Finding consensus on the most popular blockchain games
Filtering data providers for better data
Transparency is a truism of the blockchain world but that doesn’t mean good data is easy to find.
This is quickly becoming clear when it comes to tracking the performance of dapps.
Sure, the number of companies providing such data is ever increasing. And, as we will see, comparative data for most individual dapps seems reliable. The significant issues, however, are viewed from the macro level.
Each data provider takes a different approach, both in terms of what blockchains they cover, and how they categorize different types of apps; is CryptoKitties a game or a marketplace for example?
More seriously, how do data providers deal with data manipulation; already a talking point in the industry. This is particularly prevalent when it comes to categories such as gambling and high risk dapps on the EOS and TRON blockchains.
Structurally too, transparency is starting to break down, notably in terms of how to measure the activity happening on Ethereum-based sidechains such as Loom Network. This is especially problematic when attempting to compare transaction activity for dapps across Proof of Work blockchains such as Ethereum and Proof of Stake blockchains like EOS/TRON. But that’s another discussion for another time.
In the meantime, here’s a simple comparative process that looks at how four leading dapp data companies rank the top performing blockchain games in terms of one metric — users over the past 24 hours — and attempts to come to a consensus about the top 5 game dapps overall.
It starts with raw data
The starting point for our comparison is simply to look at the top 5 games by the number of users in the past 24 hours as providing by DappRadar, Dapp.com (Dapp.com) , Dapp Review and State of the Dapps.
As can be seen, at this raw data level, there’s some agreement (everyone lists EOS Knights) but also a fairly large amount of disagreement.
One of the main reasons for this is the number of blockchains each provider supports.
- DappRadar is the most restrictive, only covering dapps on Ethereum, EOS and TRON (3 chains)
- Dapp.com also covers dapps running on Steem and TomoChain (5)
- State of the Dapps doesn’t support TRON or TomoChain, but does add GoChain, POA and xDai (6).
- DappReview covers dapps on the main three chains, Steem, IOST, NEO, Ontology, Nebulas and BOS (9)
Given the majority of these blockchains host very little activity, this doesn’t matter too much except for the case of Steem, which despite being something of a legacy blockchain at this juncture, has two games listed in the top 5 table; Steem Monsters and DrugWars.
The details matter more
Ignoring this for the time being, it’s important to recognise there is strong consensus between the providers when it comes to the one dapp all of them list in their top 5.
DappRadar, Dapp.com and DappReview record EOS Knights as having 4,800 users while State of the Dapps is slightly higher at 4,877.
We can see this level of consensus over total users pers dapps also holds when we remove all dapps that aren’t duplicated across data providers.
The user numbers for EOS Knights, My Crypto Heroes and 0xUnivese are in close agreement, while DappRadar, Dapp.com and State of the Dapps are also similar in terms of order and positioning.
One of the reasons for DappReview’s position as an outlier is down to how it categorizes dapps. For example, both DappRadar and Dapp.com list Crazy Dogs Live (DappReviews’ #1 game) within their Gambling categories.
However, the companies also measure very different user levels: DappRadar listing 32 users and Dapp.com 0 users compared to DappReview’s 6,579.
Fishing Master is a similar case. Both DappRadar and Dapp.com list it under Gambling, and although DappRadar records 2,300 users, Dapp.com lists 0, with State of the Dapps listing 3.
Clearly, many of the gambling dapps on EOS and TRON should be considered edge cases due to high levels of data manipulation and bot activity they attract.
Looking for best case agreement
But let’s return to the original goal of this process: to find consensus between data providers in terms of the top 5 most popular game dapps.
Some assumptions have to be made and in this case, I decided to only allows dapps listed by all four providers.
As DappRadar doesn’t cover Steem and State of the Dapps doesn’t cover TRON, this meant only dapps running on Ethereum and EOS were considered. I also went with the consensus view (3 vs 1) that CryptoKitties was a game, not a marketplace.
The result is a consensus top 5 table that — filtered in these ways — shows remarkable agreement between all four data providers.
In fact the only dapp with more than 1% variation is Blockchain Cuties with State of Dapps recording 596 users (+8% compared to the average).
In conclusion, on an individual dapp level, there is strong consensus between the four data providers about the number of users within the Games category.
However, users of these services need to be aware of each data provider’s specific approach to the dapp ecosystem in terms of the blockchains they support and how they categorize dapps.
Further work need to be carried out to see whether this level of consensus holds for other key dapp performance indicators such as the number of transactions and their monetary value.
It is also important to point out that dapp categories such as Gambling and High Risk are not covered by these conclusions.
And, more generally, it’s always the case that great care should be taken when directly comparing the performance of dapps from the same category across different blockchain types; i.e. comparing games running on Ethereum to those running on EOS and TRON.
