How ICOformula rates, ranks and scores ICOs

Welcome to ICOformula! If you're on this page chances are you're either an ICO enthusiast looking into how we rank ICOs, why we have different scoring systems, and why we rank projects differently to other ICO rating platforms, or, you are involved with an ICO project in a professional capacity and looking to see how/why we have ranked your project with the score it has.

Introducing the "Formulatron"

We have created a dynamic mechanism to calculate the rankings and named it the "Formulatron" – in a bid to explain what the Formulatron is and provide some insight into how it works, what it measures and why we built it – below is a guide to what goes on behind the scenes at ICOformula and a little history of its creation.

It all started in early 2017 as the ICO boom started to gain serious momentum. Initially it wasn't that difficult to spot a reasonably promising ICO, there weren't many to choose from at the time, and the sector was already showing similar attributes to the dot com IPO boom in the traditional investment markets of years gone by. Throwing darts at a list of projects to pick them out as "winners" was plausible.

As hype started to build around the money to be made, various ranking and rating concepts and models began to appear online in spreadsheet from – most of which were derivatives of an initial concept promoted by Ian Balina.

Dealing with the spreadsheet swamp

Whilst spreadsheets are convenient to analyse reasonable amounts of data, keeping them updated and maintaining them is not easy, and automating the process is a headache at the best of times. Adding more data and additional metrics (such as community engagement and social media footprint etc.) becomes a serious challenge in such a rapidly growing space.

Hence the creation of the Formulatron – a more robust way of rating ICO's at scale, both relatively in conjunction with the recent market performance of completed ICO's and independently for each project – and further still to order the rankings such to display which projects show potential from an investment perspective going forward, whilst also allowing for additional scoring metrics to be added as they become available, and account for social and sentiment data as close to real time as possible.

We also decided the Formulatron should be built in such a way that it outputs multiple scores for any given ICO to account for cryptocurrency market conditions at the time, as well as public sentiment.

No small task by any stretch of the imagination!

So, without further ado – some explanations

The core scoring starts in a similar way that most of the early spreadsheets did:

  • Idea – How important is the Idea? Does the idea/proposition make sense? Is it original? Does it solve a problem? Does the market exist? Is it hard to replicate?
  • Team – who are they and what is their experience?
  • Advisors – again, who are they? What is their experience? Or are they included for branding purposes?
  • Prototype/product – Is it operational? Is it at MVP stage? If so, is it clearly described and shows promise? or is it a vague demo version of an idea? Or, is it just a "PoP"? (proof of PDF – i.e. literally nothing but a whitepaper and nothing else exists yet)

In addition to the four core metrics we also take into account the competition in the space: Market competition – Essentially whether a competitor could easily recreate the same

idea/project. Complexity takes into account whether the project has a patent, large capital needed for production and/or physical expansion.

The core metric scores are compiled by a team of experienced independent analysts, the results of which are aggregated to create each individual metric score.

Tokenomics

We then take into account the economic state and structure of the project. Fund raising targets vary wildly from project to project, however it's not just about the total Dollar volume being raised – the Formulatron takes into account the following data (when known/available):

  • Dollar volume targeted for the raise (ICO hardcap)
  • Founder/Team/Advisor share amounts
  • Founder/Team/Advisor vesting periods (how long are their tokens locked in for after the ICO)
  • Token economics modal – are unsold tokens burnt, vested or shared?
  • Percentage of total tokens being issued for ICO investors
  • Price of the token during the ICO period

All of the economic metrics are scored to take into account how favourable the situation is for potential investors – for example: If the volume of tokens being retained by founders is higher than tokens available for ICO investors – a lower score is applied. Similarly, if there is no vesting schedule for founders/team/advisor allocations – a lower score is applied because the potential is there for insider dumping of tokens once the project hits exchanges and shows low confidence by the creators of the project.

On certain individual metrics for any given project, negative scores can result which will reduce the overall score of a project.

Social media scoring

The next set of data we stuff into the Formulatron is from the world of social media and messaging platforms.

First the follower counts from the following platforms – (when known/available) and calculate scores based on volumes across the dataset:

  • Twitter
  • Facebook
  • Reddit
  • Telegram

We then pull in additional social data and calculate engagement scores (when known/available) for:

  • Twitter
  • Facebook
  • Reddit
Benchmarking

Once all the data is in the Formulatron – we then have to calculate and apply the benchmarks to which success can be measured. This is done by taking the top performing completed ICO's that are now trading on exchanges as tokens. The initial selection criterion is highest ROI. Metrics are then taken from the top performing Tokens (within the ICOformula dataset) and the Formulatron then calculates the "sweet-spot" for the following data points:

  • Token price at ICO
  • Normalised market cap at ICO
  • Available tokens
  • Global market cap at ICO

The sweet-spot data is then used to calculate how close any new ICO is to the top performers across multiple metrics for every ICO we list.

High minimum requirements for score data

Every ICO on ICOformula by default has the core scoring metrics at the very minimum, and sufficient economic metrics to have the sweet-spot scoring applied – without those scores, the Formulatron cannot do its job and the project would be classed as a "pre-ico" and not included in the calculations – to date all ICO's have more than just the core metric and benchmarked scores.

Updating and including additional data

Because we constantly update the data, if new data becomes available (change in token price or increase/decrease in the tokens available, or addition of a social media account not available at the time of the initial listing) the data in the Formulatron is updated, recalculated and new scores updated on ICOformula.com.

Accounting for missing data

Missing or unavailable data for the ancillary metrics would obviously cause an imbalance in the final scores – we accounted for this by reweighting and normalizing at several points throughout to ensure a balanced and fair scoring mechanism based on the facts available.

Maintaining fair and balanced scores

In addition to including more metrics with which to calculate rankings – we were also very mindful of the increasing number of projects and promoters of them attempting to "game" or manipulate ICO rankings on popular ranking sites that fully disclose the methods – having additional metrics reduces that likelihood of course – but to provide a genuinely clean and fair ranking and rating system we have established a few policies to which we adhere:

We purposely don't write reviews for every ICO – we did a handful in the early days, but after looking at the overall marketing methods used by many ICO promotors it became clear very quickly that reviews can be biased – especially when significant sums of cash are offered from marketing agencies and even the ICO projects themselves when asking for "favourable reviews". Instead, we tend to use the project abstract from whitepapers, or the closest equivalent.

We are happy to advertise ICO's to our user base via our email notifications – but doing so will not directly improve any scores displayed on ICOformula – it can and does aid with publicity which in turn will increase the social metrics for quality projects. Offering us a bunch of cash to "up the score" is point-blank refused – primarily because it detracts from the point of the system in the first place, the complexity of the Formulatron makes it close to impossible, and,... it's just wrong.

Why ICOformula is significantly different from other rating sites

Unless this is the first ICO rating site you have ever seen, you'll be familiar with the typical ratings/rankings/scores published by the vast array of rating sites online. Personally I have yet to find any that have gone so far as to score ICO's beyond just a single score, with many sites beginning to have significant numbers of projects all being scored within a very tight range – either due to internal manipulation by the rating site as a result of advertising cash or the offer of tokens or external manipulation by the projects (i.e. constructing their marketing materials specifically to gain high ratings prior to submission).

Obviously this sort of activity by ranking sites is no help to the individual trying to find ICO's that are all judged using a constant process, few take into account the bigger picture when it comes down to the state of the market and perhaps most importantly, I have yet to see another site that accounts for the volume of ICO's being bought to market. As an example of the last point:

Imagine you are intending to invest in a handful of ICO's – you only have a finite amount of funds with which you are prepared to risk. If most of the ICO's on a rating site all sit within a scoring range of 4 and 5 out of a total of 5 – how do you decide? They cannot possibly all be as good as each other!

Independent, relative AND rank scores for each ICO

To avoid a similar issue occurring on ICOformula the Formulatron has been built to effectively generate 3 scores.

  • An "Independent Score" - Rates the ICO with a percentage score based upon the data of the project
  • A "Relative Score" – Rates the ICO with a percentage score, relative to all the active ICO's currently listed on ICOformula and performance of Tokens listed on ICOformula
  • A "Ranking Score" – The position of the ICO as it relates to the listed ICO's using the "Relative Score" – Position 1 being the highest ranking project

As I'm sure you can imagine – this has not been a five minute job thrown together to ride the cryptocurrency "craze" – far from it. We've built this as tool initially for ourselves, and whilst initially considered making it a paid service – we have chosen to make it free to use. Why? Amongst other things, the world of crypto is still very much in its infancy and we're humble enough to know that there will always be room for improvement given constructive feedback from active users in the long term.

Future plans, improvements and additions

We have multiple versions of the Formulatron running with different parameters and configurations tied in to several different machine learning algorithms so we can continually amend and improve our calculations – in addition we also run extensive analysis on historical data to hunt for correlations and additional metrics that could potentially improve results further still.

In addition, we're already working hard with the knowledge gained thus far to develop prediction tools for live tokens on the exchanges – whilst prediction engines are not that difficult to create, building and maintaining one that can cope in a sector that changes so rapidly is another matter entirely!

Some random stats
  • Over 200 separate excel books were created in the process of building the Formulatron
  • More than 1,400 individual spreadsheets
  • Close to 1,000 cups of coffee
  • An unruly amount of swearing (usually due to Excel)
  • At one point we had to archive over 3 million data points to prevent system crashes
  • Almost 500 pages of research
The code behind the Formulatron

Before you ask, No – it's not open source and no, we won't send you a copy of it (no matter how nicely you ask!). The fact that we want to remain as independent as possible and create completely unbiased ratings means distributing the code and algorithms would instantly open up the possibility of manipulation – exactly the same reason Google or any other search engines don't open source their code.

Still awake? What next...

So if you want to see the result of our sweat, tears and many sleepless nights, simply click here to create your ICO formula account and get started.

DISCLAIMER - ICOformula.com does not provide investment advice. The use of this website and any data or information about Cryptocurrencies, ICOs and Token sales, including but not limited to Score, analysis, ratings and rankings and any links or non-proprietary content does not constitute any advice or incitement to investment. You use this site and its data entirely at your own risk. All information is for informational purposes only. Please read our full Terms and conditions for more information and ALWAYS do your own due diligence prior to making investment decisions.

SHARE THIS ARTICLE...
Need to showcase your ICO to the World?

Our ICO advertising services are launching soon. Take advantage of our early bird discounts.

Serious about ICOs and Crypto? Discover, learn, rank, score, analyse, track, guide and plan your Crypto investments...all in one place.
Connect with and follow ICOformula...