Building Fair NFT Based Games

A statistical breakdown of Season 1 of Non-Fungible Heroes — Interactive Quests

A sense of fairness has always been crucial in game design, but it is doubly important in games designed around NFTs, as the amount of investment from players is an order of magnitude higher than your average indie or even AAA title.

With Season 3 of Interactive Quests (IQ) kicking off this past week, we thought it would be a good idea to look back on some of the stats from our very first season and see if we hit our original design goals.

What Is IQ?

A quick recap, if you’re not familiar with Interactive Quests. Interactive Quests is the Non-Fungible Heroes idle game where players can soft-stake their NFH NFTs on quests. If their characters are successful in completing their quests, they will earn EXP, level-up, and ultimately earn $LORE, our ERC-20 token that can be traded or used to buy items, NFTs, and unique services in our Loremporium Marketplace.

Quest success is based off character levels as well as the ability of the user to match traits from their own NFTs with the traits of the quest patron.

Goals

Going into Season 1, our primary goal was to design a game that felt fair to both holders of a small amount of NFH ecosystem NFTs as well as our whales that held many of our NFTs. The primary question we had to answer was, “How do we make it so additional characters (NFTs) provides gameplay benefits, but with diminishing returns?”

The most significant lever we had to pull in this regard was character count. How many quests can a user run at once, and how many characters can be sent on each quest.

Want the short version? The secret number was 12. 4 Quests with 3 characters per quest.

This wasn’t completely a gut check, but instead was based on a snapshot we did of our entire collection during the game design phase. 12 represented the largest portion of our user base. After 12 the total wallet count started to drop off drastically.

But what happened after 12? While the player can’t have more than 12 characters out on quests at a time, having more than 12 characters does provide the player with more opportunities to match high value traits with the quest patron.

IQ Player Composition

That’s enough preamble. Let’s get into the meat of this. We were just discussing user snapshots, so let’s discuss NFH users and NFT distribution per wallet for IQ players.

All data was compiled, sorted, and analyzed by @Codename. Data used in this analysis comes from a June 16th 2022 snapshot. The majority of the insights provided by this article are pulled directly from Codename’s original analysis.

In order to better understand populations being serviced by NFH IQ, the active wallet population was broken into six subgroups based on the number of NFTS held.

If we then look at the number of wallets for each category, along with the number of NFTs represented by each wallet category we get:

This nominal grouping of the NFH population provides us with a not-unexpected insight: we have a large population of small holders, and a small population of large holders.

• Top 5% of Wallets hold 41% of NFTs
• Top 16% of Wallets hold 63% of NFTs

Community Composition

But how active is that population? This first table shows the number of active wallets vs the total wallets holding NFH NFTs.

A relatively small amount, but let’s get to the next table before we discuss findings. This table shows active NFTs vs the entire collection without regard to wallets.

Suddenly the active percentage jumps from 8% to 27%. So what can we take away from this?

1. That the community is a relatively small but tight-knit group of individuals.
2. That a small group of supporters hold a larger percentage of the collection.
3. That the $LORE distribution method is being accepted by “Ambassador” level community members.

$LORE Distribution

The next graph shows $LORE earned vs the number of NFH NFTs contained in the earning wallet.

At the top-tier view, it is seen that high $LORE capture has no correlation to wallet size. There are roughly equal populations between REGION 1 and 2. If we look specifically at Region 2, where “Ambassador” class wallets are mostly contained, we see that IQ provides no advantage to Ambassadors to over-compete. Ambassadors are unable to capture a sum of $LORE proportional to holdings.

This graph was pretty significant to us internally as it showed that we had hit our number 1 design goal which was to create a game that rewarded more substantially for player engagement rather than simply rewarding based on NFT quantity.

Let’s zoom in on the lower-left of this graph and see how $LORE is captured at the mid level.

Similar to the top-tier view, we see that medium $LORE capture has no correlation to wallet size. There is roughly equal populations between REGION 1 and 2. IQ provides no competitive advantage for Titanium and Platinum tier against Gold tier. Gold tier holders were able to out compete against Titanium and Platinum.

Finally, let’s exam the lowest level view of the results.

At the low-level view, it is seen that small $LORE capture does have a correlation to wallet size. We see a linear relationship of earnings for Bronze and Silver tier holders.

IQ provides a disadvantage for Bronze, Silver, and lower portion of Gold tier holders causing them to “under-compete”. Top performers facing these conditions were able to capture ~2500 $LORE per NFT.

However, we’ve saved the juiciest table for last.

$LORE Distribution Totals and Averages

Key Takeaways:

Ambassador, Titanium, and Platinum Tiers were roughly equal in $LORE distribution. Holders from these groups can compete equally to optimize $LORE earnings. As the number of NFTs in these tiers increases, the return on investment diminishes.

Gold, Bronze, and Silver tiers were limited in $LORE distribution. Holders from these groups can compete but were pushed to expand their roster in order to optimize the $LORE earnings from their collection.

While not included in the above chart, single NFT owners were able to earn on average 500 $LORE over the season. With the highest performer earning 2500 $LORE. More than enough to still be able to make purchases on the $LORE marketplace.

Final Thoughts

We definitely hit our initial goals with an even distribution of $LORE between both small and large wallets. However, because $LORE distribution past 12 character per wallets was so aggressively even, it’s possible the design could come across, at best, agnostic towards whales, and at worst, antagonistic towards them.

In seasons since, we have drastically increased the bonuses that come from trait matching in order to provide more benefit to holding more than 12 characters but without completely destabilizing the design that has let small wallets be competitive with their whale counter parts. In addition, we’re tracking significantly more battle metrics since the first season, so look forward to more insights soon!

Want to try out Non-Fungible Hero’s Interactive Quests for yourself? Pick-up a Hero or Villain on secondary, or mint a God while the mint is still open!