There's an old gambler's wisdom about knowing when to hold and when to fold. The manifesto borrows a cousin of it: know when to roll 'em. Manage your effort, dial in your approach, and don't push your luck. Only kick off a push when it's the right one — and just as importantly, know when the current one is done.
This is the principle people resist most, because it argues against a thing we've been taught to admire: relentless effort.
More effort is not more progress
Somewhere we absorbed the idea that working harder is always virtuous, that the team grinding late is the team that cares. But effort and progress are different quantities, and past a certain point they stop correlating. The tenth revision of a paragraph nobody will read. The gold-plating on a feature whose core hypothesis is still untested. The all-nighter to hit a date that didn't need hitting. That's effort spent like water, producing nothing but tiredness.
Pushing harder feels like doing something. Often it's the most expensive way to avoid the real decision, which is: is this still worth the push?
Manage your work effort, dial in your approach and don't push your luck. Only kick a push when it's the right one.
Diminishing returns have a shape
Most pushes follow the same curve. Early effort buys enormous value — the thing goes from nothing to working. Then the curve bends. You're now polishing, hardening, handling the long tail of cases that may or may not matter. The value of each additional hour drops, sometimes steeply.
The skill is feeling where you are on that curve. The first 80% and the last 20% cost about the same, and the last 20% is only worth it sometimes — for the parts that genuinely need it. Knowing which 20% deserves the effort, and which is luck-pushing, is most of the craft of shipping.
Don't push your luck
"Don't push your luck" is a specific warning. It's the moment when a thing is working, the result is in, and you decide to keep going anyway — one more refactor, one more feature, one more clever optimisation. Sometimes that's right. Often it's how a finished thing gets broken, or a clean result gets muddied, or a team that should be resting burns itself out on a push that already paid off.
A finished push has a beginning, a middle, and an end. Honouring the end — actually stopping, shipping, and moving on — is harder than it sounds, because stopping feels like giving up and continuing feels like dedication. Reframe it: stopping at the right moment is the dedication. It's what lets you spend the next push's effort where it'll actually count.
Roll the right one
The flip side is the start. "Only kick a push when it's the right one" is permission to not start things — to let an idea sit until it's earned the effort, rather than reflexively beginning work to look busy. A push you start for the wrong reason consumes the same effort as a good one and returns less.
So manage the effort like the finite resource it is. Start the right pushes. Spend hard where the curve is steep. And when a push is done, have the discipline to roll 'em, rack up the chips, and walk to the next table.
