-cap scale a coin sits.
10.5.1 How the crypto Regime is built
The crypto Regime puts trend and positioning on one scale from -100 to +100 for every perpetual the platform tracks. It has four inputs.
Trend. The trend input compares fast and slow moving averages of price at three different speeds, from a few weeks to several months. Each gap is divided by price and by the coin's recent volatility, so a 10 percent move counts for less on a coin that swings 10 percent in a normal week. The three speeds are averaged and blended with the directional intensity reading from the momentum indicator (Part 7), and each piece is scaled against its own typical size over the trailing year, so a quiet coin and a wild one land on comparable numbers.
Funding. The funding rate is compared with its own trailing year: how far today's rate sits from the year's average, in standard deviations. That distance is passed through a curve that flattens the tails, so one freak funding print moves the reading only a limited amount, and then smoothed over a few days. Positive funding, longs paying shorts, pushes the Regime up. That is deliberate: the Regime describes which way the crowd leans, and a crowd paying up to be long is a bullish lean.
Skew, on BTC and ETH only. These two coins have a listed options market with usable quotes, so their 30-day skew, stated as call minus put, gets the same treatment as funding: distance from its trailing year, flattened at the tails, smoothed. Calls bid over puts pushes the reading up. Every other coin is read on trend and funding alone.
Open interest. Open interest has no direction of its own (Part 5), so it never pushes the reading up or down by itself. It is measured in coins, so a rally alone leaves it unchanged, scored against its own trailing year the same way, and then used as a multiplier. Heavy open interest amplifies whatever trend and funding say, because more leveraged money stands behind the lean. Thin open interest damps it.
The directional inputs (trend and funding, plus skew on BTC and ETH) are averaged, scaled up to the Regime's range, multiplied by the open interest factor, averaged over the last few days so a single session cannot whip the number around, and capped at plus and minus 100. The trailing windows count calendar days, because crypto trades every day of the year. Plus or minus 70 is the extreme zone, the same as in every other asset class.
Two consequences matter for trading it. First, the Regime reads the crowd with the trend: a coin falling hard while shorts pay heavy funding scores deeply negative, even though that exact configuration is the squeeze setup in the HYPE example below. The Regime tells you the crowd is short and levered; whether that is a trend to join or a crowd to fade is the trader's call. Second, the open interest multiplier means two coins with the same trend and funding can sit at different readings, and the Lens shows which input carries each one.
10.5.2 The lens and the screener
The crypto lens and screener organize the whole market the way their futures counterparts do. The lens plots each coin on a standard-deviation scale with a dot per reading, and its Regime view puts the total beside the inputs that build it. The Regime vs 1M Move chart on the crypto Global page sorts the market into the four quadrants from the Regime lesson, with the month's move measured against realized volatility.
10.5.3 BTC: the Regime at the extremes, against the levels
The Regime describes conditions. The levels on the chart decide where a trade goes on. BTC over 2026 shows the two working together at both ends of the scale.
The bottom. BTC broke down from its 86k to 94k range at the start of February 2026. The platform's BTC Regime reached -75 on 9 February, as the first flush wicked toward 60k. Price spent the spring in a range, failed at the 82k line in May, and broke lower again in June. Through June the Regime sat between about -55 and -73, and the low close of the move came on 30 June at about 58.6k, right on the line along the top of the 2024 demand zone. The 2024 lows were where buyers had last stepped in size, and price came back to them with every input in the Regime pointing down: trend negative, funding negative, the crowd short and still pressing. That combination, an extreme reading into a level with history, is where a short trade is late and a long trade has a defined place to put a stop.
The top. From mid-August price broke out of the base, from about 64k to 78k in a week, and the Regime went from below zero to about +72 by 26 August. The second leg in September carried price to an 86.6k close on 21 September, into the bottom of the supply zone left by the January breakdown, and the Regime peaked at +74 two days later. The supply zone is where sellers who were trapped in the 86k to 94k range get their exit at breakeven, so price meeting it with the crowd fully long and funding positive is the mirror of the June low: a long is late, and the levels give a short a place to lean against.
Since then. Price drifted lower through October and lost the 82k line, the old 2025 support that capped the May rally, then pushed below 80k. The BTC Regime closed 2 November at about +16, its lowest since the August breakout, and momentum on the chart is below zero. The next levels below are the 74.5k base of the August breakout and the 200-day average near 72.5k.
The general point carries to any coin. An extreme Regime reading on its own describes the present, and extreme readings can persist for weeks in a strong trend. An extreme reading that arrives at a level with a clear history, a demand zone the market last bounced from or a supply zone it last broke down from, is where the trend trade has the least room left and a counter-trade has the clearest stop. The zone-drawing method is in the technical analysis part (Part 8).
10.5.4 Positioning against the trend: HYPE
The other end of the playbook is a crowd leaning hard the wrong way. HYPE in the middle of May is the setup in its clearest form.
In mid-May HYPE was consolidating under resistance while holding its moving averages as support, longs had just been shaken out on large liquidations after that week's news, and funding was low.
Open interest was building into the plus-two zone while funding sat around minus two: a lot of new positions, and the crowd paying to be short.
The chart was basing under resistance and holding its moving averages. The positioning said the crowd was short and levered, and the liquidation flush had already cleared the weak longs. The trade was against the crowd, and the technical base gave it a level to lean on and a place for the stop.
Order book depth, computed from spot, is the last check. Spot is where real buyers and sellers transact, while perps carry the leverage (Part 5). Depth leaning the way you want to trade is confirmation; depth leaning against you is a reason to size down or skip.
10.5.5 WLD: a worked trade start to finish
WLD in May 2026 ran the HYPE playbook end to end, and it started from the bottom of the Regime scale.
A coin coming off an extreme low Regime is a candidate for a long: the selling is crowded and the trend has nowhere further to lean. The reading alone is a watchlist entry. The trade waits for price to show strength, a reclaim or a break above a level that capped it, and that is what WLD did in late May. Momentum flipped positive, open interest pushed above plus two, and funding was strongly negative: the trend turning while the crowd sat short and levered.
The entry and exit came from the chart. A zone near 0.28 had capped price through April and May. In late May price reclaimed it, above the moving averages, and that reclaim was the entry. The coin ran to about 0.72 in mid-June, and the position rode the ATR trailing stop until a daily close back through it, around 0.53 in late June.
10.5.6 Options on BTC and ETH
BTC and ETH carry full options data, so every options strategy from Part 9 applies to them as it does to equities: the volatility risk premium, calendars, and the directional structures from the Execution lesson, financed spreads and risk reversals included, since the skew is quoted.
10.5.7 The systematic version: Crypto L/S
Everything above is discretionary. For traders who want the same two ingredients, trend plus derivatives data, as a set of rules, the systematic section publishes Crypto L/S. It is a daily long/short book across the most liquid crypto perpetuals: long the coins in the strongest trends, short the weakest, with derivatives positioning data refining which trends it trusts and how much risk it takes. The book is volatility-targeted, with a choice of 20, 30 or 40 percent annualized volatility, and rebalanced every day. Its backtest covers about five years of data, a shorter record than the equities book, and it deserves the caution that implies. The strategy page carries the full backtest.
10.5.8 Risk and the stop
Risk runs about 1 to 3 percent per trade, scaled to conviction, with the same ATR trailing stop as futures: one and a half times the daily ATR, trailed behind the position, exited only on a daily close through the line, and only ever moved in the trade's favor. Crypto wicks through everything, so the close rule matters more here than anywhere else. Size by dollar risk to that stop, and keep the position small enough that the stop closes a losing trade long before price nears the exchange's liquidation level. Placing the orders, venues, and the choice between perp, spot and options structures are in the Execution lesson.
End of lesson 10.5Directional crypto
10.6 Execution: futures, stocks and options structures
The earlier lessons end with a view: a direction, a level on the chart, and a read of the trend and the crowd behind it. This lesson turns that view into a position: the linear instrument first, then option structures built on skew, then three Overview tools that help with strikes and timing, and finally sizing.
10.6.1 Linear first: futures, stocks and perps
The simplest expression of a directional view is linear: long or short the future, the stock or the perpetual. The P&L moves one for one with price, with no implied volatility to overpay for, no theta and no strike to choose. Most directional trades in this part are best expressed this way, and options earn their place only when they add something specific.
A linear trade has three parts, and all three come from the chart.
- Entry at the level. The price action read gives the level: the reclaimed range high, the resistance that flipped to support, the trendline break. The entry sits at or near it, on the close that confirms it or on the retest.
- Stop beyond the level, or the ATR trail. The stop goes where the setup is wrong: below the reclaimed zone for a long, above the broken support for a short. Once the trade moves away from entry, the ATR trailing stop takes over.
- Size by dollar risk. Decide how much of the account the trade may lose, divide by the distance from entry to stop, and that gives the size.
The ATR trailing stop is the same rule across futures, stocks and crypto. Take 1.5 times the daily ATR, plot it as a line behind price from the daily close, and let it ratchet in your favour as the trade works. Three rules govern it:
- Exit on a daily close through the line. An intraday wick through it is noise, and crypto in particular wicks through everything. A daily close beyond it says the move you were riding has changed.
- Trail only in your favour. The line moves up under a long and down over a short, and it never moves back to give a losing trade more room. Loosening a stop is how a planned small loss turns into a large one.
- Size to the stop. The dollar distance to the line times the position is the risk.
A futures multiplier makes margin a poor proxy for exposure, and a perp venue offers far more leverage than any sane position needs. Size by the dollar loss to the stop, never by margin or by venue leverage, and keep perp positions small enough that ordinary volatility never brings price near liquidation, so the ATR stop is always what closes a losing trade. The arithmetic also shows when a market is too big for the account. A crude oil contract carries a multiplier of 1,000, so a stop two dollars away is two thousand dollars of risk on one contract. On a 100,000 dollar account at 1 percent risk, that trade does not fit even at one contract, and the right answer is a smaller contract, a closer stop that still respects the level, or a pass.
Risk per trade follows the asset-class lessons: about 1 to 4 percent on futures and 1 to 3 percent on crypto, scaled to conviction, and the same dollar-risk logic for stocks. Part 11 covers the sizing machinery underneath these numbers.
10.6.2 Options and skew
Options add one thing a linear position cannot: the ability to buy one part of the distribution and sell another, at prices the options market sets. Skew tells you which part is expensive. Throughout, skew is CALL minus PUT: a negative reading means puts are bid, a positive reading means calls are bid, and the skew z-score measures today's reading against its own history, with 2 as the extreme threshold (Part 3 covers skew in full).
An extreme skew reading gives you a fork. Either the crowd paying up for one wing is right and the move continues, or the demand marks exhaustion and price reverts. The structure follows the side of the fork you take.
10.6.2.1 With the trend: call and put spreads when skew is elevated
When the view is that the move has further to go and skew is elevated in that direction, the outright option is expensive because the crowd has bid that wing. A spread fixes the price problem. Buy the option near the money and sell the further out-of-the-money option on the same side, the rich wing the skew flagged. For an up move with calls bid, that is a call spread; for a down move with puts bid, a put spread. The short leg finances the long one, so the cost and the maximum loss both fall, and the debit you pay is the most you can lose.
The trade-off is a capped payoff. USO in 2026 shows it. Skew screamed call demand, and the chart said the same thing independently.
Call spreads bought on that reclaim in February were profitable, and the parabolic continuation ran straight through the short call. The spread captured a defined chunk of the move and gave up the tail. That is the deal when you sell a wing to finance the trade: a cheaper, defined-risk position that cannot catch a parabola. Use the spread when the skew is steep enough that selling the wing meaningfully cheapens the long option. When skew is flat, the plain long call or put keeps the whole tail for a small extra cost.
10.6.2.2 Mean reversion: the 25-delta risk reversal
When the view is that the extreme is exhaustion, the structure is a risk reversal, at roughly 25 delta on each side. It fits when the leg you sell carries much richer implied volatility than the leg you buy. Sell the bid wing, the one the crowd has paid up for, and use the premium to buy the option on the other side. With the strikes around 25 delta, the structure often goes on for close to zero cost.
Two examples, in call-minus-put terms:
- Calls bid to an extreme (positive skew z-score past 2), view that the rally is exhausted. Sell the 25-delta call, buy the 25-delta put. The position is short, financed by the call buyers' overpayment.
- Puts bid to an extreme (negative skew z-score past minus 2), view that the panic is overdone. Sell the 25-delta put, buy the 25-delta call. The position is long, financed by the put buyers' overpayment.
Compare the implied volatility of the two strikes before placing it. The edge in the structure is the gap between them; when the two wings price at similar IV, there is nothing to harvest and a linear position does the same job more cleanly.
The short leg is naked, so the risk is open-ended in the direction you are fading. A risk reversal always carries a stop, with no exceptions: the same ATR trailing stop as the linear trades, placed on the underlying's price, and the whole structure comes off on a daily close through the line. Size it by the dollar loss to that stop, the way a future is sized, never by the near-zero premium it costs to open.
10.6.3 The TradingRiot distribution: picking strikes and targets
Each symbol's Overview carries the momentum readings and, for every equity, every futures market, and BTC and ETH, a Price distribution chart. It draws two curves over where price may sit at a chosen tenor (10 to 90 days on equities and futures, 7 to 60 days on crypto):
- Implied (blue). The market's view: a lognormal curve at the 30-day implied volatility, centred so its mean is today's price. It is what the option prices say.
- TradingRiot (teal). The platform's own view. Its width blends implied volatility with recent realized volatility, it uses ex-earnings IV when an earnings print falls inside the tenor, and its tails are heavier than a plain lognormal. Its centre is tilted off today's price by the 25-delta skew z-score, the post-earnings drift signal and the momentum reading, a small shift that grows only when those readings sit at extremes. How the inputs are weighted is proprietary.
A vertical line marks today's price, and the footer prints the 1, 2 and 3 sigma price ranges for both curves, plus the move each curve implies over the tenor in percent.
Read it in three steps when choosing strikes.
The gap between the curves is the price of options. When implied volatility runs well above realized, the teal curve is narrower than the blue one: the options market is pricing a bigger move than recent trading has delivered, so long options are expensive and a spread, or selling the rich wing, suits the trade better than a naked long option. When the two curves are close in width, buying the option outright is fairly priced.
The tilt shows where the model leans. A teal curve shifted toward your direction says skew, drift and momentum agree with the view. A teal curve leaning against your view is the signals disagreeing with you, and that is worth one more look at the chart before any order.
The bands give strikes and targets. The TradingRiot 1 sigma range is a realistic zone for the move over the tenor, so it suits the short strike of a spread and the first profit target of a linear trade. The 2 sigma level marks a stretch move, a sensible place for the far strike when you want room, and a strike beyond it is a lottery ticket. For a risk reversal, the wing where the blue curve carries clearly more weight than the teal is the one the market is overpaying for, which is the leg to sell. A target set beyond the model's 2 sigma range needs a reason the chart can name.
On SPOT above, all three steps point the same way. The curves are close in width, so a long call is fairly priced and needs no financing; the tilt agrees with a bullish view; and the TradingRiot 1 sigma top near 676 is a realistic 30-day target, with the 2 sigma level near 762 as the stretch. A bullish trader could buy a call near the money and, if a spread is wanted to cut cost, sell the strike near the 1 sigma top.
The Strike Selector runs the same model strike by strike. It takes a symbol, a direction (bullish or bearish), an expiration and a mode (Single, a naked option, or Spread, which adds a short strike on the same side to cap the cost), then prints the distribution inputs and a table of strikes with the model's probability of finishing in the money, the edge of that probability over the market's delta, the expected value, the odds, a Kelly percentage and a rating.
10.6.4 Volatility Regime and Directional Intensity: confluence for price action
Two more series sit side by side under the momentum readings on each symbol's Overview. Both are built from daily returns over a trailing 21-day window, and each chart draws its mean and a one standard deviation band over the visible period, so a reading can be judged against the symbol's own history.
Volatility Regime is the average absolute daily move over the window minus its own moving average. Positive means daily ranges are larger than their recent norm and volatility is expanding; negative means ranges are shrinking and volatility is compressing. Zero crossings mark the transitions. It is read in percent of price per day, so it compares a market with its own past, and the levels on one symbol say nothing about another.
Directional Intensity is the average daily return over the window divided by the average absolute daily move. It runs from minus 1 (every day down) to plus 1 (every day up). The reading is the share of the total distance travelled that turned into net progress: at 0.3, price made net headway equal to 30 percent of everything it moved. Near zero, the market moved plenty and went nowhere, which is chop.
Neither series is a signal on its own, and both look back only 21 sessions, so they describe the recent tape and lag turns. Their use is confirming or filtering a setup the chart has already given you.
A breakout. Price breaks a range high that capped it for weeks. The setup has the best backing when Volatility Regime has been negative, the market coiled, and turns up through zero as the break happens, while Directional Intensity rises from around zero toward its upper band. That is a market leaving compression with one-way participation, the textbook start of a trend, and it suits a linear entry on the ATR trail or a long call. Research finds volatility clusters and mean-reverts, so compression tends to end in expansion, and the Volatility Regime shows when it has begun. A breakout where Directional Intensity stays near zero is a break inside chop, and it has a higher chance of falling back into the range. SPOT, the trend example from the equities lesson, shows the confirming version at the right edge of both charts: price clearing its August and September highs, Directional Intensity at a one-year high, and Volatility Regime turning up through zero. Read the two together, since expansion has no direction of its own: SPOT's February spike came with Directional Intensity negative and price falling, this one with price rising.
A fade into a level. Price runs into a well-defined resistance after a strong move. The fade has the best backing when Directional Intensity is rolling down from an extreme toward zero, the one-sided buying running out, and Volatility Regime is past its upper band and turning lower, the expansion exhausting itself. That combination at a level is the setting for the mean reversion structure: a short against the level with the stop beyond it, or the 25-delta risk reversal when skew makes the wing you sell rich. A fade into a market where Directional Intensity is still climbing and volatility is still expanding is standing in front of a trend that has not slowed, and it is the trade to skip.
Holding a trend. A position on the ATR trail with Directional Intensity holding above zero and Volatility Regime calm is a trend doing its work. A spike in Volatility Regime with Directional Intensity collapsing is worth a journal note. The exit still belongs to the trailing stop.
10.6.5 Sizing and management
Long options and debit spreads. Risk 1 to 2 percent of the account per position. The debit is the most the trade can lose, so the size of the debit is the size of the risk. These trades are convex, small losses most of the time and an occasional large win, which is why the size stays small.
Risk reversals. Sized by dollar risk to the ATR stop on the underlying, smaller than a linear trade with the same view, because a gap can jump the stop on the naked leg.
Linear trades. Sized by dollar risk to the stop: 1 to 4 percent on futures, 1 to 3 percent on crypto, scaled to conviction.
Stops. A stop is either trailed by ATR or set at price action structure. The ATR trail ratchets behind the trade and exits on a daily close through the line. A structure stop sits beyond the level that defines the trade (the zone, the swing low or high, the broken range) and moves only when price builds a new structure, such as a higher low in an uptrend. Pick one per trade and size to it.
Taking profit on long options. A long option decays every day, faster as expiry nears. When a position is in profit and the move stalls, theta hands the gains back, so take profits as the trade works or trail it, and avoid sitting on a winner into expiry. A linear position on the ATR trail is managed the other way: it is closed only by the stop, because the uncapped upside is the point.
End of lesson 10.6Execution: futures, stocks and options structures