Entry, Stops, Position Sizing, and a Complete Trading Plan

28.1 Analytical Bias, Trading Opportunity, and Entry Trigger

Analytical bias, trading opportunity, and entry trigger progressively narrow the decision. Bias asks which direction or market state the evidence favors. The overall market, trend or range context, structural phase, relative strength, and price-volume results eliminate some explanations, but price can still develop in different ways. For example, improving demand within a possible accumulation makes an upward departure worth preparing for, yet price may remain sideways, test again, or invalidate accumulation. Bias therefore determines what to seek first, without directly generating an order.

A trading opportunity constrains direction to a location with boundaries. A post-Spring test, LPS candidate, or retest after a valid breakout differs from the range center because nearby structural boundaries can test the current interpretation. If the expected response appears, entry-to-protection distance can be calculated; if it fails to appear or the boundary is validly breached, the plan can be canceled. There must also be remaining room from expected execution to a conservative target and feasible execution under costs, liquidity, and portfolio capacity. Even a correct direction offers no tolerable opportunity if price is already near supply ahead, invalidation is too distant, or exit is unreliable.

An entry trigger is new information within the opportunity zone distinguishing mere arrival from a result achieved by the expected side. A long trigger may be defined as an upward-shifting close after downward movement stops, a break of a local swing high, or aggressive buying progress after repeated selling fails to depress price. Reverse this for shorts. Stopping only shows that the old direction temporarily failed to expand its result; a subsequent turn eliminates some explanations of a brief pause. Rules must specify the observation scale, completion condition, and order type. Strength appears cannot establish when an order becomes valid. After triggering, the actual executable price changes stop distance, target room, and quantity, so a trigger grants execution eligibility rather than an unconditional instruction to fill.

The three levels have different vetoes. Wait if bullish but not yet at a bounded location; a reached location without directional change remains untriggered; even after a trigger, abandon the trade if a gap moves execution far from the boundary, target room is consumed, or portfolio capacity is insufficient. Reviews can then distinguish directional judgment, location selection, signal confirmation, and execution constraints instead of using an eventual rise to justify poor location or an over-budget order.

28.2 Tests, LPS/LPSY, Breakout Retests, and Trend Continuation

The Spring and Its Test

A Spring is a downward penetration of support in a possible accumulation range followed by a rapid return inside. Its context should include a stopped decline, recognizable range boundaries, and no sustained acceptance below the range. The Spring itself can be an early participation point, but a limit fill while the probe is still falling cannot be treated as a completed Spring in advance. A more robust trigger comes when price recovers support and exceeds local resistance. If there is a separate test, examine whether its decline and trading activity contract, support holds, and price strengthens again.

Cancel test participation if price breaks the Spring low again and remains below it, the test expands downward with clear supply results, or overall market and instrument context cease supporting longs. A protective stop may sit beyond the local boundary invalidating this test trade or beyond the entire Spring structure. The former is closer but more vulnerable to normal movement; the latter permits deeper tests but increases unit risk. Initially target internal resistance and the range ceiling. Manage a subsequent valid breakout using new information rather than assume at entry that a trend must follow.

An Upthrust and its test in a possible distribution range use the reverse logic: price breaks resistance and returns inside, a renewed probe cannot regain resistance, and sellers then depress price to create the short opportunity. Cancel the reversal plan if trading continues above the upper boundary.

LPS and LPSY

The Last Point of Support (LPS) usually occurs during a pullback in late accumulation or after a sign of strength. It is not any higher low: it needs prior upward evidence, an explainable support or breakout location, and contracting pullback distance with difficulty moving lower. Triggers may include a demand bar after the pullback stops, a local downtrend-line break, or an upward turn in smaller swings. Cancel if the pullback penetrates the original range deeply, breaks the support underlying the LPS, and gains acceptance below. Place protection around this pullback's invalidation boundary; initial targets may be the prior high, range ceiling, or a nearer actual supply zone.

The Last Point of Supply (LPSY) is a weak rally in late distribution or after weakness appears. Sellers should already have driven price down; the rally then recovers poorly near resistance, the breakdown level, or the lower range boundary, before price turns down again. Regaining resistance with sustained acceptance cancels the short plan. LPS/LPSY participation usually adds confirmation beyond trading the initial shock, but more confirmation does not necessarily mean a more distant stop. A clear local pullback boundary can shorten stop distance; an oversized trigger bar can instead increase unit risk at a confirmed entry.

Breakout retest

A breakout retest first requires distinguishing valid departure from a brief boundary crossing. A valid breakout generally makes net progress beyond the structure, avoids an immediate complete return inside, and shows acceptance in the new area through time, trading, or swings. The retest checks whether old resistance becomes support or old support becomes resistance. The trigger should occur at the retest location; a historical breakout is not a reason to chase.

For a long retest, sustained trading back inside the range or a rally failing to overcome local supply weakens continuation and may cancel the plan; reverse this for shorts. Protection may be beyond the retest swing or a deeper structural boundary, representing different trade hypotheses. First reference highs/lows ahead, opposing structure, identified high-volume areas, or Chapter 26 candidate targets, constraining theoretical projections by the nearest real obstacle.

Trend continuation

Trend continuation applies after price leaves the old balance area and forms directional swings. The same-scale trend must remain intact, with the latest demand zone supporting the advance or supply zone restraining a declining market unbroken. Wait for a countertrend pullback instead of chasing the end of a directional move. Candidate locations include a previous breakout area, predefined support or resistance, or small reaccumulation or redistribution structures. Modern traded-price tools may add value-area or high-volume node (HVN) edges as location aids. These advanced supplementary tools, fully defined in Advanced Appendix I, are not required for classic trend continuation.

The trigger occurs when weakening pullback magnitude and activity are followed by price crossing a local swing point in the original trend direction. Cancel if the pullback develops an opposing structure, breaks the latest support underlying a long or resistance underlying a short, or leaves insufficient target room. Continuation protection often depends on recent local structure and is sensitive to noise. Moving protection beyond a distant larger structure lowers the chance of a local stop but raises unit risk and reduces size. Management may take partial profits at prior highs/lows or hold after a valid break; choose the rule before ordering.

28.3 Early, Confirmed, and Retest Participation

Early participation occurs when an event first supplies directional-change evidence, such as a Spring recovering the range, an Upthrust returning inside, or a first turn at a range edge. This excludes uninterrupted outward extension but not another breach after recovery, further range expansion, or an opposite response lasting only one short swing. Participants accept these unresolved possibilities for earlier execution and potentially more target room. Earlier does not guarantee a better fill: the shock bar may be wide, market slippage larger, and protection beyond the complete extreme may still create high unit risk.

Confirmed participation waits for a completed test, changed local highs/lows, or a demand/supply bar crossing a key swing. New facts do not generically make the trade safer; they eliminate specific explanations such as an incomplete event, an ineffective turn, or an unproven boundary. Confirmation costs movement already made in the expected direction, potentially worsening entry and reducing room. However, a new local swing can sometimes offer a nearer, logically complete protective boundary than the original extreme. Measure information gained and price paid separately; a conservative label cannot replace actual distance and remaining-room calculations.

Retest participation adds another boundary test. A breakout shows that opposing liquidity on one side of the old boundary was insufficient; a return that holds and turns again further supports changed trading conditions across that boundary. An increasingly deep retest restoring back-and-forth trading inside instead weakens the breakout. Waiting costs opportunities when a strong market never returns, and queue position or depth may prevent a limit fill. Define the acceptable zone, renewed-turn trigger, and cancellation conditions beforehand; a deep return inside cannot indefinitely be called a normal retest.

The three methods may be mutually exclusive or staged within one risk budget. With an early position already open, later confirmation or retest is not an unrelated new opportunity: a shared boundary failure may damage all portions simultaneously, so recalculate the entire position under the common adverse scenario. Missing an early fill creates a lost opportunity; it does not improve the later signal's validity. Once price leaves the planned zone, wait for new structure and boundaries rather than rename chasing as confirmation.

Four Participation Locations

Sequence does not determine price advantage or risk distance. Expected execution, a valid protective boundary, costs, and remaining target room jointly determine the result. Each location still needs its own entry signal.

Figure 28-1 Four Participation Locations (schematic)

28.4 Price Advantage, Information, and Potential Risk

Price advantage comes from expected execution relative to invalidation and the usable target, not simply buying low or selling high. Near the lower boundary while price falls unimpeded, a shorter chart distance does not supply a long advantage. Only after stopping and turning does price near the boundary combine with directional evidence. More confirmation eliminates more erroneous explanations but often worsens entry and consumes room. If confirmation creates a new local boundary, unit risk may shrink again. Information, execution price, and protective boundary change independently, with no fixed conversion among them.

Waiting also creates opportunity cost and selective fills. Waiting for confirmation may miss a move that never pulls back; a limit order seeking a better price may fill more often in weaker moves while the strongest never return. This sample selection makes average outcomes after fills differ from outcomes of all original opportunities. Compare alternatives by recording unfilled and canceled opportunities too, and convert expected entry, protective exit, costs, and slippage into the same monetary risk.

Potential reward to risk is potential after-cost net gain to a conservative target divided by potential after-cost net loss to the protective exit. It compares outcome magnitudes at given prices, boundaries, and targets, not the probability of the structural interpretation being correct. Moving the stop closer than hypothesis invalidation mechanically raises the ratio but can label normal fluctuations as failure. Moving the target to a remote projection enlarges the numerator without strengthening evidence for reaching it. If price, information, boundary, costs, and portfolio capacity cannot jointly meet the plan, abandoning the trade is the decision this comparison produces.

28.5 Hypothesis Invalidation, Discretionary Exit, and Protective Stops

Ultimate structural invalidation tests the larger analytical model; individual trade invalidation tests a narrower participation rationale. A possible accumulation exiting downward, sustaining trading outside, making new lows, and encountering blocked rallies progressively excludes a Spring or temporary breach. One upward swing failing to reach the opposite side shows only that the rally missed its expected area, not that the entire accumulation model is disproven. Conversely, a long based on an LPS holding local support loses its entry condition as soon as that support breaks validly. A larger range that may later become accumulation does not justify unlimited expansion of this local risk.

Derive protective-stop location backward from the local hypothesis failure. State which boundary should not be crossed within normal movement if the view is right; allow noise appropriate to the chosen scale; then recalculate loss at an actually submittable price. An arbitrary fixed-money stop may exit while the rationale remains valid; moving it inside normal swings to shrink the sizing distance changes the hypothesis being tested. A farther boundary reduces ordinary fluctuation triggers but increases unit loss. Resolve the budget conflict by reducing quantity or declining the trade, not inventing a nearer stop without structural meaning.

Discretionary exit addresses progressively worsening evidence before the stop triggers. A breakout expected to extend quickly that instead stalls, returns inside, and resumes opposing movement makes a simple recovery-by-waiting explanation less credible with each fact. A test expected to narrow on lower volume but instead expand in both downward activity and net result likewise need not wait for the outermost boundary. Time alone does not invalidate a trade; the issue is absent expected progress within the specified time while adverse exposure continues. Rules must state scale, expected progress, and conditions for reducing, closing, or tightening protection, rather than treat intraday discomfort as evidence.

A protective stop addresses fast movement or insufficient time to reassess. Trigger price S only changes order state and does not guarantee a fill. Gaps, thin books, price limits, or trading interruptions may produce an exit worse than S. Structural invalidation updates analysis, discretionary exit responds to gradual counterevidence, and protective orders constrain execution risk. They may occur at different locations. A stop shows only that this participation hypothesis or execution boundary was reached; it does not prove the whole structure wrong. Nor does a still-valid larger structure justify canceling protection on an individual trade.

28.6 Market, Limit, and Stop Orders

A market order requests execution against available quotes as soon as possible, prioritizing fill probability rather than price certainty. The visible best bid and ask are only part of liquidity. If size exceeds that level, the remainder may trade against worse quotes within venue price protections, trading bands, and other rules; suspensions, rejections, or disappearing liquidity may still prevent a full fill. Fast markets, openings, and gaps can change quotes before the instruction arrives. When estimating buys at Ask and sells at Bid, the spread is already in expected execution and must not be added again as a cost.

A limit order caps the buying price or floors the selling price but need not wait. A buy limit at or above the ask, or sell limit at or below the bid, may execute immediately if opposing quantity exists. Only the nonmarketable remainder waits or cancels according to time in force and venue rules. Even a touch of the limit may leave no or partial execution because of queue priority, insufficient quantity, or a quick departure. Position and protection systems must use actual quantities in fill reports, not submitted quantity as assumed holdings.

Stop orders separate when an exit activates from how it executes. The trigger may reference last trade, Bid, Ask, or mark price. Meeting it activates the next order rather than establishing a fill. A stop-market order seeks execution and may sweep multiple levels; a stop-limit order remains price constrained and can leave exposure open if the market jumps through it. A trailing stop moves the trigger by platform formula without recognizing structure and may exit on a normal pullback during volatility expansion while the original hypothesis holds. Check the venue's actual trigger source, sessions, time in force, and order rules.

OCO links the intended cancellation of profit-taking and stop orders, but a fill report and cancellation confirmation are separate events. After one side fills partly or fully, the other may already be triggering in transit or not yet canceled. Manual modifications, delays, and platform differences can cause both to fill or even reverse the position. After every fill, cancellation, and modification, reconcile actual holdings, remaining orders, and protected quantity. Define responses to disconnection, rejection, and racing fills before entry.

28.7 Risk Budget, Unit Risk, and Position Sizing

Position sizing converts a structural judgment into a quantity the account can tolerate. Let B be the permitted planned-loss budget, P planned entry, S the protective-stop reference, q quantity, and m the contract multiplier per price unit for a linear product. B must be a fixed positive monetary amount determined before quantity. Choosing affordable quantity first and then reverse-engineering a larger B removes its constraint. P should approximate executable price rather than the chart's ideal line; S is the protective order's planned trigger. Both differ from structural confirmation price: the market may hit S and exit before later conditions such as a close below or a failed next rally confirm structural failure. Waiting for complete structural confirmation before stopping makes actual loss distance greater than that used in sizing.

Initially assume risk budget, profit/loss, and costs are in one currency and ignore nonlinear effects. P and S share price units, so |P−S| is a price distance. Multiplier m specifies monetary profit/loss per position unit for a one-unit price change. Thus |P−S|×m is the monetary loss for one position unit from P to S, and q×|P−S|×m is the whole position's planned loss. Absolute value expresses distance only; direction must still be checked separately. This formula converts distance and quantity into money without improving structural reliability. Sizing keeps planned loss within budget for a given entry and stop; foreign currency, costs, and nonlinear payoffs are addressed in 28.8.

At entry, a long's initial protective stop normally lies below P and a short's above P. If reversed, verify order direction and trigger rules. Moving protection to entry after favorable progress does not reduce economic risk to zero: entry/exit costs and slippage can still create loss. Add expected round-trip fees per unit, stop slippage, and costs not already reflected in execution as c: r_unit＝|P−S|×m＋c. c must cover a complete round trip per unit. Do not add spread again if Ask/Bid execution already includes it, or slippage already incorporated in expected fills.

The expression |P−S|×m＋c applies to initial entries whose stop is on the loss side. If a held position's stop moves into profit, use the product's actual position profit/loss function to reassess scenario risk from the current state to protective exit. Do not keep calling the absolute distance from original entry an expected loss. Initial R₀ remains fixed.

With approximately linear costs and quantity increment u, theoretical size is B÷r_unit and actual size rounds down by increment: q＝u×floor[(B÷r_unit)÷u]. floor takes the greatest integer no larger than its argument, so round down to a tradable increment, never to nearest. This division is meaningful only when B＞0, u＞0, and r_unit＞0. If any is zero or negative, check entry, stop, multiplier, costs, and currency conversion; zero unit risk never permits infinite size. Minimum order quantity may differ from u. A computed q below that minimum means the trade is infeasible at the current stop distance and budget, not that size should be forced up to the minimum.

Rounding down guarantees estimated loss stays within budget only if unit risk does not change with quantity. Minimum commissions, tiered rates, taxes, notional-based fees, changing increments, or size-dependent impact costs make cost more than a constant per unit, so the floor formula alone is insufficient. Use the complete constraint L(q; P, S, cost)≤B and select the largest q among orders meeting minimum size and increment rules that satisfies it.

L includes underlying loss from expected entry to protective exit plus all related costs; cost is a function of quantity, order type, and market conditions. If the complete L(q) is known to be nondecreasing, enumerate valid quantities from small to large or use a search algorithm for the largest qualifying value. Knowing that one cost component increases does not establish that the whole L does. With rebates, minimum fees, or fee-tier jumps, calculate candidate quantities separately rather than assume proportionality. Protect the actually filled quantity after a partial fill. After partial exits, additions, or stop changes, recalculate using remaining quantity, current stop distance, and incurred costs.

Risk-based quantity is only one ceiling. Actual q is also constrained by available margin or buying power, exchange/broker limits, liquidity, and portfolio limits; use the executable amount satisfying all. Margin permission does not mean the account should bear the associated stop loss, and risk-budget permission may still be unexecutable for insufficient margin. Liquidity checks must include capacity near the stop: estimating a large exit from the best book level understates multi-level execution cost.

Define risk budget as a tolerable account-currency loss subject to per-trade, daily, portfolio, and extreme-scenario limits. It is neither invested principal nor margin. Margin is performance collateral required by the broker or clearing system; planned loss estimates entry-to-protective-exit movement and costs. Low margin and high leverage permit larger notional positions but do not shorten stop distance or increase loss capacity. Gaps, suspensions, price limits, disconnections, or rejected stops can make actual loss exceed B. B is a planned ceiling under normal execution, not a guaranteed maximum loss. Reduce size or decline entry if stress losses are intolerable.

Risk Budget and Quantity Calculation

Round down, never to nearest. If q is below the minimum order size, the trade is infeasible at the current stop distance and budget. Do not calculate quantity when B, u, or r_unit is zero or negative. Rounding down guarantees expected loss does not exceed B only when unit risk is independent of quantity; otherwise use the complete loss function L(q; P, S, cost)≤B.

Figure 28-2 Risk Budget and Quantity Calculation (from Section 28.7 of this course)

Risk Budget, Margin, and Principal

With quantity, price, and multiplier unchanged, changing leverage settings does not automatically change notional exposure but does affect margin requirements. With price and multiplier unchanged, notional exposure changes with actual quantity. For example, with available margin unchanged, increasing permitted leverage and trade quantity increases notional exposure.

Gaps, slippage, and insufficient liquidity may cause actual loss to exceed risk budget B.

Figure 28-3 Risk Budget, Margin, and Principal (from Section 28.7 of this course)

28.8 Multipliers, Minimum Ticks, Currencies, Costs, and Nonlinear Products

Contract multiplier m converts quotation changes into profit/loss. A one-price-unit change produces m settlement-currency units of gross profit/loss per linear contract, or q×m for q contracts. Tick size specifies the permitted quotation increment; tick value is the monetary value of one tick. For a contract settled linearly in one settlement currency, tick value＝tick size×m. Check entry-to-stop ticks as |P−S|÷tick size; underlying risk is also ticks×tick value×q. With tick value＝tick size×m, fixed tick size, and identical inputs, the two calculations necessarily agree; their difference cannot test order-price validity. Check each submittable order price individually against venue price levels. If the formulas disagree, check tick value, tick size, multiplier, quantity, and currency consistency.

Limits, stop triggers, and other submittable prices must follow applicable minimum increments. Each actual fill should also meet venue rules, though the quantity-weighted average of multiple fills may not fall on a permitted price level. Retain individual fills and use the average to calculate actual position cost. Do not mechanically round to nearest: rounding a long's sell stop upward may trigger earlier, while rounding downward increases risk distance. First derive a submittable stop using direction and platform rules, then recalculate quantity. Round theoretical quantity down to valid minimum shares, contracts, or increments. Some products use different ticks in different price bands or special order-type rules. Check each expected order price's band rather than apply the current price's tick to all risk.

Quotation currency, profit/loss settlement currency, and account base currency can be three different currencies. Quotation currency defines price units, settlement currency the currency in which contract profit/loss first arises, and base currency the units of budget B. If L is first calculated in settlement currency, convert it using the correct exchange-rate direction. Whether the rate means B per unit A or the inverse determines multiplication or division; do not infer this from the pair's name alone. State the rate source and estimate time in the plan and allow for adverse changes, because the position may carry unhedged currency risk as well as instrument risk.

Costs should include entry and expected exit commissions, exchange and clearing fees, taxes, borrow or financing costs, funding rates, and reasonable slippage. Carry costs can be expenses or income, but uncertain future income must not be used to enlarge size; stress scenarios should also consider rate changes. Minimum commission is a fixed per-order component, notional-based fees vary with value, and split fills can increase charge counts. Do not force all into one constant per-unit c. If executable Ask buys and Bid sells already include spread, do not add it again. If P and S use midprices, include crossing the spread in expected execution or costs.

The linear formula applies only where profit/loss is approximately proportional to price difference. Some inverse contracts define face value in quotation currency, settle in the underlying asset, and use reciprocals of entry and exit prices in the payoff; follow the contract's exact specifications. Equal upward and downward distances do not yield symmetric quantities of settlement asset, so fixed m cannot describe risk. Options depend simultaneously on underlying price, implied volatility, time remaining, and exercise terms. Even at the same underlying S, executable option prices can differ with time and volatility. Knock-in, knock-out, daily-reset, and path-dependent products can also produce different results according to the sequence of prices.

Do not calculate nonlinear-product risk directly as |P−S|×m. Use official contract specifications and the payoff function with direction, quantity, actual entry, stop scenario, multiplier or face value, settlement currency, exchange rate, and costs, requiring total loss L(q; P, S, cost)≤B. Quantity, multiplier or face value, and currency-conversion factors should be positive; the product payoff handles direction. Preserve the income/expense signs of funding, interest, and rebates without relying on uncertain income in sizing. Calculate valid quantities individually for fixed increments, and do not simply multiply one unit's loss if loss is not strictly linear in quantity. For options or path-dependent products, include time to the stop, volatility changes, gaps, and contractual valuation conventions, using the conservative loss required by account rules.

Slippage estimates must match order type, session, book depth, and event risk. A market stop normally becomes an execution-seeking order after triggering and can fill beyond the trigger; a stop-limit controls price but may not fill. Use historical slippage distributions of comparable orders at comparable times for normal conditions and separate stress scenarios for gaps or abrupt liquidity loss. Averages hide tails, and samples may omit the forthcoming event environment. Do not size on guaranteed execution at S. If stressed losses exceed account limits, reduce quantity, avoid the event, choose a more liquid instrument, or refrain from trading instead of deleting slippage from the formula.

28.9 Portfolio Correlation, Concentration, and Abnormal-Market Contingencies

Multiple trades cannot be assessed only against individual budgets. Instruments may share an industry, index, funding currency, or seemingly different exposures that all depend on rising risk appetite. Low historical correlation in normal conditions does not ensure independence under stress; correlations change and several stops may simultaneously fill worse in a gap. Taking the square root of summed squared individual risks does not by itself establish diversification.

Group portfolio exposures by direction and risk source, then calculate each group's net position and loss under common adverse scenarios. Separately assess correlated longs in a market decline, correlated shorts in a squeeze, and cross-currency positions during exchange-rate jumps. Check concentration by instrument, industry, expiry month, venue, broker, and liquidity session. Relative strength ranks candidates but is insufficient reason to allocate more risk; several top-ranked instruments may still share one factor.

Abnormal-market plans must specify actions: order type or manual response when an opening gap passes the stop; limits on new positions during interruption or price-limit conditions; backup channels during platform disconnection; protection quantity synchronization after partial fills; whether to pause automation on abnormal data; and whether to reduce risk before major scheduled events. Identify tail risks a stop cannot solve, absorbing uncertainty through smaller positions and portfolio limits.

Review every addition on the portfolio after inclusion, not as an isolated trade. A new trade losing alongside existing positions in the same scenario may require rejection or reduction despite passing its own budget. After reductions, hedges, or exits, update actual net exposure. Nominally opposite directions are not automatically a complete hedge unless maturity, basis, multiplier, and liquidity genuinely match.

28.10 A Complete Trading Plan

A complete plan first defines information cutoff, market, transaction-data coverage, sessions, and price treatment, then states analytical bias and opposing evidence. Identify the participation model—Spring/Test, LPS/LPSY, breakout retest, or trend continuation—its structural phase, specific trade zone, and basis for target room. State the primary scenario conditionally: if price reaches a specified zone and a defined trigger appears, prepare to execute. The alternative explains how opposite development changes direction; remain flat if no branch qualifies.

Participation conditions include location, scale, trigger-completion method, order type, and validity period. Distinguish trigger from expected fill and predefine responses to unfilled limits, partial fills, opening gaps, and post-trigger slippage beyond the limit. Do not impulsively chase after price departs. Missing the original trigger permits only a predefined reentry model or a newly completed plan.

Cancellation conditions apply from preparation through order submission until execution: for example, an accepted opposing breach, changed overall-market bias, trigger waiting time exceeded, exhausted target room, or worsening liquidity or borrow conditions. Stop submitting or cancel unfilled orders when they occur. For partial fills, cancel the remainder while retaining protection and discretionary-exit arrangements for the filled position. List post-entry local stops, discretionary-exit evidence, and ultimate larger-structure invalidation separately rather than assign every function to a vague exit if wrong.

Risk conditions include (B, P, S, m), tick size, tick value, trading and settlement currencies, exchange-rate source and conversion time, fees, normal and stressed slippage, unit risk, and final order quantity meeting minimums and increments. List post-addition portfolio risk sources and concentration. Check stops, profit targets, and OCO against platform rules. After any fill, recalculate with actual price and quantity; if over budget, prioritize reducing to an acceptable quantity.

Management rules should define price and time expectations, first and final targets, reduction proportions, independent conditions permitting additions, stop movement, and discretionary exits. Add only on new structural evidence when the whole position again satisfies portfolio budget, never merely to average down a losing position. Specify which items may change with new evidence, whether the trader or software executes them, and when new rules take effect, preserving previous versions. Analysis, opportunity, trigger, orders, risk, and management then form one consistent plan.

Plan status must also be explicit. Prepared means context and location qualify but the trigger is absent; active means orders can execute within the validity period; partially filled requires protection based on actual quantity and a decision on remaining orders; holding begins tracking price and time expectations; closed requires canceling all associated orders and saving final fills. A cancellation condition makes the plan void. A later return to the same zone must not automatically revive old orders; create a new version using new information. State transitions should follow market events, fill reports, or explicit manual instructions so system and human records do not disagree.

Before submission, perform a final consistency check: intended data source, correct contract and expiry, consistent side, quantity, price units, and currency, protection covering actual holdings, and existing exposure to the same risk source. After submission, use the broker or venue's order ID, status, and fill reports rather than infer success from a local display or strategy request. Record rejections, disconnections, duplicate fills, and manual takeover. Only when these operational fields accompany structural judgment can the written plan be executed in fast markets rather than serve merely as retrospective commentary.

Elements of a Complete Trading Plan

Figure 28-4 Elements of a Complete Trading Plan (from Section 28.10 of this course)