COMPARISON GUIDE · OCTOBER 11, 2026
Schedule 1 guide: compare effect values, not just effect counts
This guide solves one decision: how to compare two calculator outputs when both show the same number of effects. The figures come from an independently executed saved engine. They are not completed sales, production measurements or a promise about a later game build. The comparison keeps the base product and number of additions fixed.
A controlled check of 256 recipes
We used OG Kush and every ordered pair of the engine’s 16 ingredients. Repetition is allowed, so the grid contains 16 × 16 = 256 recipes. Each run has exactly two additions. We grouped the resulting recipes by their final effect count, then counted distinct rounded sell values and found each group’s minimum and maximum.
| Final effects | Ordered recipes | Sell-value range | Distinct sell values |
|---|---|---|---|
| 2 | 19 | $39–$64 | 11 |
| 3 | 237 | $45–$84 | 38 |
The two-effect group contains 19 ordered recipes and 11 distinct sell values. The three-effect group contains 237 recipes and 38 distinct sell values. The group counts sum to 256; an ordered recipe is a separate input even when it produces the same effects or price as another input. These counts are not visitor data or probabilities of an in-game outcome.
A count of three therefore does not identify one value. In this grid it covers $45 through $84, a spread of 84 − 45 = $39. The two-effect group spans $39 through $64, a spread of $25. The groups overlap: a two-effect result at $64 exceeds a three-effect result at $45 by $19.
Keep the recipes attached to the extremes
A range is useful only if its endpoints can be reproduced. The following rows are the actual minimum and maximum sell-value examples selected from each count group. Ingredient order is shown with arrows. The weight column is the sum of the final effect multipliers.
| Ordered additions | Count | Final effects | Weight sum | Sell value | Ingredients | After ingredients |
|---|---|---|---|---|---|---|
| Gasoline → Gasoline | 2 | Calming, Toxic | 0.10 | $39 | $10 | $29 |
| Mouth Wash → Mouth Wash | 2 | Anti-Gravity, Balding | 0.84 | $64 | $8 | $56 |
| Cuke → Gasoline | 3 | Calming, Euphoric, Toxic | 0.28 | $45 | $7 | $38 |
| Cuke → Mega Bean | 3 | Glowing, Cyclopean, Foggy | 1.40 | $84 | $9 | $75 |
Gasoline → Gasoline retains Calming and Toxic. Its $10 ingredient expense leaves $29 after ingredients. Mouth Wash → Mouth Wash finishes with Anti-Gravity and Balding; its $8 expense leaves $56. Both have two final effects, but their sell values differ by $25 and their after-ingredient values differ by $27. The cost difference adds $2 to the value difference.
Read the final effect weights, then calculate value
The saved engine stores OG Kush’s base value as $35 and rounds the final unit sell value from 35 × (1 + sum of final effect weights). The important word is final. Ingredients may transform effects that were present earlier, so the starting effect list is not a substitute for the output list.
For Cuke → Gasoline, the final weights are Calming 0.10, Euphoric 0.18 and Toxic 0.00. Their sum is 0.28. The calculation is 35 × 1.28 = 44.8, which becomes $45 in this engine. Three names appear in the result, but one contributes no value multiplier.
For Cuke → Mega Bean, the final weights are Glowing 0.48, Cyclopean 0.56 and Foggy 0.36. Their sum is 1.40. The calculation is 35 × 2.40 = $84. It also has three names; their weights are different. The raw-value difference is 84 − 44.8 = 39.2, while the displayed rounded-value difference is $39.
The same check explains the two-effect endpoints. Calming plus Toxic totals 0.10, giving 35 × 1.10 = 38.5 and a rounded $39. Anti-Gravity plus Balding totals 0.54 + 0.30 = 0.84, giving 35 × 1.84 = 64.4 and a rounded $64. Effect count is absent from this formula; effect identity supplies the weights.
Use a comparison ledger instead of one score
Record four fields beside each candidate: ordered additions, final effects, rounded sell value and total ingredient cost. Compute after-ingredient value by subtracting the fourth field from the third. Keep actual base production expense separate rather than silently treating it as zero.
For the three-effect pair in the first answer, the after-ingredient difference is (84 − 9) − (45 − 7) = 75 − 38 = $37 per unit. It is smaller than the $39 sell-value difference because the higher-value recipe costs $2 more in ingredients. For a hypothetical batch of 20 identical units, the difference is 20 × 37 = $740. This is a derived quantity comparison, not evidence of 20 sales.
If both candidates have the same measured base production cost P per unit, their net difference remains $37: (75 − P) − (38 − P) = 37. If the candidates have different production costs, customer payments or deductions, those differences need their own recorded inputs. This experiment does not supply them.
Questions when a result looks stronger
Does one more effect always make a recipe more valuable?
No. In this fixed two-addition grid, the $64 two-effect endpoint exceeds the $45 three-effect endpoint. Check final weight sums before interpreting the count. This is a concrete counterexample in the saved engine, not a statement about every longer recipe.
Can two outputs have the same count and different costs?
Yes. The two-effect examples use $10 and $8 of ingredients; the three-effect examples use $7 and $9. Count describes the final set size, not the expense needed to reach it. Record cost independently.
Can this grid choose the best recipe in the whole game?
No. It fixes one base product and exactly two additions, uses standard mixing, and excludes longer sequences, seeded mixing and actual customer requirements. It can expose a faulty comparison shortcut within that scope. A different decision needs a separately defined test.
Make the next comparison auditable
When two candidates have equal effect counts, compare their final names first. Reconstruct the value weights, confirm the rounded sell values and then subtract ingredients. Retain both input sequences with the result. That makes the $84 versus $45 comparison explainable instead of reducing it to “both have three effects.”
The audit record contains all 256 ordered recipes, effect codes and names, multiplier sums, costs, sell values, group endpoints and source SHA-256. The group ranges, distinct-value counts and paired arithmetic are original calculations from those saved runs.