[Research] Map increased item quantity (IIQ% ) and its role in map sustainability

INTRODUCTION
For the past few months I've been recording my T7+ map runs and collecting data on my map drops in order to determine the most efficient mapping strategy currency-wise. During my research the following criteria were taken into account:
- map increased item quantity (IIQ%),
- pack size %,
- boss kill.
After running nearly 200 maps I decided to run the numbers in order to determine how much of an influence do the aforementioned variables have on map sustainability. While boss kills had a marginal effect on map drops and while my map returns were significantly affected by pack size % which was to be expected... I've got rather unusual results when comparing low IIQ% and high IIQ% of which the map drop coefficient were nearly identical. In order to determine whether or not IIQ% achieved through map mods have any effect on map returns I decided to run a series of low IIQ% maps as well as collect various data from the community.

RESEARCH BOUNDARIES
In order to make this research as legitimate as possible I determined and implemented a number of rules that helped me establish the boundaries within which this research was conducted, namely:
- high-tier maps (T12+) were excluded from the research due to insufficient data consisting of low IIQ% map runs,
- mid-tier maps (T7-11) were divided into two subtier brackets, namely T7-9 and T10-11,
- each bracket was then divided into two categories, namely <60% IIQ and >100% IIQ for T7-9 and <75% IIQ and >100% for T10-11,
- each category has to consist of at least 100 samples,
- each sample has to include pack size.

SAMPLE ANALYSIS [EXAMPLE]
Hereby an example of how I collected and evaluated data for one specific bracket.

Out of 224 T7-9 samples I recorded for the purpose of this research 114 are a result of my own maps that I've ran over the course of this challenge league. Hereby an example of data I collected during this research.

In order to explain the data hereby the legenda I implemented so that a 3rd party reader can get a firm grasp of what the data actually shows.

Any map that map the criteria fit for this research were taken into account, for instance if Zana spawned a T8 map with low IIQ% and pack size the map automatically became part of the data.

ANALYSIS
LOW MID-TIER (T7-9)
Not all map drops were being included in the research data. For T7 map drops to be valid those drops had to consist of a map of tier equal or higher, which effectively excluded low tier maps from the research. For T8-9 the same rules applied with the addition of drops one tier lower. As a result if a T7 dropped in a T8 map it was classified as valid.

As you can see despite a significant difference in IIQ% (40%+) the sustain drop coefficient for low IIQ% maps was only 0.02 lower than that for high IIQ% maps. Both IIQ categories include around 10% samples being affected by double pack size, so that one cannot conclude that lower IIQ% maps had a significantly higher double pack size rate.

HIGH MID-TIER (T10-11)
For this category map drops two tiers below were also included as they can play a significant role in map pool sustainability [through trading up].

When it comes to T10-11 maps the map drop coefficient is 0.12 lower for maps with low IIQ% based on a 202 sample base. This can be explained by the fact that nearly 52% of high IIQ% maps included double pack size as opposed to only 20% of low IIQ% maps.

CONCLUSION
Based on the data consisting of a grand total of 424 samples it is possible to conclude that map increased item quantity (IIQ) achieved through a variety of map modifiers has marginal if any effect on potential map drops. This allows the playerbase to save a significant amount of currency otherwise spent on achieving as high IIQ%/map as possible next to pack size. I don't expect GGG to post any reaction as I understand the devs want the community to figure things on their own. I'd like to hereby thank the creators of Path of Maps as well as the redditor Shinyblu for allowing their data into this research.

Really nice that you shared this!

It seems like i "wasted" quite some currency on the quantity that I could have saved for buying high tier maps.

I am quite new to the game so this kind of information is more than welcome...


Thank you and keep it up,

Shamain

Nice effort and a good approach to shed some light on the whole map drop system.

Two issues I have with the analysis:

1) Despite the effort, I believe 224 map runs is still a rather small sample size. For future analysis, a couple of players should group up to increase the sample size.

2) For me, a rather large proportion of maps comes from vendoring 3 lower tier maps. As you did not include these and their IIQ-dependent drop rates in the analysis, I suspect the result being biased.

What do you think?

after a certain point its more viable to sell the 3 component maps separately and buy 2 of the higher tier than it is to actually combine them. That is if buying the higher tier was actually worth it.

unfortunately due to mechanics ggg themselves introduced. its basically get to t11 , sell everything above t11.

"

ben_hure написал:

Nice effort and a good approach to shed some light on the whole map drop system.

Two issues I have with the analysis:

1) Despite the effort, I believe 224 map runs is still a rather small sample size. For future analysis, a couple of players should group up to increase the sample size.

2) For me, a rather large proportion of maps comes from vendoring 3 lower tier maps. As you did not include these and their IIQ-dependent drop rates in the analysis, I suspect the result being biased.

What do you think?

1) It's a total of 424 samples (224 for T7-9 and 202 for T10-11), but I intend to continue the research in Ascendancy.
2) I think it's a personal preference. You also have to remember that you need 27 maps of the same tier and type to upgrade three tiers higher. That's a highly unlikely outcome in a research consisting of 424 samples. Personally I didn't notice lower tier maps to drop more often in high IIQ% maps, but if more players share your feeling I'll make sure -3 is also included in the follow-up research.

This data will surely help me in Ascendancy farming.

Thanks for your effort!

You should have been recording lower level map drops too, bucket them all in one category.

Why?

Because they are part of your map pool in the IIQ part of their formula. Two unfortunate consequences on your analysis:

1) you're lowering your sample size
2) GGGs IIQ formula might be favouring the lower end of the map pool (I agree that these maps are useless but it would be nice to confirm), thus you may be skewing your analysis.

I'll make sure I broaden the boundaries of my research in Ascendancy. You've to remember that I started recording maps for my personal purposes. I've ran so many high IIQ% maps I couldn't believe how much currency I've wasted.

Tkz for sharing. :)

I always felt being cheated doing 100%+ IIQ maps and next map i do with like 20 IIQ, i get 3 times more drops lol... It has been like this for ages. You might say its RNG, but over few years of PoE thats how it feels about the game.