Story Point Estimation Techniques for Better Velocity
Compare Planning Poker, T-shirt sizing, affinity mapping, and other estimation techniques. Learn which method produces the most predictable velocity.
Detailed Explanation
Estimation Techniques That Improve Velocity Predictability
Story point estimation directly impacts velocity accuracy. Inconsistent estimation leads to unpredictable velocity. Here are the most popular techniques.
Planning Poker
The most widely used method in Scrum:
- Each estimator has cards with Fibonacci values (1, 2, 3, 5, 8, 13, 21)
- A story is presented and discussed
- Everyone reveals their estimate simultaneously
- If estimates diverge, the highest and lowest explain their reasoning
- Re-vote until consensus
Best for: Teams of 3-9 people, detailed sprint planning
T-Shirt Sizing
A faster, rougher technique:
XS = 1 point (trivial change, <1 hour)
S = 2 points (simple, half a day)
M = 5 points (moderate, 1-2 days)
L = 8 points (complex, 2-3 days)
XL = 13 points (very complex, needs splitting)
Best for: Roadmap-level estimation, large backlogs
Affinity Mapping
- Write all stories on cards
- Silently arrange them from smallest to largest
- Group similar-sized stories
- Assign point values to each group
Best for: Initial backlog estimation, re-calibration sessions
Which Produces the Best Velocity?
Research and practice suggest that Planning Poker produces the most accurate estimates for sprint-level planning because it forces discussion and surfaces hidden complexity. However, the best technique is the one your team uses consistently.
Tips for Better Estimation
- Always estimate relative to a well-understood reference story
- Never estimate alone -- group estimation catches blind spots
- Re-estimate stories that have been in the backlog for months
- Track estimate-vs-actual to calibrate over time
Use Case
Use this guide when establishing an estimation process for a new team or when your team's velocity has become unpredictable due to inconsistent estimates.