Trade Stocks and Commodities with the Insiders: Secrets of the COT Report
Larry Williams · 2005 · 224 pages
Williams builds a genuinely useful set of tools from the CFTC's weekly Commitments of Traders report: an index that ranks the commercials' current position against three years of history, and a follow-up formula (WILLCO) that measures them against total open interest. The catch is that almost none of it applies to individual stocks, despite the title.
The one idea
The Commitments of Traders report splits every futures market into three groups each week: commercials (the hedgers, producers and end users of the actual commodity), large traders (mostly commodity fund managers trading in 100-lot size), and small traders (the public). Williams' argument is that the commercials are the smart money because they operate in the cash market every day and have superior information, while large traders are trend followers who pile in near the end of moves and small traders are, by the record, wrong most of the time. The book's job is to turn those three weekly position counts into something you can act on.
The core tool is the COT Index: take this week's net long/short reading for a group, find where it falls between the highest and lowest readings of the past three years, and express that as a percentage. Above 80 percent means the group has done more buying, relative to its own recent history, than at almost any point in three years; below 20 percent means the opposite. The insight Williams keeps hammering is that the absolute position (net long or net short) matters far less than the relative one: the commercials can be net short and still be at their most bullish reading in three years, because they were even more short before.
Later in the book Williams refines this into WILLCO (Williams Commercial Index): instead of comparing the commercials to their own three-year range, you compare the commercials' short position to total open interest, on a 26-week stochastic. A reading of 75 to 80 percent or higher is his buy trigger; a reading near 55 to 65 percent short-to-OI has historically preceded tops in currencies. The idea is to weight the commercials against everyone else in the market at once, not just against their past selves.
What it actually teaches
The book builds its toolkit in a rough sequence, each layer adding perspective the previous one lacked.
BUILD THE COT INDEX (CHAPTER 4)
For any group (commercials, large traders, or small traders), take this week's net long/short value, subtract the lowest value of the past three years, and divide by the range (highest minus lowest) of the past three years, then multiply by 100. Williams' own worked example: current value 350, three-year low -150, three-year high 750. (350-(-150))/(750-(-150)) x 100 = 33%. Above 80 percent is read as bullish for that group's positioning; below 20 percent is bearish. For commercials, treat 80/20 as buy/sell zones; for the public and for large traders, Williams says to fade them at the same 80/20 extremes because both groups are shown, chart after chart, buying tops and selling bottoms.
READ THE PUBLIC AS A CONTRARY INDICATOR (CHAPTER 5)
The small-trader net position is charted against price. Williams' rule pair: sell short when the public index is extremely high, buy long when it is extremely low, then do the exact opposite of what the public is doing as a group. His cotton example: the public held its largest net long position on July 3, 1998, the exact start of a six-and-a-half-year bear market, and its smallest (near net short) on June 6, 2003, the exact start of the next major rally.
TREAT LARGE TRADERS AS TREND FOLLOWERS TO FADE (CHAPTER 6)
Large traders (fund managers, mostly) are shown getting most bullish at market highs and most bearish at market lows across gold, bonds, and cattle. Williams explains this as mechanical: different funds trigger off different lookback highs (20-day, 40-day, 26-week, 52-week breakouts), so the longer a trend runs, the more of them pile in, right up until there is no one left to push price further. His rule: use the same COT-index math on the large-trader position, but read above 80 as a sell setup and below 20 as a buy setup, the reverse of how he treats the commercials.
LOOK INSIDE OPEN INTEREST, NOT JUST AT IT (CHAPTERS 8-9)
Williams rejects the textbook rules (OI up + price up = bullish, etc.), citing his own count that they work only about a third of the time. His fix is to ask who is causing a change in open interest: if OI rises because the commercials are buying, that is a different signal than if it rises because the public or large traders are buying. He also argues low open interest itself is a buy signal, since it means the public has left a market, and it is usually the commercials who stay behind.
COMBINE COMMERCIALS WITH TOTAL OPEN INTEREST: WILLCO (CHAPTER 10)
The final indicator, WILLCO, is a 26-week stochastic of (commercial net position / total open interest), used as commercial shorts as a percent of OI. In his Japanese yen example, once commercial short selling passed 55 percent of total OI, a top followed; by late November 2004 it had reached 64 percent. His generic buy trigger for WILLCO is a reading of 75 to 80 percent or higher in the bullish direction; on the sell side he cites levels above 65 percent preceding 1988-1991 yen tops. He pairs WILLCO with a same-day open-to-close 'professional buying' index (ProGo, a 14-day average of the open-to-close change) to look for divergence between price and pro accumulation before timing entries with a simple trend line break.
What it looks like Monday morning
For a futures trader, the concrete action is to start pulling the CFTC's free weekly COT report and building the three-year COT index for the commercials in whatever markets you already trade, watching for readings above 80 or below 20 as areas of interest rather than trade triggers on their own. Williams is explicit that the index tells you when a market is set up, not when to enter; he still uses a plain trend-line break on price for the actual entry.
For someone drawn in by the word 'stocks' in the title, Monday morning is a letdown: there is no official COT report for individual equities, and the book's answer for stocks is a proprietary substitute Williams calls a proxy index, which he says he is 'not yet willing to fully reveal.' The honest move is to treat this book as a futures and currency book that happens to have one speculative chapter on stocks, and not build an equity strategy around a formula the author does not publish.
| Input | Value |
|---|---|
| Current week's net position | 350 |
| Lowest value, last 3 years | -150 |
| Highest value, last 3 years | 750 |
| Calculation | (350 - (-150)) / (750 - (-150)) x 100 |
| Result | 33% (relatively neutral, not bullish or bearish) |
Readings above 80% are read as bullish for the commercials and above 55-65% short-to-OI (WILLCO) as bearish for the same group; below 20% is read as bearish for commercials but as a buy signal when applied to the public or large-trader positions, since Williams treats those two groups as contrarian indicators.
The absolute position they had was not nearly as important as how the current week's reading fit into the grand schemeLarry Williams, Trade Stocks and Commodities with the Insiders
Where it fails
- The stock-market promise in the title is not delivered. Chapter 11, 'A New Twist on the Commercials: Using Them for Stocks,' admits there is no COT reporting for individual equities, so Williams substitutes a self-built 'proxy index' modeled on commercial buy-weakness/sell-strength behavior. He states outright that he is 'not yet willing to fully reveal the index' and instead points readers to his paid website to watch it updated. A reader who buys this book specifically for a stock-trading edge gets a handful of chart examples (Microsoft, Citigroup, Intel, Starbucks) and no replicable formula.
- The commercial category was split apart by regulators four years after this book published. The book treats 'commercials' as one undifferentiated smart-money bloc, which was accurate for the report format available in 2005. In 2009 the CFTC introduced the disaggregated COT report, which breaks the old commercial category into producers/merchants, swap dealers, and managed money, because swap dealers (banks facilitating client positions, not necessarily directional bets) had been muddying the old commercial figures. A reader applying this book's rules to the modern legacy COT report is working with a coarser, less accurate version of the same signal the book describes.
- The core case studies are all winners, chosen after the fact. Every chart in Chapters 4 through 10 (gold, cocoa, the British pound, crude oil, silver, soybeans, lumber, the yen) is presented as a case where the index called the turn. There is no systematic backtest anywhere in the book showing the win rate, drawdown, or false-signal rate of the COT index or WILLCO across a fixed universe of markets and dates; every example is a hand-picked chart with vertical lines already drawn on the winning entries.
- The 26-week WILLCO window and the 55/75/80 percent trigger levels are stated, not derived. Williams says the 26-week lookback is 'the best, across the board' after noting 'it varies from time to time and market to market,' and the specific trigger numbers (55 percent short-to-OI for a yen top, 75 to 80 percent for a WILLCO buy) come from a handful of named examples rather than a stated optimization or out-of-sample test. This is the edition-reviewed figure; the book gives no indication these thresholds were revisited in any later printing.
- It assumes you already have COT-aware charting software. The indicators are built and shown inside Genesis Financial Technologies software (the book credits Genesis for every chart), and the WILLCO formula is given as a named function ('Stochastic Custom (COT Commercials/Open Interest, vara)') meant to be typed into that platform. A reader without a charting package that can pull historical COT data and build custom stochastics has to build the plumbing themselves before any of the method is usable.
- Stock-index and financial futures are called out by the author as a weak fit for open interest analysis. Williams states plainly that the S&P 500 and Dow futures have a 'different set of problems' because open interest resets at quarterly contract expiration and includes heavy arbitrage flow, and that the reader should 'learn not to rely on OI by itself for these contracts.' That is one of the largest and most heavily traded futures categories, and the book's own author flags it as an exception rather than folding it into the general rule.
Who it is for
Buy it if
You already trade futures or currencies and want a concrete way to read the weekly Commitments of Traders report, with a specific formula for the COT index and WILLCO that you can rebuild in your own charting software and test yourself.
Skip it if
You came for the word 'stocks' in the title, or you want a book that hands you a backtested, ready-to-run system rather than a set of chart examples and an invitation to build the tools yourself. Traders without access to historical COT data or custom-indicator charting software will spend more time on infrastructure than on the method.
As an Amazon Associate I earn from qualifying purchases. It costs you nothing extra, and it does not change what gets recommended or what the verdict says.