This post is about the method I used for determining the most efficient maps for Rainbow Goblin hunting based on the Goblin spawn rate for that map and the time it takes me to run those maps. This is also a followup post to this thread I created a few weeks ago: Some Thoughts and Data After Killing-100 Rainbows
Here is what the finished product looks like, and here is the link to all the data: Goblin Tracker (Responses) - Google Sheets
Everything below this line is dry and boring, so if you are not interested in the details of how I arrived at my conclusions, then there is no need to read further.
There are many things you can do to make your route more efficient such as learning the map layouts, using a speed build, and lowering the difficulty, among many other things that I will not go into here.
In order to determine which were the best maps, I needed to know what the exact spawn rate was for each map, and also how long it takes to fully search that map, including travel time to get there. I did this by creating a Google Form where I could easily enter data on my other monitor as I was hunting, and then submit it to a spreadsheet where I could analyze it on-the-fly.
Here is a link to a copy of the Form I was using to collect my data: Goblin Tracker Demo
Here is a link showing all of the collected data in its raw form as well as my conclusions: Goblin Tracker Demo (Responses) - Google Sheets
This is what the form looks like:
I then started to collect data for each map. This took a full month, and I killed a little over 1,000 Goblins over the course of around 120 runs and 2,600 maps visited.
It is important to note how I collected my data. I made a conscious decision before I entered a map about whether or not I was was going to full clear it, and if so, I recorded the results, whatever they were. If I was only traveling through a map, and not fully searching it, then I did not record the results even when I found a Rainbow Goblin there. This was to preserve data integrity.
After I had collected data on at least 10 runs for each map, I had a decent idea of the spawn rate for the maps, and I also had recorded the average time it took for me to run each map.
All that was left to do was to sort the maps by the average time it would take to encounter a goblin on that map. Since most maps have less than a 100% spawn chance, the time it takes to encounter a goblin is always higher than the time it takes to run the map. For example: If the spawn rate for a map is 35%, and it takes 20 secs to run that map, then the average time to find a goblin on that map is 57 secs.
I also combined multiple floor maps such as Moon Cave 1+2. Since no one ever visits only the first floor, I treat these maps as a single destination. This causes some maps to appear to have a higher than 100% spawn rate since I am counting both floors.
For a few maps such as the Desolate Sands and attached caves, I calculated them a bit differently. The real destination is the caves, not the huge map of the Desolate Sands. Also, some of the caves do not always spawn, so that makes it a bit trickier. For that map, I took 3/4 of the time it takes to full search the main map, plus 100% of the time for the maps that always spawn, then 1/3 of the time for the Vile Cavern since it has a 33% spawn rate. Then for the percentages, I used the same calculations, more or less. Also, the Veiled Treasure has a 100% spawn rate for a goblin, but it is not very common, and you have to complete the Ancient Devices event in order to have a chance to spawn it. However, I have also kept the raw data as well on the other sheets in the spreadsheet if you want to see a breakdown of the maps by floor.
I would suggest that if you are contemplating putting together your own Optimal route, that you record your own run times, and use your own data to determine how long it takes you to run the maps. Also, you might be less efficient in maps that have lots of dead-ends than you do in open maps.
Then of course, there is personal preference to consider when putting together a route. I would suggest only searching maps that you enjoy running.
In any case, I wanted to share the method that I used to compile a Goblin farming route that works best for me. At the very least, you can look at the spawn rates for each map and make your own decisions.
One last thing: More data means more accurate results, and for some maps, I had a hard time getting some maps to spawn, so I do not have nearly as much data on them as I would like. Look at the number of times I have visited a map, and take that data with a grain of salt.
Please let me know if you found this useful, and if you think there is a way that I can improve my results or if you would have gone about this in another way.