The Fibonacci Sequence: A Quick Refresher
The Fibonacci sequence is a series of numbers where each value is the sum of the two before it: 0, 1, 1, 2, 3, 5, 8, 13, 21, 34, 55, 89… For planning poker purposes, the practical deck starts at 0 and typically ends at 21 (or 34 for teams that estimate epics).
The key property of the sequence is that the gaps between numbers grow as the numbers grow. The gap between 1 and 2 is 1. The gap between 8 and 13 is 5. The gap between 13 and 21 is 8. This is not a coincidence — it is the reason the sequence was chosen for agile estimation.
Why Fibonacci Works for Planning Poker
Human beings are good at relative comparisons and poor at absolute predictions. We can reliably judge "this is bigger than that" but struggle to say "this will take exactly 14 hours." The Fibonacci scale takes advantage of comparative reasoning while eliminating the false precision of finer-grained scales.
The gaps prevent unproductive debates. If your scale includes 6 and 7, teams will spend time debating the difference between them — a distinction that is almost certainly not meaningful. Fibonacci removes those intermediate options. You cannot pick 6 or 7; you must choose between 5 and 8. This forces a cleaner decision: is this story closer to "medium" or "large"?
Large gaps reflect genuine uncertainty. The jump from 13 to 21 is not arbitrary — it reflects the reality that a 21-point story is substantially more uncertain than a 13-point story. Our confidence in estimates degrades rapidly as task complexity grows. A story you feel confident estimating at 3 points is qualitatively different from one where 13 feels like a lower bound.
The Full Planning Poker Deck Explained
A standard Fibonacci planning poker deck contains the following cards:
- 0 — the story is already done, or requires no effort (documentation update, config change).
- ½ or 0.5 — trivial work, often omitted in favour of rounding up to 1. Some teams use it for very small chores.
- 1 — a simple, well-understood story. Minimal complexity, minimal uncertainty.
- 2 — straightforward work with clear scope and minor complexity.
- 3 — moderate complexity. More moving parts than a 2, but still well-understood.
- 5 — notable complexity. Requires thought, touches multiple components, or has some uncertainty.
- 8 — significant complexity. The team will need to think carefully about approach. May benefit from a design session before estimation.
- 13 — large and complex. Consider splitting. Deliverable in one sprint by a capable team, but carries meaningful uncertainty.
- 21 — very large. Almost always a signal to decompose before committing.
- ? — insufficient information to estimate. Request clarification or schedule a spike.
- ☕ — the team needs a break.
Fibonacci vs. Other Planning Poker Decks
Powers of two (1, 2, 4, 8, 16, 32) — similar philosophy to Fibonacci but with larger gaps. Good for teams that find Fibonacci debates unproductive. The tradeoff is less granularity in the middle range.
T-shirt sizes (XS, S, M, L, XL) — non-ordinal and non-numeric, which makes velocity calculation impossible but keeps early backlog triage fast and accessible for non-technical stakeholders.
Linear scales (1–10) — provide the false comfort of precision without the benefits. A linear scale invites exactly the kind of "6 vs. 7" debates that Fibonacci is designed to prevent.
Hours — useful when stakeholders need time-based forecasts, but reintroduces all the problems of absolute estimation. Hour estimates are person-specific, context-specific, and don't aggregate well into team-level velocity.
For most agile teams doing sprint-level estimation, Fibonacci is the right default. Start with it, and deviate only if you have a specific reason.