Maximizing Crafting Improvement Cost Efficiencies
Obviously many players are only concerned with getting that 100% chance of success when improving gear. Especially if the underlying gear is not easily replaceable. But for those of us who have enough mats for some wiggle room, I made this spreadsheet to help calculate the most cost-efficient number of improvement materials (tempers, tannins, or resins) that you should use to save gold in the long-run. It shows the average total cost of bringing a single piece of gear up through the quality ranks.
The spreadsheet uses conditional formatting only, no macros. It is not editable because too many variables need to be entered to make that feasible for multiple users, so you'll have to download it to use it properly. The conditional formatting might only show up once you make the file editable on your computer.
It requires the entry of several variables by the user; most of them are the costs of the various required (and one optional) materials. If you do not plan on using a Trait material, just leave the cost at zero. For everything else, I recommend first using the current market replacement costs. If you are swimming in materials, you should still assign them a cost to represent their value to you (perhaps representing what you could sell them for, or the opportunity cost of farming them).
Only the highlighted cells in the second column need to be changed. Don't forget your rank in the appropriate crafting Expertise skill. Once this data is entered, the lowest average total cost for each level of improvement should be highlighted in the charts. Use the number of improvement materials associated with that cost to maximize their efficiency over the long-run. You obviously can stop improvement at any time, but each level will include the cost of the previous levels.
The data is all laid out in a chart instead of just giving you the most efficient answer so that you can compare. Sometimes the savings of 90% success is so slight over 100% that it may not be worth it to you. It all depends on the costs of different materials. Usually (but certainly not always) it's better in the long-run to get as close to 100% without using that last improvement material, which doesn't give as big of a boost to the success rate as the others did (e.g., if each one gives 30%, that fourth is only giving an extra 10%), and might not be worth the lower cost of the other materials that would be required to start over from scratch. This is mainly because improvement materials are more rare, and cost more than the other materials.
Obviously, following this method you are going to have the occasional failures, but remember that it will save you gold in the long-run. It's especially important to keep costs down if you are regularly selling your crafted gear. You'll also see exactly how much the Expertise skills are worth.
For those into discrete mathematics, crafting improvements follow a geometric distribution. "Expected number of attempts" is equal to the expected value of that distribution. As long as the probability of success is close to 100%, the standard deviation will be small. For those who are into mass production and sales of crafted gear, you might want to try analyzing variance with negative binomial distributions; let me know if you do!
Link to the Spreadsheet on Google Docs
The spreadsheet uses conditional formatting only, no macros. It is not editable because too many variables need to be entered to make that feasible for multiple users, so you'll have to download it to use it properly. The conditional formatting might only show up once you make the file editable on your computer.
It requires the entry of several variables by the user; most of them are the costs of the various required (and one optional) materials. If you do not plan on using a Trait material, just leave the cost at zero. For everything else, I recommend first using the current market replacement costs. If you are swimming in materials, you should still assign them a cost to represent their value to you (perhaps representing what you could sell them for, or the opportunity cost of farming them).
Only the highlighted cells in the second column need to be changed. Don't forget your rank in the appropriate crafting Expertise skill. Once this data is entered, the lowest average total cost for each level of improvement should be highlighted in the charts. Use the number of improvement materials associated with that cost to maximize their efficiency over the long-run. You obviously can stop improvement at any time, but each level will include the cost of the previous levels.
The data is all laid out in a chart instead of just giving you the most efficient answer so that you can compare. Sometimes the savings of 90% success is so slight over 100% that it may not be worth it to you. It all depends on the costs of different materials. Usually (but certainly not always) it's better in the long-run to get as close to 100% without using that last improvement material, which doesn't give as big of a boost to the success rate as the others did (e.g., if each one gives 30%, that fourth is only giving an extra 10%), and might not be worth the lower cost of the other materials that would be required to start over from scratch. This is mainly because improvement materials are more rare, and cost more than the other materials.
Obviously, following this method you are going to have the occasional failures, but remember that it will save you gold in the long-run. It's especially important to keep costs down if you are regularly selling your crafted gear. You'll also see exactly how much the Expertise skills are worth.
For those into discrete mathematics, crafting improvements follow a geometric distribution. "Expected number of attempts" is equal to the expected value of that distribution. As long as the probability of success is close to 100%, the standard deviation will be small. For those who are into mass production and sales of crafted gear, you might want to try analyzing variance with negative binomial distributions; let me know if you do!
Link to the Spreadsheet on Google Docs
1
