Last updated: December 2025
This analysis is based on real calculator outputs from testing various scenarios. All findings are derived from actual algorithm results, not theoretical assumptions.
Package Efficiency Analysis
Not all RP packages are created equal. Here's the efficiency (RP per Euro) of each package in the EUR region:
| Package | RP | Price | RP per € | Bonus vs Smallest |
|---|---|---|---|---|
| Smallest | 575 RP | €4.99 | 115.2 | - |
| €10.99 Package | 1380 RP | €10.99 | 125.6 | +9.0% |
| €21.99 Package | 2800 RP | €21.99 | 127.3 | +10.5% |
| €34.99 Package | 4500 RP | €34.99 | 128.6 | +11.6% |
| €49.99 Package | 6500 RP | €49.99 | 130.0 | +12.8% |
| €99.99 Package | 13500 RP | €99.99 | 135.0 | +17.2% |
| €244.99 Package | 33500 RP | €244.99 | 136.8 | +18.8% |
| Largest | 60200 RP | €429.99 | 140.0 | +21.5% |
The €10.99 Package Mystery - Solved!
Test Results:
- Standard scenario (80 pulls, 0 starting RP): €10.99 is NEVER used
- High starting RP (10000 RP, 50 pulls left): €10.99 appears 4 TIMES in optimal strategy!
- Few pulls needed (20 pulls): €10.99 appears once in optimal
Pattern Discovered: The €10.99 package is actually optimal when you need to "top up" a small amount. When you already have significant RP, buying a €99.99 package would be massive overkill. The €10.99 package fills the gap perfectly!
Example from testing: €10.99 -> €34.99 -> €10.99 -> €10.99 -> €10.99 was optimal for 10000 starting RP with 50 pulls needed.
The Two Optimization Strategies
1. Optimal Strategy (Minimize Expected Cost)
This strategy considers the probability of getting the item early. It uses the formula:
Expected Cost = Σ(Package Cost × Probability You Need It)
Key insights:
- Early packages in the strategy have 100% probability of being needed
- Later packages have decreasing probability as you might drop the item before needing them
- Cheaper packages early can be optimal even if they have worse RP/€ efficiency
- The algorithm evaluates packages based on expected value, not just efficiency
2. Most Bonus RP Strategy (Minimize Cost at Pity)
This strategy assumes worst-case: you need all pulls until pity. It optimizes for:
Minimum Total Cost to reach Pity RP
Key insights:
- Always prefers packages with highest RP/€ ratio
- Minimizes leftover RP at pity
- Ignores probability entirely - assumes you need everything
- Often uses larger packages exclusively (€99.99, €49.99)
Real Patterns From Testing
Finding #1: Strategy Convergence at Low Drop Rates
Test Case: 0.2% drop rate (very low), 80 pulls needed
- Optimal:
€99.99 -> €34.99 -> €99.99 -> €4.99 - Pity:
€99.99 -> €34.99 -> €99.99 -> €4.99 - IDENTICAL!
Finding #2: The €49.99 Spam Strategy
Test Case: 2% drop rate (high), 80 pulls needed
- Optimal:
€49.99 -> €49.99 -> €49.99 -> €49.99 -> €49.99 - Pity:
€34.99 -> €99.99 -> €99.99 -> €4.99 - Expected cost difference: €11.61
Finding #3: The €244.99 Package is Nearly Useless
Testing 6 different scenarios, the €244.99 package appeared only ONCE (in expensive pulls scenario with pity strategy). It's almost never optimal because:
- Too expensive for early purchase (high risk if you drop early)
- Efficiency gain over €99.99 is only 1.8 RP/€
- Two €99.99 packages give more flexibility
Finding #4: Small Packages First is Real
Standard scenario (80 pulls, 0.5% drop):
- Optimal:
€21.99 -> €21.99 -> €49.99 -> €99.99 -> €49.99 - Pity:
€34.99 -> €99.99 -> €99.99 -> €4.99
The optimal strategy starts with TWO €21.99 packages despite their lower efficiency. If you get lucky and drop early, you save money. The pity strategy jumps straight to large packages because it assumes worst-case.
Finding #5: €99.99 Dominance
Across all 6 test scenarios, the €99.99 package appeared in 5 out of 6 strategies (all except the "few pulls needed" case). It's the sweet spot of efficiency and cost.
Algorithm Performance Stats
Real Computation Data
From our test runs, here's what the algorithm actually does:
- Standard case (80 pulls): Generated 3,835 strategies
- High RP (50 pulls): Generated 1,551 strategies
- Expensive pulls: Generated 5,010 strategies
- Few pulls (20): Generated 1,021 strategies
Observation: The number of strategies grows with RP needed, not just pulls. Expensive pulls (1000 RP each) generated the most strategies because more RP is needed total.
Unique Solutions Found
Interesting pattern - the algorithm finds many strategies with identical costs:
- Standard case: 3,835 strategies → only 3,075 unique expected costs
- This means ~20% of strategies are tied for cost!
- Example:
€21.99 -> €99.99 -> ...often has same expected cost as€99.99 -> €21.99 -> ...
Counterintuitive Discoveries
Discovery #1: Order Sometimes Doesn't Matter
In the standard test, these had nearly identical expected costs (within €0.03):
€21.99 -> €21.99 -> €49.99 -> €99.99 -> €49.99€21.99 -> €49.99 -> €21.99 -> €99.99 -> €49.99€49.99 -> €21.99 -> €21.99 -> €99.99 -> €49.99
But only the first one is truly optimal! The algorithm found 5 top strategies all within €0.06 of each other.
Discovery #2: High Drop Rate Drastically Changes Strategy
The 2% drop rate case showed the most dramatic shift:
- Expected cost: €144.76 (optimal) vs €156.37 (pity) = €11.61 difference!
- Worst-case cost: €249.95 (optimal) vs €239.96 (pity) = €9.99 difference
The paradox: The optimal strategy has HIGHER worst-case cost but MUCH lower expected cost. You pay more if unlucky, but save significantly on average.
Discovery #3: The €4.99 Package Has a Purpose
The smallest package appeared in pity strategies more than optimal ones! It's used as a "filler" to minimize leftover RP:
- Pity example:
€34.99 -> €99.99 -> €99.99 -> €4.99 - The €4.99 tops off the exact RP needed instead of buying a larger package with more waste
Discovery #4: Strategy Diversity Increases With Drop Rate
- Low drop (0.2%): Only 2,888 unique costs from 3,835 strategies (75%)
- High drop (2%): 3,233 unique costs from 3,835 strategies (84%)
Higher drop rates create more differentiation between strategies because order matters more when you're likely to drop early.
Final Insight: This calculator evaluates thousands of possible package combinations in milliseconds. The optimal strategy is often counterintuitive - starting small, using "inefficient" packages in the right context, and carefully balancing expected vs worst-case costs. The mathematics of gacha optimization is far more nuanced than "just buy the biggest package."