Physics Lecture YouTube Problem Extraction Workflow
Watching someone else solve a physics problem feels like learning. It is mostly entertainment with equations. A physics lecture YouTube problem extraction workflow turns the video into a small set of solvable skeletons you can re-do with the player closed—because that is what exams and problem sets actually test.
The job sentence
Before play:
Job: From this lecture I will extract N problem types, each with given → find → model → pitfall, then re-solve at least one cold.
If you cannot name N (usually 2–5 for a one-hour lecture), you are about to take prose notes about math instead of building a practice set.
Pass 1 — Map the lecture by cognitive mode
Tag sections so you do not treat a derivation the same way as a worked example:
| Tag | What it is | First-pass behavior |
|---|---|---|
concept | Definition, symmetry argument, “why this model” | Timestamp + one-line claim |
derive | Formal derivation | Timestamp; do not copy every line yet |
demo | Qualitative experiment / simulation | What should happen + why (as taught) |
worked | Full problem solution | Mark start/end times; extract later |
trap | Common error callout | Capture verbatim-ish |
Play talk/concept faster if you can; drop to 1× for algebra-heavy boards and for the moment the lecturer chooses a coordinate system.
Done check: list the problem types in the video without looking at equations.
Pass 2 — Extract problem skeletons (not transcripts)
For each worked segment, pause after you understand the setup—or jump back once—and write a skeleton without the full algebra:
Skeleton template
- Given (symbols + units if stated)
- Find (target quantity)
- Model / assumptions (point mass, no friction, small angle, ideal gas, electrostatics approx, …)
- Principle choice (Newton II, energy, momentum, Faraday, …) — why this, not that
- Key equation set (names or final forms, not every intermediate scratch)
- Pitfall the lecturer flagged (sign, reference point, double-counting, mixing frames)
- Time range of the worked example
Weak extract: “They used energy and got the speed.”
Strong extract: Block slides off hemisphere. Given R, m, μ=0. Find speed at angle θ where contact lost. Model: frictionless first; constraint N=0 at loss. Principle: energy + radial Newton. Pitfall: using energy alone for loss-of-contact. 31:10–38:40.
Copying every board line trains handwriting, not problem selection.
Pass 3 — Re-solve cold (the real study)
Same day:
- Cover your skeleton’s algebra hints; keep only Given / Find / Model.
- Solve on blank paper with the video closed.
- Uncover and compare strategy, not just the final number.
- If you diverged early (wrong principle), rewatch only the principle-choice minute—not the entire simplification.
If you can only follow along while they write, you have not extracted a problem yet; you have rented a solution.
Pass 4 — Variation drills (exam transfer)
For each skeleton, invent one cheap variation:
- Change a parameter (add friction, change initial speed)
- Change the ask (find tension instead of speed)
- Change the representation (graph interpretation instead of numeric)
You do not need perfect textbook quality. You need to notice whether you understood the model choice.
Derivation sections: extract claims, not calligraphy
For derive tags, capture:
- Starting assumptions
- Intermediate named result (theorem, identity)
- Final boxed relation + validity limits
- One sentence: when you would use this result on a problem
Then try to reproduce the derivation once with cues, once colder if it is exam-critical. Daily lecture time is usually better spent on problem skeletons unless the course exams are derivation-heavy.
Exam-week physics video triage
| Block | Minutes | Action |
|---|---|---|
| Inventory | 10 | List all skeletons across unit videos |
| Cold solves | 40 | 3–5 problems from skeletons, mixed |
| Patch | 20 | Timestamp seeks for principle-choice misses only |
| Formula sheet honesty | 10 | Write relations from memory; mark gaps |
Watching solution compilations at 2× is not this table.
Pitfalls
- Solution tourism — many videos, zero cold solves.
- Equation collecting without assumptions.
- Unit blindness — numbers without dimensions in your skeleton.
- Ignoring the trap callout — that minute is often the exam.
- Calculator theater — arithmetic correctness hiding a wrong model.
Mini worked pattern: extract once, vary twice
Suppose a lecture works a block-on-incline problem with friction, then casually mentions the frictionless limit. Your notes should not be two full transcripts. Extract one skeleton with friction in the assumptions slot, then add a two-line variation:
- Variation A: μ → 0; which term drops; does the principle choice change?
- Variation B: ask for minimum μ to prevent slide; what becomes the unknown?
That is physics lecture YouTube problem extraction in practice: one video segment becomes a small family of problems you own. If you only copy the numeric answer from the board, you cannot generate Variation B on an exam.
When lectures chain three related demos (energy, then momentum, then both), write an explicit chooser line: “If the ask is speed after a rough slide → energy + nonconservative work; if the ask is direction after a collision → momentum (and energy only if elastic as stated).” Chooser lines prevent principle salad under time pressure.
Sign and coordinate discipline
Add a permanent line to every mechanics skeleton: positive direction / zero point for potential. Many “I followed the solution” failures are sign errors, not concept failures. If the lecturer spends time on a free-body diagram, timestamp that minute and make a card: “Redraw FBD for problem type T.”
Optional watch-page support
An on-page panel can help you chapter-mark worked segments, chat “what principle did they choose at 31:00?”, and turn pitfalls into Study flashcards. It cannot replace blank-paper solves. Use summaries to navigate; use paper to learn.
If you want that loop on YouTube—summary, chapters, chat, Study flashcards—SummarizAI is a Chrome extension built for students studying from lecture videos. The free plan works as a student trial. Keep Pass 3 non-negotiable either way.
Frequently asked questions
Should I pause and copy every derivation line?
Not on Pass 1. On a derivation you must reproduce for credit, do a guided recreate, then a cold recreate. For problem-solving courses, prioritize skeletons.
What if the lecture is conceptual with few numbers?
Extract “predict / explain” prompts as problems: initial conditions → qualitative outcome → which principle decides. Cold-explain with the player closed.
How many skeletons per lecture?
Usually 2–5 high-quality ones beat 12 half-copied solutions.
Is watching a second YouTuber’s solution helpful?
Only after you attempted cold. Otherwise you just collect accents for the same algebra.
What about lab or demo videos?
Map predicted vs observed as taught; extract the measurement claim and uncertainty note if present. Still separate from homework problem skeletons.
What is a physics lecture YouTube problem extraction workflow?
It turns the video into a small set of solvable skeletons (usually 2–5 per hour) you can re-do with the player closed: given → find → model → principle choice → pitfall, then at least one cold solve and a cheap variation. Watching someone else solve a physics problem feels like learning; exams and problem sets test whether you can select a model and solve it yourself.
Related guides
- CS Lecture YouTube Study Stack: Map, Code Claims, Practice
- The Best Way to Summarize YouTube Lectures for Exams
- Active Recall From YouTube Lectures: A Practical Study Loop
- How to Study From YouTube Lectures Without Rewatching the Whole Video
- How to Stop Rewatching Lecture Videos and Still Remember Them
- How to Retain Information From Video Lectures (Beyond Rewatching)
Try SummarizAI on your next lecture
SummarizAI is a Chrome extension that adds a summary, chapters, and Study flashcards on the YouTube watch page. The free plan is a student trial—no need to leave the lecture tab.
Start the free student trial