5 people fully committed to a B+ strategy beats 5 people half-committed to an A strategy.
When you average two people's ideas, you get what a random person on the street would think—because you cut the extremes. You become a clone company taking poor decisions.
| Domain | Why |
|---|---|
| Real-time video games (LoL, CS:GO, Dota) | Millisecond decisions, no time to debate |
| Poker | Incomplete information, commitment signals strength, hesitation = death |
| Early-stage startups (<$10M ARR) | Resource-constrained, must bet on one thing |
| Disruptive tech (new category) | No existing playbook, must invent |
| Turnarounds / crisis | Survival mode, speed > correctness |
| Sales (closing deals) | Momentum matters, confidence sells |
| Creative products (film, games, design) | Vision coherence > feature checklist |
| Fast-moving markets (AI, crypto) | Windows close quickly |
| Military (tactical, real-time combat) | Hesitation = casualties |
| Startups competing against incumbents | Asymmetric warfare, need bold bets |
| Domain | Why |
|---|---|
| Turn-based strategy (Civilization) | Time to think between moves |
| Chess (preparation, not game itself) | Analysis benefits from multiple perspectives |
| Mature businesses (>$100M, stable market) | Optimization > innovation |
| Regulated industries (healthcare, finance) | Compliance errors costly |
| Infrastructure / platform (AWS, databases) | Reliability > speed |
| Scientific research | Truth-seeking, not speed |
| Manufacturing / operations | Process improvement is convergent |
| Legal / contracts | Precision matters more than speed |
| Large enterprise sales | Multiple stakeholders must align |
| Government / policy | Legitimacy requires buy-in |
| Variable | Single Strategy Wins | Consensus Works |
|---|---|---|
| Clock speed | Fast decisions required | Slow, time abundant |
| Reversibility | Irreversible, must commit | Reversible, can iterate |
| Information | Incomplete, uncertain | Complete, knowable |
| Competition | Zero-sum, real-time opponents | Positive-sum, no direct competitor |
| Error cost | Omission worse (missing opportunity) | Commission worse (wrong action) |
| Case | What Happened |
|---|---|
| Pontiac Aztek (GM) | Focus groups + multiple internal teams = "horrible least-common-denominator vehicle" |
| Healthcare.gov (2013) | 55+ contractors, multiple agencies, no single decision-maker = crashed on launch |
| Kodak | Invented digital camera in 1975, internal consensus protected film business, bankruptcy 2012 |
| Nokia | Engineer showed touchscreen prototype, management said "that's not how phones work", lost to Apple |
| Blockbuster | Board squabbles + complacency = rejected Netflix for $50M, worth $0 today |
| NATO Kosovo 1999 | "War by committee" - any member could veto any target on air strike list |
| Napoleon's enemies | Coalition armies required consensus among nations with different objectives. Napoleon had unified command. He won 53 of 60 battles. |
| Case | Why It Worked |
|---|---|
| Wikipedia | Low cost of iteration, reversible decisions, no time pressure |
| Linux/Open Source | Modular architecture, benevolent dictator for core decisions |
| Scientific consensus | Truth is objective, being wrong > being slow |
| Apollo 11 | Clear singular mission, hierarchical command, NASA admin made final calls |
| Toyota Production System | Manufacturing optimization is convergent (one right answer) |
| Jury system | False conviction (commission) worse than false acquittal (omission) |
- Military: A platoon half-flanking, half-charging gets slaughtered
- Rowing: 8 rowers at 80% in sync > 8 rowers at 100% out of sync
- Physics: Force split across vectors cancels out. Two equal forces at 180° = zero movement
- Music: A band playing the same wrong note together sounds intentional. Different "right" notes at different times = chaos
- Thiel: Averaging opinions = consensus = already priced in = no edge
"The clock speed of your environment determines whether unity or diversity wins."
Napoleon won 53 of 60 battles not because he had better troops or better maps. His enemies had committees. Austria, Prussia, Russia—coalition armies where every decision required consensus among generals who didn't trust each other. Napoleon had one brain. While they debated whether to flank left or charge center, he was already behind their lines.
His enemies explored 10^6 paths in the decision tree and chose the average. Napoleon explored 2^4 paths and chose the first one that wasn't stupid.
Speed is strategy. A mediocre plan executed violently today beats a perfect plan executed next week.
His generals learned: "If you start to take Vienna, take Vienna." No second-guessing. No averaging. No consensus.
His coalition enemies had more soldiers, more resources, more time to plan—and they lost precisely because they used that time. They optimized. He moved.
Draft for essay: "Clock Speed Theory: Why 5 People Committed to B+ Beats 5 People Split on A"