Agile-and-Scrum: Agile and Scrum, 2020 Guide - Velocity, Burndown/Burn-up Charts, and Forecasting

Agile & Scrum (2020 Guide): Velocity, Burndown/Burn-up Charts, and Forecasting

A Hyper-Practical, Zero-Fluff Study Guide

1. What This Is & Why It Matters

You’re a Scrum Master or Product Owner on a team that just inherited a legacy project with no historical data. The CTO asks, "When will we hit MVP?" Your gut says "3 months," but stakeholders want hard dates. Velocity, burndown/burn-up charts, and forecasting are your superpowers here—they turn gut feelings into data-driven predictions.

Why this matters in production: - Without velocity, you’re guessing. Guessing leads to missed deadlines, overpromising, and death marches. - Without burndown/burn-up charts, you can’t spot blockers early. Teams waste sprints "hoping" they’ll catch up. - Without forecasting, you can’t answer "When will Feature X ship?" with confidence. Stakeholders lose trust.

Real-world scenario: You’re leading a cloud migration. The team says, "We’ll finish in 6 sprints." But after Sprint 1, you realize they’re completing only 20 story points instead of the planned 40. Do you: A) Hope they’ll magically double output next sprint? B) Use velocity to adjust forecasts and reset expectations? (Spoiler: B is the only way to avoid a death march.)

2. Core Concepts & Components

? Velocity

• Definition: The average number of story points a team completes per sprint (e.g., "Our velocity is 35 points").
• Production insight: Velocity is not a KPI—it’s a planning tool. If you use it to "measure productivity," teams will inflate estimates to game the system.
• Key detail: Only count done work (Definition of Done). If a story is "90% done," it counts as 0 in velocity.

? Story Points

• Definition: A relative unit of effort (e.g., "This feature is 5 points, that one is 13").
• Production insight: Story points are not hours. If your team starts saying "1 point = 1 day," you’ve failed. Points measure complexity, not time.
• Example: A "Login Page" might be 3 points. A "Payment Integration with Stripe" might be 21 points.

? Burndown Chart

• Definition: A graph showing remaining work (y-axis) vs. time (x-axis) in a sprint.
• Production insight: A flat line in a burndown means blockers. A steep drop at the end means crunch time. Neither is healthy.
• Example: Day 1: 40 points Day 5: 25 points Day 10: 0 points (ideal)

? Burn-up Chart

• Definition: A graph showing completed work (y-axis) vs. time (x-axis), with a line for total scope.
• Production insight: Burn-ups are better for scope changes. If the "total scope" line jumps mid-sprint, you’ll see it immediately.
• Example: Sprint 1: 30/50 points Sprint 2: 60/70 points (scope increased by 20)

? Forecasting

• Definition: Using velocity to predict when a backlog will be completed.
• Production insight: Forecasts are probabilistic, not promises. Always give ranges (e.g., "We’ll finish between 5–7 sprints").
• Formula: Forecast = (Total Backlog Points) / (Average Velocity) Example: 200 points backlog / 35 velocity = 5.7 sprints-6 sprints.

? Yesterday’s Weather

• Definition: A forecasting rule: "Assume next sprint’s velocity will match the last 3 sprints."
• Production insight: This is the most reliable way to forecast. If your team did 30, 35, and 28 points in the last 3 sprints, plan for ~31 points next sprint.
• Why it works: Teams are consistent. If they suddenly double output, it’s usually due to lower quality or inflated estimates.

? Monte Carlo Simulation (Advanced Forecasting)

• Definition: A statistical method to predict completion dates using random sampling of past velocities.
• Production insight: Use this for high-stakes forecasts (e.g., "When will we hit GA?"). Tools like Jira or Excel can run this for you.
• Example:
• Input: Last 5 sprints’ velocities: [30, 35, 28, 32, 31]
• Output: "85% chance of finishing in 6–8 sprints."

3. Step-by-Step Hands-On: Forecasting a Backlog

Prerequisites

• A Scrum team with at least 3 sprints of historical data.
• A backlog with estimated story points.
• A tool (Jira, Excel, or even a whiteboard).

Step 1: Calculate Average Velocity

1. List the last 3–5 sprints’ completed story points. Example: Sprint 1: 30 Sprint 2: 35 Sprint 3: 28 Sprint 4: 32 Sprint 5: 31
2. Calculate the average: (30 + 35 + 28 + 32 + 31) / 5 = 31.2-31 points (round down)

Step 2: Estimate Total Backlog

1. Sum all story points in the backlog. Example: Feature A: 21 Feature B: 13 Feature C: 8 Bugs: 5 Tech Debt: 3 Total = 50 points (Note: Only count "ready" stories—no epics or unrefined items.)

Step 3: Forecast Completion

1. Divide total backlog by average velocity: 50 points / 31 velocity = 1.6 sprints-2 sprints
2. Add a buffer (because life happens):
3. Best case: 1 sprint
4. Likely: 2 sprints
5. Worst case: 3 sprints

Step 4: Create a Burn-up Chart (Excel Example)

1. Open Excel and create a table: | Sprint | Completed | Total Scope | |--------|-----------|-------------| | 1 | 30 | 50 | | 2 | 60 | 50 | (Note: "Total Scope" can increase if new work is added.)

2. Insert a line chart:

3. X-axis: Sprint
4. Y-axis: Points

5. Lines: "Completed" and "Total Scope"

6. Interpretation:

7. If "Completed" line flattens, the team is blocked.
8. If "Total Scope" line jumps, scope is increasing.

Step 5: Adjust for Reality

• If velocity drops: Reforecast. Example: If velocity drops to 25, 50 points takes 2 sprints.
• If scope increases: Update the "Total Scope" line. Example: If scope jumps to 70 points, 70 / 31 = 3 sprints.
• If stakeholders demand faster delivery:
• Option 1: Reduce scope.
• Option 2: Increase velocity (but this is hard—usually requires more people or less work).

4.-Production-Ready Best Practices

? Velocity Best Practices

• Never compare velocities across teams. A 50-point team is not "better" than a 20-point team. Points are relative.
• Use a rolling average (last 3–5 sprints). Don’t use just the last sprint—it’s noisy.
• Exclude outliers. If a sprint had a 50-point anomaly (e.g., a team member was out sick), exclude it from the average.
• Re-estimate if velocity changes. If velocity drops from 35 to 25, reforecast the backlog.

? Burndown/Burn-up Best Practices

• Update charts daily. If you wait until the end of the sprint, it’s too late to fix blockers.
• Use burn-ups for scope changes. Burndowns hide scope creep.
• Add a "trend line" to burndowns. This shows if the team is on track to finish.
• Color-code blockers. Example: Red = blocked, Yellow = at risk, Green = on track.

? Forecasting Best Practices

• Always give a range. Example: "We’ll finish in 4–6 sprints (85% confidence)."
• Use Monte Carlo for high-stakes forecasts. Example: "There’s a 70% chance we’ll hit MVP by Q3."
• Reforecast after every sprint. Velocity changes—adjust plans accordingly.
• Communicate uncertainty. Example: "This forecast assumes no major scope changes or team disruptions."

? Tooling Best Practices

• Jira: Use the "Velocity Chart" and "Release Burndown" reports.
• Excel/Google Sheets: Build your own charts (see Step 3).
• Physical boards: Use sticky notes for burndowns (great for co-located teams).
• Advanced: Use tools like ActionableAgile for Monte Carlo simulations.

5. Common Mistakes & Traps

6.-Exam/Certification Focus

Typical Question Patterns

1. "What is velocity?"
2. ? Wrong: "How fast the team works."

3. Correct: "The average number of story points a team completes per sprint, used for forecasting."

4. "When should you use a burn-up vs. burndown chart?"

5. ? Wrong: "Burn-ups are for releases, burndowns are for sprints."

6. Correct: "Use burn-ups when scope changes frequently (e.g., product backlog). Use burndowns for fixed-scope sprints."

7. "How do you forecast a backlog?"

8. ? Wrong: "Divide total points by the team’s capacity."

9. Correct: "Divide total backlog points by the team’s average velocity (last 3–5 sprints)."

10. "What’s the biggest risk of using velocity for performance reviews?"

11. ? Wrong: "It’s not accurate."
12. ? Correct: "Teams will inflate estimates to hit targets, making velocity meaningless."

Key Trap Distinctions

• Velocity vs. Capacity:
• Velocity = historical output (points/sprint).
• Capacity = available time (e.g., "We have 80 hours this sprint").

• Trap: Forecasting with capacity instead of velocity.

• Burndown vs. Burn-up:

• Burndown = remaining work (good for sprints).
• Burn-up = completed work + total scope (good for releases).

• Trap: Using a burndown for a release with changing scope.

• Story Points vs. Hours:

• Story points = relative complexity.
• Hours = time.
• Trap: Estimating in hours (leads to micromanagement).

Scenario-Based Question

"Your team’s velocity is 30 points/sprint. The backlog has 150 points. Stakeholders want it done in 4 sprints. What do you do?" --Wrong: "Work harder!" --Correct: 1. Forecast: 150 / 30 = 5 sprints. 2. Communicate: "We’ll finish in 5 sprints. To hit 4, we’d need to reduce scope by 30 points or increase velocity (unlikely)." 3. Options: - Negotiate scope. - Add team members (but this may slow velocity short-term). - Accept the 5-sprint timeline.

7.-Hands-On Challenge

Challenge: Your team’s last 5 sprints: [25, 30, 28, 32, 29]. The backlog has 200 points. Forecast the completion date using:
1. Simple average velocity.
2. Yesterday’s weather (last 3 sprints).
3. Monte Carlo (simulate 100 sprints using random sampling).

Solution:
1. Simple average: (25 + 30 + 28 + 32 + 29) / 5 = 28.8-29 points/sprint 200 / 29 = 6.9-7 sprints

1. Yesterday’s weather: (28 + 32 + 29) / 3 = 29.6-30 points/sprint 200 / 30 = 6.6-7 sprints

2. Monte Carlo (Excel):

3. Use =RANDBETWEEN(25,32) to simulate 100 sprints.
4. Average the results (e.g., 28.7-7 sprints).
5. 85% confidence range: 6–8 sprints.

Why it works: - Simple average is easy but noisy. - Yesterday’s weather is more stable. - Monte Carlo gives a range, not a single number.

8.-Rapid-Reference Crib Sheet

9.-Where to Go Next

1. Scrum Guide 2020 – Official Guide
2. ActionableAgile – Monte Carlo Forecasting Tool
3. Jira Tutorial – Velocity & Burndown Charts
4. Book: "Agile Estimating and Planning" by Mike Cohn (for deep dives on forecasting).

Final Pro Tip: "Velocity is like a car’s speedometer—it tells you how fast you’re going, not how far you’ll get. Always pair it with a map (backlog) and a destination (goal)."