Based on Larry Williams, Long-Term Secrets to Short-Term Trading (2nd Edition, 2011)
Larry Williams is one of the most celebrated short-term traders in history. His credentials include:
Williams did not come from a Wall Street background. He is largely self-taught, having begun trading in the 1960s while working as a journalist in Oregon. His approach is empirical and data-driven — he tests everything, trusts nothing that cannot be verified statistically, and treats trading as a probability game, not a prediction game.
The book's title captures its central paradox: short-term trading success requires long-term thinking. Most short-term traders fail because they think short-term — they react to noise, chase momentum, and have no framework for distinguishing signal from randomness.
Williams argues that profitable short-term trading rests on a surprisingly small number of structural truths about how markets move:
1. MARKET STRUCTURE — Prices move in identifiable short-term cycles (3-bar patterns, swing points).
2. VALUE ZONES — Markets oscillate between overbought and oversold; buy value, sell premium.
3. SMART MONEY — Commercial hedgers (tracked via COT data) are the best contrarian indicator.
4. PATTERN SETUPS — Specific price patterns (Oops, Smash Day, etc.) create high-probability entries.
5. MONEY MGMT — Position sizing and risk control determine survival more than entry quality.
All five must be understood as an integrated system. A pattern without a money management framework is gambling. COT data without timing is too early. Timing without structure is noise.
Williams defines short-term trading as trades lasting one to five days, occasionally extending to two weeks. He is not a scalper and is not a swing trader in the multi-week sense. His time horizon is long enough to capture meaningful moves but short enough to avoid the large adverse excursions that destroy longer-term positions.
The key insight: short-term traders make money from the daily range expansion, not from the trend. A short-term trader's profit comes from buying near the low of a move and selling into the range expansion that follows — or the reverse for shorts. The underlying trend provides directional bias, but the daily range is the profit engine.
Williams acknowledges that short-term price movement is extremely noisy. However, within that noise, certain structural regularities persist:
Williams distills his understanding of short-term moves into a single principle:
Short-term price lows are made when the market closes below the prior day's low, then fails to follow through to the downside. Short-term highs are made when the market closes above the prior day's high, then fails to follow through to the upside.
This principle underlies nearly every entry and exit pattern in the book. It is the grammar of short-term price action.
Contrary to momentum-following approaches, Williams is fundamentally a mean-reversion trader on the short-term time frame. His core behavioral rule:
This works because short-term extremes tend to be emotional overreactions. The crowd panics at lows and gets euphoric at highs — both states are temporary.
Williams observes that short-term markets move in a natural 3-bar rhythm: a swing low forms, price rallies for one to three bars, a swing high forms, price declines for one to three bars, and the cycle repeats. This is the market's "heartbeat."
The 3-bar cycle is not a rigid count. It is a structural tendency — a way of identifying the minimum unit of a short-term swing. Sometimes the cycle stretches to four or five bars; sometimes it compresses to two. But three bars is the modal duration.
Williams defines swing points precisely:
Short-Term Low (Swing Low):
Low[i] < Low[i-1] AND Low[i] < Low[i+1].Short-Term High (Swing High):
High[i] > High[i-1] AND High[i] > High[i+1].Williams extends the concept to create a fractal structure:
This creates a hierarchy:
Long-term swing points
└── Intermediate-term swing points
└── Short-term swing points (3-bar cycle)
└── Individual price bars
Trend direction is determined by the sequence of swing points:
| Pattern | Trend |
|---|---|
| Higher swing highs + higher swing lows | Uptrend |
| Lower swing highs + lower swing lows | Downtrend |
| Mixed or overlapping swing points | Trading range / no trend |
A short-term trader uses the intermediate or long-term trend for directional bias and the short-term swing points for entry timing. You buy short-term swing lows in an intermediate uptrend. You sell short-term swing highs in an intermediate downtrend.
Williams %R is a momentum oscillator that measures the current close relative to the high-low range over a lookback period. It is mathematically related to the Stochastic Oscillator but inverted.
Formula:
%R = (Highest_High(n) - Close) / (Highest_High(n) - Lowest_Low(n)) × (-100)
Where n is the lookback period (Williams typically uses 10 periods for short-term trading).
Scale: The indicator ranges from 0 to -100.
Williams %R is not a simple overbought/oversold indicator used mechanically. Williams himself warns against the naive approach of "buy when %R is below -80, sell when above -20." That approach loses money because strong trends can keep %R pinned in extreme zones for extended periods.
The correct usage:
In Williams's system, %R is used primarily as a timing confirmation filter, not as a standalone signal generator. A typical sequence:
Step 1: Identify intermediate uptrend (swing point analysis).
Step 2: Wait for short-term pullback (price approaches or makes a swing low).
Step 3: Confirm %R is in oversold territory (-80 or below).
Step 4: Wait for a specific entry pattern (Oops, Smash Day, etc.).
Step 5: Enter the trade.
The %R reading at step 3 increases the probability that the entry pattern at step 4 will lead to a successful trade. Without the oversold condition, the same pattern is less reliable.
Williams's accumulation/distribution work focuses on the relationship between the open and close relative to the day's range. The core insight:
The WAD indicator builds a cumulative running total:
If Close > Open:
AD_today = Close - True_Low
(True_Low = min(Low, Prior_Close))
If Close < Open:
AD_today = Close - True_High
(True_High = max(High, Prior_Close))
If Close == Open:
AD_today = 0
WAD = Cumulative sum of AD_today values
The power of WAD is in divergences between the indicator and price:
Williams uses WAD divergences as a structural filter — a divergence alone is not a trade signal, but a trade signal that occurs during a WAD divergence is significantly higher probability.
The Commodity Futures Trading Commission (CFTC) publishes the Commitment of Traders report every Friday (data as of Tuesday close). It breaks open interest into three categories:
| Category | Who They Are | What They Do |
|---|---|---|
| Commercials | Producers, consumers, hedgers | Hedge business risk; trade based on actual supply/demand knowledge |
| Large Speculators | Hedge funds, managed futures, CTAs | Trade for profit based on models and momentum |
| Small Speculators | Individual retail traders | Trade based on emotion, tips, and hope |
Williams considers COT data to be the single most valuable tool available to futures traders. His framework:
Williams normalizes COT data into a percentage index to compare current positioning against historical extremes:
COT_Index = (Current_Net_Position - Min_Net_Position(n)) /
(Max_Net_Position(n) - Min_Net_Position(n)) × 100
Where n is the lookback period (Williams often uses 3 years / 156 weeks).
Interpretation:
COT data is slow-moving — it changes weekly and reflects positioning that may have built over months. It is therefore used as a background condition, not a timing tool:
COT favorable + Trend favorable + %R timing + Entry pattern = HIGH PROBABILITY TRADE
COT unfavorable + Everything else favorable = LOWER PROBABILITY (reduce size or skip)
Williams emphasizes that trading against the commercials — even with perfect timing — dramatically reduces win rates over large samples.
Williams was one of the earliest traders to rigorously test seasonal tendencies in commodity and stock index markets. His findings:
Williams documents several persistent seasonal effects:
Seasonality acts as another structural filter in the layer cake:
Layer 1 (slowest): Seasonal tendency — bullish or bearish for this time of year?
Layer 2: COT positioning — are commercials confirming the seasonal bias?
Layer 3: Trend (swing point structure) — is the intermediate trend aligned?
Layer 4: %R timing — is the market oversold/overbought in the direction of the setup?
Layer 5 (fastest): Entry pattern — has a specific trigger occurred?
All layers are filters. You only trade when multiple layers align. The more layers that confirm, the higher the probability.
A large range day is simply a day with an unusually wide high-low range — typically in the top quartile of range sizes over the past 20 to 60 trading days.
Calculation:
Range_today = High - Low
Average_Range = SMA(Range, 20)
Large_Range_Day = Range_today > (Average_Range × multiplier)
Williams typically uses a multiplier of 1.5 to 2.0. A range 150-200% of the average is a large range day.
Large range days are important because they reveal institutional conviction:
Williams's key observation: the day after a large range day is one of the best short-term trading days. Specifically:
Large range days reset the emotional state of the market. They attract attention, bring in late participants, and create the conditions for the specific entry patterns described in the next section.
This is Williams's most famous pattern and perhaps the simplest. It exploits the tendency for gap openings to fail.
Setup (Long):
Setup (Short):
Why it works: Gap openings against the prevailing trend are typically emotional reactions — overnight news, panic, or euphoria. When the gap fails to hold (price reverses back through the prior day's range), it traps the gap traders on the wrong side and creates a burst of momentum in the reversal direction.
Stop placement: Below the session low (for longs) or above the session high (for shorts). The risk is defined by the distance from entry to the extreme of the gap.
The Smash Day captures a reversal of a short-term extreme.
Setup (Buy Smash Day):
Setup (Sell Smash Day):
Why it works: The Smash Day represents a failed attempt at continuation. The close beyond the prior day's extreme suggests momentum — but when that momentum is immediately reversed, the aggressive traders who acted on it are trapped. Their forced liquidation fuels the reversal.
A variation of the Smash Day that looks at the close relative to the open rather than the prior day's range.
Setup (Buy Hidden Smash Day):
Setup (Sell Hidden Smash Day):
Key difference from regular Smash Day: The Hidden Smash Day does not require the close to exceed the prior day's range. It focuses on intra-day buyer/seller dominance through the open-close relationship.
An outside day (also called an engulfing bar) is a bar whose range completely encompasses the prior bar's range.
Setup (Buy Outside Day):
Setup (Sell Outside Day):
Filter: Outside days are most significant when they occur at swing points — particularly when they form at the end of a 3-bar (or longer) pullback in an established trend.
Not all patterns are equal. Williams's implied hierarchy by reliability:
Highest : Oops! in direction of trend with COT/seasonal confirmation
Smash Day at intermediate swing point with %R confirmation
Middle : Hidden Smash Day with trend alignment
Outside Day at swing point
Lowest : Any pattern against the intermediate trend
Any pattern without %R confirmation
Williams states clearly: the exit is more important than the entry. A good entry with a bad exit produces losses. A mediocre entry with a good exit can still produce profits.
His exits fall into three categories:
Williams is a strong advocate of time stops — exiting a trade after a fixed number of days regardless of profit or loss.
Rule: If the trade is not profitable within 3 trading days of entry, exit at the close or the next open.
Williams uses the average daily range to set price targets:
Target (long) = Entry Price + (Average_True_Range(n) × multiplier)
Target (short) = Entry Price - (Average_True_Range(n) × multiplier)
Typical multipliers range from 1.0 to 2.0× ATR depending on market conditions and the strength of the setup. In high-conviction setups (multiple confirming filters), Williams may use a higher multiplier. In marginal setups, a lower one.
Every trade must have a predefined stop loss. Williams's stop placement rules:
One of Williams's distinctive exit methods: exit on the first profitable opening.
Williams states repeatedly that money management — not indicators, not patterns, not market selection — is the single most important determinant of trading success. He argues that he could give a trader a system that is right only 40% of the time and still make money, provided the money management is correct.
Williams was one of the first trading authors to popularize the Kelly Criterion for position sizing:
Kelly % = W - [(1 - W) / R]
Where:
W = Win rate (probability of a winning trade)
R = Win/Loss ratio (average win size / average loss size)
Example: If a system wins 55% of the time and the average win is 1.5× the average loss:
Kelly % = 0.55 - [(1 - 0.55) / 1.5] = 0.55 - 0.30 = 0.25 (25% of capital per trade)
Williams warns that full Kelly is too aggressive for real-world trading. He recommends:
Williams's practical implementation for most traders:
Position_Size = (Account_Equity × Risk_Percentage) / Dollar_Risk_Per_Contract
Where:
Risk_Percentage = Maximum percentage of equity risked per trade (1-5%)
Dollar_Risk_Per_Contract = Distance from entry to stop in dollar terms
Example: $100,000 account, 2% risk, $500 risk per contract:
Position_Size = ($100,000 × 0.02) / $500 = 4 contracts
Williams makes an often-overlooked point about drawdown mathematics:
| Drawdown | Gain Needed to Recover |
|---|---|
| -10% | +11.1% |
| -20% | +25.0% |
| -30% | +42.9% |
| -50% | +100.0% |
| -75% | +300.0% |
This asymmetry means that avoiding large drawdowns is more important than achieving large gains. A 50% drawdown is not twice as bad as a 25% drawdown — it is four times as bad in terms of the effort required to recover. This is why money management, not entry quality, determines survival.
Williams uses sentiment indicators to measure crowd psychology. His principle: when the crowd reaches an extreme of bullishness or bearishness, prices are near a reversal point. This is not contrarianism for its own sake — it is based on the observation that when virtually everyone who is going to buy has bought, there are no more buyers left, and prices must fall (and vice versa).
Williams does not rely on any single sentiment indicator. He looks for consensus across multiple sentiment measures:
Strong bullish sentiment setup (look for buys):
- COT: Commercials heavily net long
- Advisors: Majority bearish
- Put/Call: Elevated (fear)
- Volatility: Elevated (panic)
- Media: Negative headlines
Strong bearish sentiment setup (look for sells):
- COT: Commercials heavily net short
- Advisors: Majority bullish
- Put/Call: Depressed (complacency)
- Volatility: Low (complacency)
- Media: Positive headlines
When three or more sentiment measures reach extremes simultaneously, the probability of a reversal increases substantially.
Williams describes his complete trading approach as a cycle — a repeating process of analysis, setup identification, trade execution, and management:
┌──────────────────────────────────────────────────────────────────┐
│ THE WILLIAMS CYCLE │
│ │
│ 1. STRUCTURAL ANALYSIS │
│ - Identify intermediate trend via swing points │
│ - Note seasonal tendency for this market / time of year │
│ - Check COT positioning (weekly update) │
│ │
│ 2. CONDITION ASSESSMENT │
│ - Is %R in the correct zone (oversold for buys, etc.)? │
│ - Is the accumulation/distribution line confirming? │
│ - Are sentiment indicators at extremes? │
│ │
│ 3. PATTERN RECOGNITION │
│ - Watch for entry patterns (Oops!, Smash Day, etc.) │
│ - Grade the pattern: how many confirming factors? │
│ │
│ 4. EXECUTION │
│ - Calculate position size via fixed fractional method │
│ - Place entry order with predetermined stop │
│ - Define exit: time target, price target, or FPO │
│ │
│ 5. MANAGEMENT │
│ - Monitor for exit criteria daily │
│ - Never move stops in the adverse direction │
│ - Exit when any exit condition is triggered │
│ │
│ 6. REVIEW │
│ - Log the trade (entry, exit, reason, result) │
│ - Update system statistics (win rate, R-multiple) │
│ - Adjust position sizing if equity has changed significantly │
│ │
│ [Loop back to Step 1] │
└──────────────────────────────────────────────────────────────────┘
Before any trade, Williams (implicitly) runs through a checklist:
□ Intermediate trend direction identified (swing point structure)
□ COT positioning is favorable (or at least neutral)
□ Seasonal tendency is favorable (or at least neutral)
□ %R is in the correct zone for the trade direction
□ A specific entry pattern has triggered
□ Stop loss is defined and position size calculated
□ Risk on this trade does not exceed maximum per-trade risk
□ Total portfolio risk does not exceed maximum portfolio risk
□ Exit method is predetermined (time, target, or FPO)
If any item cannot be checked, the trade is either skipped or size is reduced.
Overtrading. Taking trades that meet only two or three criteria instead of waiting for full alignment. Williams's own championship year involved a relatively small number of high-conviction trades, not hundreds of marginal ones.
Ignoring money management. Focusing on entry perfection while risking too much per trade. "The market does not care about your analysis. It only cares about your position size."
Moving stops. Once a stop is set, it must not be moved farther from the entry. Moving stops in the adverse direction is the single most destructive habit a trader can develop.
Holding losers, cutting winners. The psychological tendency to let losses run (hoping for recovery) while taking profits quickly (fearing reversal). Williams insists on the opposite: cut losses immediately and let winners reach their targets.
Trading against the commercials. Taking long positions when commercial traders are heavily net short (or vice versa). The COT data is freely available — ignoring it is trading with a blindfold.
Assuming seasonality is destiny. A seasonal tendency is a probability tilt, not a guarantee. Traders who enter seasonal trades without confirming price action are gambling.
Curve-fitting. Optimizing system parameters to fit historical data perfectly. Williams warns that any system that looks perfect on historical data is almost certainly overfit and will fail in real-time. Robust systems look "merely good" on historical data.
Emotional trading. Making decisions based on how you feel rather than what the data shows. Williams notes that his worst losses have always come when he deviated from his system.
Failing to keep records. Without a detailed trade log, you cannot identify which elements of your system are working and which are not. You are flying blind.
Not understanding the game. Short-term trading is not investing. It is a probability game with a definite edge, played over many repetitions. Treating any single trade as important is a fundamental misunderstanding.
Date: Hypothetical trading week.
Step 1 — Structural Analysis:
Step 2 — Condition Assessment:
Step 3 — Pattern Recognition:
Step 4 — Execution:
Step 5 — Management:
Step 6 — Review:
"Short-term trading is not about predictions. It is about reactions. You do not need to know where the market is going. You need to know what you will do when it gets there."
"The biggest secret to short-term trading is that you need long-term thinking. The irony of this game is that the short-term traders who survive are the ones who think about the long run."
"Money management is the most important chapter in this book. It is the most important chapter in any trading book. If you learn nothing else, learn this."
"I have seen traders with great systems go broke, and I have seen traders with mediocre systems get rich. The difference was always money management."
"The commercials are the smart money. They know more about their markets than you ever will. When they are buying, you should be buying. When they are selling, you should be selling. It is that simple, and that hard."
"The Oops! trade works because it exploits a very human mistake — the gap opening is an emotional reaction, and when it fails, the emotional traders are trapped."
"Most traders spend 90% of their time on entries and 10% on exits. It should be the reverse. A great entry with a terrible exit is a losing system. A mediocre entry with a great exit can still make money."
"My winning percentage is not as high as most people think. What matters is not how often you win, but how much you make when you win and how little you lose when you lose."
"If you are not keeping records of every trade, you are not a trader. You are a gambler who happens to be using a brokerage account."
"The worst thing that can happen to a new trader is to make money right away. It teaches all the wrong lessons."
End of specification. This document covers the core concepts, patterns, indicators, money management framework, and implementation logic from Larry Williams's Long-Term Secrets to Short-Term Trading (2nd Edition). It is intended as a reference for systematic implementation, not as a substitute for reading the original work.