Shipwright Archive
Thread: Experimentation @ 0(zero) risk... Critical Failure!?
Lunariel wrote:
Here's a model of how it works.
Result = Rating + Roll
Risk = 100 - Rating (Any rating at 100 or higher shows as zero risk)
Rating: Calculated from your experiment skill and modified by average MA on the resources and how many points you spend. 900 MA is a 10 skill bonus, 100 MA is a -10 skill penalty. Each extra point used at the same time is a -5 skill penalty. The formula for the rating is:
50 + (Average_MA - 500)/40 + Experiment_Skill - 5*Points_Used
Average_MA follows the same rules as if it had a hidden 'MA 100%' experiment line. It is weighted by resource quantity and resources without MA are not counted. If none of the resources have MA, the Average_MA is considered zero which is not fair for Chefs and BEs. Points_Used is 1 if you only use one point.
Roll: Random roll. Rolling below a limit will always give a critical failure regardless of your rating and above another limit will always give amazing success. (Think of it as in RPGs where an original 1 or 20 on 20-sided die overrides your base chance.) Research City, Bespin Port and maybe crafting stations and tools modifies your roll before checking the limits and thus lessens your chance of critical failures. As well as improving your chance for amazing success. (The impact of crafting stations and tools seems to be fairly small if any.)
Risk: This is your chance of getting any kind of failure. Note that the lower results like marginally successful are not technically considered failures even though it feels like it.
At zero risk you can no longer get any type of failure except critical from the override on the roll.
Great Success gives you 7% and Amazing Success gives 8.05%. Good success and the others gives 3.5% and less. Normal Failure gives -7% and Critical Failure gives -14%. These percentages are multiplied by the number of points you spend. If you drop down to 0% from failures you can no longer increase the percentage.
Some general advice on experimenting. For conservative experimenting you should spend at least 2 points each time. Using 2 points effectively halves your chance of critical failure without reducing you chance of great success noticably. You may even do fine with 3 or 4 points at Master.
If you are experimenting on multiple lines and absolutely need almost only amazing successes for a schematic, you have to do another approach. Then you should spend as many points as possible and hope for the amazing override. The chance of getting many smaller amazing in a row is just too remote.
There are other variants. For example the bikes mentioned above where dropping to 0% really hurts. Don't start out with more than one point. With decent resources you will probably start above 14% and can survive an initial critical failure. Continue by never spending more points than the current percentage divided by 14.Stillif you drop down below 14 and get another critical nothing can save you.
Lunariel
