Core Rates: additional notes
This page contains some extra commentary on core rates that are less relevant/interesting to most players.
My personal data is a bit rough (just going by what's in rescued_doll.csv) and has a few issues:
- I very rarely use Rescue Fairy (SL5). This may slightly increase the core rate on some maps.
- Runs may very rarely fail or be cancelled partway through. This may slightly decrease the core rate. 11-6 is most vulnerable to this, as I've tried a few Rescue setups that ultimately didn't pan out there.
- Some runs do not have the "mission attempt" recorded. These are excluded as I get the number of runs by counting the number of unique mission attempts.
- If no dolls drop at all, the run won't be counted. This should be an exceptionally rare occurrence, however. (Example: if the chance for each node & mission clear is equal and 12-4E drops 4.4 dolls/attempt (of 5 checks), the probability of failing all 5 checks in a row is 0.002%. If any specific drop is guaranteed this can't happen.)
With any luck these factors are minor enough to not have a significant effect and maybe even cancel each other out. But don't take these numbers as exact to the twentieth decimal place or anything.
I suspect some incorrect claims on core rates come from a problematic assumption in analyzing collected data - namely that there is a set number of drops per run. Many enemies don't always drop something, so using the number of drops to calculate core rate will be a bit off. Some examples based on my own data:
Map | Total cores | Total runs | Total drops | Drops/run | Cores/run | Cores/5 drops |
---|---|---|---|---|---|---|
11-5 | 554 | 1469 | 5035 | 3.43 | 0.38 | 0.55 |
12-4E | 8484 | 9465 | 41689 | 4.40 | 0.90 | 1.02 |
Dividing the cores in my data by (total drops/5) on 11-5 matches up with one core rate figure I've heard but overstates the true core rate by 40-50% as it seems the enemies there have a relatively low chance of dropping anything. I've also heard 1-1.2 commonly cited as the core rate on 12-4E, and assuming 5 drops/run gives a number closer to that.
You can verify this for yourself - try counting the number of dolls you have after a few runs of 12-4E. Sometimes you'll gain 4 dolls, other times you'll gain 5.