The Method
Most explanations of world events are character studies. This one is not. The claim underneath all four books is that outcomes are produced by the structure of a game, and that the people inside it are largely interchangeable.
The test is simple: if you replaced every named individual in a situation with a different competent person facing the same rules and the same rewards, would the outcome change much? Usually it would not. What you are looking at is a structure, and structures are readable in advance.
The four questions
1. Who is actually playing?
Count the players, including the ones who never appear in the official diagram. A player is anyone who can make a choice that changes the outcome — which frequently excludes the person holding the title, and frequently includes a creditor, a supplier, a regulator, or a domestic faction that never gives interviews.
Getting this wrong is the most common analytical failure. A negotiation described as two-sided is usually four-sided, and the two extra players are the ones who decide it.
2. Which rules are actually enforced?
Not the rules that are written down. The rules that carry a real penalty when broken, applied to this player, in this decade.
A rule that is universally violated without consequence is not a constraint; it is decoration. A norm nobody has codified but everybody obeys is a hard constraint. Write down the second list, not the first.
3. What is each player genuinely rewarded and punished for?
Rewards run on a personal clock, not an institutional one. An official who will be out of office in three years is playing a three-year game regardless of what the mandate says. A family firm with a hundred-year horizon plays a different game in the same room.
Ask what happens to each player personally if they are wrong, and how long it takes for that to arrive. That interval explains more behaviour than ideology does.
4. Where does stated purpose diverge from produced outcome?
Take an institution. Write down what it says it is for. Write down what it reliably produces, measured over decades and not from its own annual report.
Where the two have diverged for a long time while remedies existed and were refused, the divergence is not failure. It is design — arrived at by selection rather than by conspiracy. Systems that produce a particular result keep the people who are comfortable with that result, and those people write the next set of rules.
That last clause matters. The method does not require anybody to have planned anything.
What comes out of it
Run the four questions and you get an equilibrium: the outcome each player arrives at when nobody can improve their own position by moving alone. Equilibria are stable. They are also, very often, bad for everyone inside them — which is the subject of Book Three.
Once you have the equilibrium you can ask the only question that matters for prediction: what would have to change for the equilibrium to break? Usually it is one of four things — a new player, an enforced rule that stops being enforced, a change in what somebody is paid for, or an outside shock that alters the payoffs. Nothing else moves a stable system.
The instrument, in one block
Every chapter of Book Two closes with the same six-line block. It is small enough to use on something in your own week.
Players — everyone who can make a choice that changes the result.
Payoffs — what each is rewarded and punished for, and on what timescale.
Constraints — the rules that are genuinely enforced.
Dominant move — the choice each player makes regardless of what the others do.
Equilibrium — where it settles when nobody can improve alone.
What would change the game — the specific event that breaks it.
What the method cannot do
It gives you structure, not timing. It will tell you that a currency arrangement is unsustainable and give you the mechanism by which it ends; it will not tell you the year. It is much better over thirty years than over three, because noise averages out and structure is what remains — the argument Book Four rests on.
It is also deliberately amoral. It describes what incentives push toward, not what anybody ought to do. A description of a mechanism is not an endorsement of it.
And it can be wrong. Book Two ends with a statement of what would falsify the framework, including where its author thinks it probably overreaches. A model that cannot fail cannot inform.