Between 1948 and 1963, twenty men wrote the papers that quietly contain the entire architecture of a real trading edge. Three of them worked in the same Bell Labs hallway. One was an outsider at IBM whose only job was to warn everyone else. The rest expanded the map.
Almost nobody has read all of them. That is the trade.
The industry has spent seventy years arguing about one layer of the stack while the other five sit in the library, free, waiting. Below is the canon. Read it in order. Then stop paying for filters.
1. The Signal Problem
Before you trade anything, you have to prove there is something in the noise worth trading.
"The fundamental problem of communication is that of reproducing at one point either exactly or approximately a message selected at another point." - Claude Shannon, 1948
Every filter, every model, every AI swarm is a machine that tries to solve this problem for markets. Shannon proved a ceiling above every one of them. The channel has finite capacity. No engineering breaks it. Most of what a filter finds is noise smoothed into a shape.
"The advantage scientists bring into the game is less their mathematical or computational skills than their ability to think scientifically." - Jim Simons
Simons ran 66% a year for 30 years. He didn't win by building better shovels. He won by picking channels where signal still existed.
The first question of every trade is not "what does my model say." It is "does this channel have signal left in it, or has the crowd already extracted it."
2. The Distribution Problem
Once you find a signal, you have to know what the tail of the distribution actually looks like.
"Clouds are not spheres, mountains are not cones, coastlines are not circles, and bark is not smooth, nor does lightning travel in a straight line." - Benoit Mandelbrot
Neither do price returns. The bell curve is the fantasy funds run on. The real distribution has fat tails, wild variance, and rough repeating patterns at every scale. The 10-sigma move is not once in a hundred lifetimes. In markets it shows up on a Tuesday.
"Risk is randomness with knowable probabilities. Uncertainty is randomness with unknowable probabilities." - Frank Knight, 1921
Every fund confuses one for the other. That confusion is where blowups live.
"All models are wrong, but some are useful." - George Box
Useful is not true. Useful is a tool with a shelf life. Every model dies eventually. The only question is whether you notice before your account does.
3. The Sizing Problem
A signal you cannot size is worth nothing.
In 1956, John Kelly proved that a single fraction of your capital maximizes the long-term growth rate of any strategy. Bet less and you leave money on the table. Bet more and the first drawdown ends the game before the edge shows up.
Ed Thorp took Kelly's paper, applied it to blackjack, then to markets, and ran 20% a year for 30 years. He'll tell you free that the signal was never the edge. The sizing was.
"It's not whether you're right or wrong that's important, but how much money you make when you're right and how much you lose when you're wrong." - George Soros
The market never pays you for being right. It pays you for sizing being right when you are.
"Rule number one: never lose money. Rule number two: don't forget rule number one." - Warren Buffett
The Kelly criterion for people who never took a math class.
4. The Compounding Problem
Even a real, well-sized edge decays. What compounds is you.
"Knowledge and productivity are like compound interest." - Richard Hamming, 1986
Hamming spent his life at Bell Labs asking why so few brilliant people did great work. The answer was not horsepower. It was refusing to quit on the one thing that was actually working.
"The big money is not in the buying and selling, but in the waiting." - Charlie Munger
Every strategy has a drawdown that looks like the end. The people who quit at that drawdown are the ones the market takes tuition from.
"Time is the friend of the wonderful company, the enemy of the mediocre." - Warren Buffett
Same for strategies. Same for accounts.
5. The Discipline Problem
You are the least reliable component in the system.
"The first principle is that you must not fool yourself, and you are the easiest person to fool." - Richard Feynman, 1974
Every backtest that fits perfectly is a lie you told yourself. Every "this time is different" is a lie the market told you and you agreed with.
"The stock market is never obvious. It is designed to fool most of the people most of the time." - Jesse Livermore
Written in 1923. Aged like a diamond.
"Being too far ahead of your time is indistinguishable from being wrong." - Howard Marks
You can be right about a signal and still bleed out for two years being right. Discipline is what keeps you at the desk when your edge is early.
6. The Reality Problem
The final layer is remembering the map is not the territory.
"Models are like passports. They open some doors and close others." - Emanuel Derman
Every model captures a piece of the market. None captures all of it. Trading a model as if it captured everything is the shortest path to a graveyard.
"The essence of risk management lies in maximizing the areas where we have some control over the outcome while minimizing the areas where we have absolutely no control." - Peter Bernstein
You cannot control the tail. You control the size, the frequency, and whether you show up tomorrow.
Andrew Lo added the modern layer with his Adaptive Markets Hypothesis. The edge you have this year decays because the world updates on your presence. The strategy that survives is the one that adapts before the market kills the old one.
The Stack
Six layers. Twenty teachers. All of them free by 1980.
- Prove the channel has signal. - Shannon
- Know the distribution has a tail. - Mandelbrot
- Size the bet to survive the tail. - Kelly, Thorp
- Compound the small edge across years. - Hamming, Munger
- Stay disciplined while the edge is early. - Feynman, Livermore, Marks
- Update when the market updates on you. - Lo, Derman
The Close
Every fund that dies misses one of the six. Every AI swarm hype thread argues about layer one. The edge was always the stack.
Twenty people wrote it down. They gave it away. The tuition is still open.
You just have to read them.
