Why does it always go worse than planned?
summary
A classroom dice-based game models a linear handoff process with a Google Sheet and ten sequential rounds; each person rolls 1–6 and passes inventory, teams bet on highest yield. Theory predicts 3.5 per roll or expected 35 vs observed 12–21 items, revealing a non-ergodic asymmetry: upstream shortfalls damage downstream but excess output isn’t reusable. Proposal: smaller teams and timeboxes, add buffers, pay down technical debt, and redesign contracts to increase skin in the game and reduce ruin risk.
AI-written summary · about
chapters (17)
- 0:00Workshop introduction
- 1:06Workshop format
- 4:40Game rules
- 15:46Debriefing results
- 21:14Ergodicity theory
- 23:51Non-ergodicity failures
- 27:45Risk exposure
- 28:50Defining ergodicity
- 29:52Software development ergodicity
- 31:02Ergodicity discussion
- 37:30Improving ergodicity
- 43:58Collaboration trade-offs
- 49:40Risk aversion biases
- 53:55Time management
- 57:31Anti-fragility principles
- 1:00:06Ludic fallacy
- 1:01:32Project contracts
related talks
47m[Workshop] NoEstimates: Forecasting with Less Effort and More Accuracy
45mStop rolling the dice: an approach to effort estimation in software development
39mGame Theory and the Art of Teamwork in Software Development
30mNavigating Organizational Dynamics at Google
43mReliability Equilibrium: The Hidden Playbook behind SRE Influence
13mPerfect Estimator in Strange New Worlds
05mAgile: "I have never" Game
50mCards Against people Mismanagement