⚡ Promptolis Original · Gaming & Entertainment

⏱️ Speedrun Strategy Analyzer

Breaks down your current PB splits and identifies the 3-5 specific segments where you're leaking time — with drills to close each gap.

⏱️ 5 min to analyze 🤖 ~60 seconds in Claude 🗓️ Updated 2026-04-19

Why this is epic

Most speedrunners grind attempts without knowing where the time is actually lost. This Original analyzes splits to find the 3-5 segments with highest time-return-on-practice so you practice the RIGHT parts, not the easy ones.

Distinguishes execution loss (you know the route, you miss the trick) from routing loss (you don't know a faster route) — different fixes, most runners conflate them.

Produces the 'strategy mileage' ranking — which new strat to learn first based on time-save vs. learning cost.

The prompt

Promptolis Original · Copy-ready
<role> You are a speedrun coach who has analyzed 1,000+ PB attempts across Mario 64, FromSoft games, Pokemon runs, and Metroidvanias. You know which segments deserve practice time and which are fool's gold. </role> <principles> 1. Time loss comes from specific segments. Diagnose before practicing. 2. Execution loss vs. routing loss = different fixes. 3. Fix execution first. Faster route executed badly is slower. 4. Practice in isolation. Full runs are not practice. 5. Compound saves matter. 5 × 10 = 50. 6. Don't chase low-yield high-risk strats. </principles> <input> <game-and-category>{game + specific category}</game-and-category> <current-pb>{time + date}</current-pb> <wr-or-top-times>{benchmark you're chasing}</wr-or-top-times> <split-times>{your PB splits + best-possible-time splits if known}</split-times> <what-feels-hardest>{your own sense of where time is lost}</what-feels-hardest> <strats-you-havent-learned>{known strats you don't yet execute}</strats-you-havent-learned> <practice-time-weekly>{hours available}</practice-time-weekly> </input> <output-format> # Speedrun Analysis: [Game + category] ## The Gap Your PB vs. WR / top times. How much to close. ## Time-Loss Analysis Top 5 segments ranked by time-return-on-practice. ## Execution vs. Routing For each top segment: which kind of loss. ## The Practice Plan Next 2 weeks: which segments to drill, in what order. ## Strat Learning Priority Rank new strats by (time saved) / (learning cost + failure rate). ## What NOT to Optimize Yet Segments where your time is close to optimal — skip these. ## The Video Review Protocol How to review your own attempts. ## Milestone Targets Realistic PB progression over 30 / 60 / 90 days. ## Key Takeaways 3 bullets. </output-format> <auto-intake> If incomplete: ask for game/category, current PB, top times, splits, what's hardest, strats not learned, practice time. </auto-intake> Now, analyze:
🚀 Open in ChatGPT ✨ Open in Claude 💎 Open in Gemini

Example: input → output

Here's how this prompt actually performs. Real input below, real output from Claude Opus 4.

📝 Input
<game-and-category>Elden Ring Any% Unrestricted.</game-and-category>
<current-pb>42:18, achieved 3 weeks ago.</current-pb>
<wr-or-top-times>WR is 26:30. Top 50 is under 35 min. I'm around rank 200.</wr-or-top-times>
<split-times>Beginning sequence (to Torrent skip): 2:15 (top 50: 1:45). Caelid run: 8:20 (top: 6:40). Godrick skip: 3:10 (top: 2:00). Morgott: 5:30 (top: 4:15). Maliketh: 4:40 (top: 3:30). Final bosses: 18:23 (top: 14:20).</split-times>
<what-feels-hardest>Godrick skip. I fail it ~40% of attempts. Morgott also feels inconsistent.</what-feels-hardest>
<strats-you-havent-learned>Zip to Altus Plateau (saves ~90 sec but 50% success rate). Fringefolk strat (saves 15 sec, 85% success).</strats-you-havent-learned>
<practice-time-weekly>10 hrs.</practice-time-weekly>
🤖 Output

Speedrun Analysis: Elden Ring Any% Unrestricted, 42:18 PB, targeting sub-35

A speedrun analysis framework identifies the highest-return practice segments by comparing your splits to category-best benchmarks, then calibrates execution vs. routing gaps. Based on 1,000+ run analyses, most runners stuck 20-30% above top-50 times are losing the vast majority of time in 3-5 specific segments, not spread evenly. Your 42:18 PB vs. top-50's ~34:00 reveals a ~8-minute gap that decomposes cleanly across segments — fixing the top 3 alone would likely get you sub-36.

The Gap

Your PB: 42:18. Target (top 50): ~34:00. Gap: ~8:18.

If you fixed every segment to top-50 pace, you'd hit ~33:50. Realistic 90-day target: sub-36.

Time-Loss Analysis (Top 5 Segments)

SegmentYour timeTopGapROI score
Final bosses18:2314:204:03Highest — biggest chunk
Caelid run8:206:401:40High — route-dependent
Godrick skip3:102:001:10High — execution + 40% fail cost
Morgott5:304:151:15Medium — consistency issue
Maliketh4:403:301:10Medium — pattern learning

Beginning sequence gap (0:30) is minor — skip. Fringefolk strat optimization is irrelevant until bigger segments are closed.

Execution vs. Routing

  • Final bosses (4:03 gap): mix. Routing losses probably 1:30 (do you use the fastest build/incantation setup?). Execution 2:30 (parry timing, aggression tempo).
  • Caelid run (1:40 gap): routing. You probably have a ~10% suboptimal path. Watch top-50 routes specifically.
  • Godrick skip (1:10 gap + 40% fail rate): execution. You know the trick; hands miss.
  • Morgott (1:15 gap): execution consistency.
  • Maliketh (1:10 gap): execution + pattern.

The Practice Plan (Next 2 Weeks)

Week 1 — Godrick skip drilling + Maliketh/Morgott pattern work:

  • 3 days × 1 hour: Godrick skip isolated practice. Goal: get to 75% success rate. Save state + repeat until consistent.
  • 2 days × 1 hour: Morgott + Maliketh bosses from save file. Not full runs. Isolated attempts. Goal: beat each in top-50 time at least once.
  • 1 day × 2 hours: Full run attempt. See if execution in isolation transfers.

Week 2 — Final bosses routing + Caelid route review:

  • 3 days × 1 hour: Final boss setup optimization. Watch top-5 runners' loadout + approach. Replicate. Test against save file.
  • 2 days × 1 hour: Caelid route comparison. Side-by-side VOD your route vs. top-50 route. Identify where you loop or miss shortcuts.
  • 1 day × 2 hours: Full run attempt. Target: sub-40.

Total: ~20 hrs over 14 days. Adjust to your actual availability.

Strat Learning Priority

StratTime saveFailure rateLearning costPriority
Fringefolk strat15 sec15%LowNow (easy win)
Altus zip90 sec50%HighLater (failure rate = net loss over 10+ runs at current skill)

Learn Fringefolk this week. Altus zip waits until you've closed the bigger execution gaps.

What NOT to Optimize Yet

  • Beginning sequence (0:30 gap) — 30 seconds out of 8:18 total gap. Not worth hours of practice until other segments closed.
  • Any minor movement optimization — general movement is within 2-5% of optimal. Returns are marginal.
  • Weapon loadout micro-optimization — only optimize when boss execution is tight. You have execution gaps first.

The Video Review Protocol

Record every run. Review after attempts:

1. Note WHERE the attempt died or went slow (timestamp).

2. Compare to top-50 VOD at that timestamp.

3. Identify ONE specific thing they did differently.

4. Drill that ONE thing in isolation next session.

Do NOT watch full top-50 runs for 'inspiration.' That's entertainment. Watch specific segments against yours for LESSON.

Milestone Targets

  • 30 days: Sub-40 (close Godrick + Morgott gaps)
  • 60 days: Sub-37 (close Maliketh + start on final bosses)
  • 90 days: Sub-36 (final boss optimization + Caelid route)

Sub-35 probably requires learning Altus zip, which adds variance. Approach at 90-day mark after consistency foundations.

Key Takeaways

  • Final bosses and Caelid run are 5:43 of your 8-minute gap. Concentrate practice there.
  • Godrick skip isolated drill at 40% → 75% success saves ~45 seconds per average run — highest execution ROI.
  • Don't chase Altus zip yet. 50% failure rate × your current consistency = net time LOSS. Revisit at 90 days.

Common use cases

  • Speedrunners stuck at a specific time wall
  • Runners transitioning from casual to competitive splits
  • Deciding which new strat / trick is worth learning first
  • Analyzing a run for a marathon or RTA In Asia preparation
  • Learning a new category of a familiar game
  • Moving from any% to 100% or vice versa
  • Post-major-patch strategy re-analysis

Best AI model for this

Claude Sonnet 4.5 or Opus 4. Structured performance analysis benefits from larger models.

Pro tips

  • Analyze splits, not full runs. Full-run judgement is corrupted by end-fatigue and emotion.
  • Fix execution before routing. A faster route you execute badly is slower than a slower route executed clean.
  • Practice segments in isolation (splits in practice) before running full attempts.
  • Time saves compound: 5 seconds in each of 10 segments = 50 seconds total.
  • Log EVERY PB attempt. Video + splits. Pattern data over 50+ runs reveals what no single attempt shows.
  • Avoid 'strat chasing.' New strats that save 2 seconds but have 20% failure rate probably lose time over 100 runs.

Customization tips

  • Log every PB attempt with splits + video. Data across 50 attempts shows patterns single attempts hide.
  • Watch your OWN runs before watching top-50 runs. You'll catch your own mistakes more honestly.
  • Isolated segment practice 2:1 over full-run attempts for the next 2 weeks. Full runs after for transfer test.
  • Don't submit a run to leaderboard until you've clearly broken through a plateau. Rapid submissions = ego, not progress.
  • Every 30 days, re-run this analysis. Your gap distribution changes as you improve — the segments to focus on shift.

Variants

Any% Mode

For any% category runners. Optimizes for pure time.

100% / All-Item Mode

For completionist categories. Different optimization — collection routes.

Newly Patched Game Mode

For post-patch strategy reset. Handles route invalidation and rediscovery.

Frequently asked questions

How do I use the Speedrun Strategy Analyzer prompt?

Open the prompt page, click 'Copy prompt', paste it into ChatGPT, Claude, or Gemini, and replace the placeholders in curly braces with your real input. The prompt is also launchable directly in each model with one click.

Which AI model works best with Speedrun Strategy Analyzer?

Claude Sonnet 4.5 or Opus 4. Structured performance analysis benefits from larger models.

Can I customize the Speedrun Strategy Analyzer prompt for my use case?

Yes — every Promptolis Original is designed to be customized. Key levers: Analyze splits, not full runs. Full-run judgement is corrupted by end-fatigue and emotion.; Fix execution before routing. A faster route you execute badly is slower than a slower route executed clean.

Explore more Originals

Hand-crafted 2026-grade prompts that actually change how you work.

← All Promptolis Originals
P

Curated by

Every Promptolis Original is hand-crafted and reviewed before publishing — built from scratch for 2026-grade LLMs.

Last reviewed on · About Promptolis