Token Metrics is a crypto investment platform that merges human capital, data, and machine learning to rate cryptocurrencies.
It’s designed to be the only platform you’ll ever need to make money in crypto.
We have channeled all our collective experience in blockchain development, machine learning, big data, and pattern recognition while working for firms like Goldman Sachs, JP Morgan, IBM, and Deloitte into Token Metrics.
Having invested in over 100 cryptocurrency projects, we have also learned from all our successes, most importantly, our mistakes and failed investments.
And all of our data and analysis is continuously improving every day, as the machine learning evolves.
Using Token Metrics, you can build and manage a custom portfolio that leverages machine learning without the need for any programming knowledge of your own.
But of course, you’ll have access to the data for yourself and contact with our team. You’ll be the one in control.
And if you’re among the first 1,000 people to subscribe to the Token Metric’s HODLer, Investor, or Professional plans today, you can get a full year’s subscription for 50% off.
We’ll let you know how to do that—and the different levels of membership we offer—in just a minute.
First, we want to show you why a data-driven approach is so important when it comes to crypto investing.
Quant trading has ruled the traditional investment markets for some time. But it’s only recently made waves in crypto trading.
Our original spreadsheet was a basic form of quant trading. It used mathematical and statistical models to analyze crypto investments.
However, the crypto market has shifted dramatically according to PwC! The only crypto funds that made money in 2018 were quant funds.
How are retail investors supposed to compete with quant funds? Especially if they’re not machine learning developers?
That’s one of the problems we’re planning to solve.
Token Metrics puts the full power of quant trading at your fingertips.
Developing the algorithms used in Token Metrics’ quant systems was not easy or cheap.
As we said, we’ve been working on this for more than two years, and put millions of dollars of our own money into the research, development, and experimentation of everything in our crypto investing career that’s led to the creation of Token Metrics.
But we think the cost and the effort were well worth it.
We're confident Token Metrics is going to help make many of its early subscribers enough money to retire on.
We realize that’s a big claim. But we have data on our side.
And more than that, we have the power of machine learning.
Remember the story about IBM’s Deep Blue supercomputer beating chess world champion Gary Kasparov?
Well, there’s a bit more to that story, which shows just how powerful machine learning has become.
You see, IBM started working on a supercomputer to beat the chess world champion back in 1985.
It took a full 12 years of development, and numerous defeats, before Deep Blue eventually beat Kasparov in 1997.
But the thing about Deep Blue was, it wasn’t really “intelligent” at all.
It needed to be fed unlimited data about the previous human vs. human matches to learn how to win.
Even Google’s world-beating AlphaGo program, which became the best Go player on the planet in 2015, worked in the same way.
It couldn’t learn for itself. It had to be spoon-fed by humans.
Then in 2017, everything changed.
In 2017, Google’s DeepMind released AlphaGo Zero.
This new software taught itself to be the best Go player that had ever existed in just three days through what’s known as reinforcement learning using neural networks.
Such systems “learn” to perform tasks by considering examples, generally without being programmed with task-specific rules.
After three days, it was able to beat the original AlphaGo program by 100 games to 0.
And it did it all without any human input, and without being fed any datasets.
It merely played against itself, and learned along the way, getting better and better with each passing second.
That is the real power of machine learning. And it is only now beginning to be realized.
The Token Metrics platform uses deep learning neural networks to continually improve our underlying cryptocurrency rating algorithms the same way Google did with AlphaGo.
The more valuable data we feed it the smarter it gets and we here at Token Metrics have worked out incredibly valuable data points throughout the years to feed it!
We’re currently leveraging genetic algorithms in our machine learning models.
Now stay with us here, it's going to get a bit technical!
A genetic algorithm is an algorithmic search process that is inspired by Charles Darwin’s theory of natural evolution.
This algorithm reflects the process of natural selection, where the fittest individuals are chosen for reproduction to produce the healthiest offspring in future generations.
The process of natural selection starts with the selection of the fittest individuals from a population.
They produce offspring which inherit the characteristics of the parents and become part of the next generation.
If the parents had the best fitness in the last generation, their offspring will have better fitness than the parents and have a better chance of surviving.
This process keeps on iterating, and in the end, we settle upon a generation with the fittest individuals.
In addition to applying genetic algorithms to crypto, we’re exploring using particle-swarm optimization algorithms to crypto as well.
Particle Swarm Optimization belongs to the field of Swarm Intelligence and Collective Intelligence.
Particle Swarm Optimization comes from the social foraging behavior of some animals, such as the flocking behavior of birds and the schooling behavior of fish.
Aagam Jain, Machine Learning Engineer
Machine learning is going to be paramount to the success of Token Metrics, and that’s why we’ve put together a strong data science team, that includes with Aagam Jain, a machine learning engineer.
Aagam is world renown for his performance in machine learning competitions.
He also makes sure our reinforcement algorithms are working smoothly.
The way that our algorithms work is that we start with our experience of investing in hundreds of cryptocurrencies in the last three years.
We score our performance as a benchmark.
We then create an initial random population of 100 algorithms. Think of this as having 100 brothers and sisters.
We test the accuracy of those 100 algorithms and declare winners.
The winners being the models that had the highest accuracy in predictions that led to better investment performance.
We then take the winning models, whether our expert opinion, or those from the initial random population, and create a second generation of another 100 algorithms, which we can call the offspring.
In other words, the best winning models create 100 new models inspired by the fittest models, just like with evolutionary natural selection.
Token Metrics creates over 500 generations a day.
Every single day.
That's 500 multiplied by 100 children equating to 50,000 total models created a day to find the most optimum models.
And that’s just for fundamental analysis.
We do the same thing for technical analysis, technology and code reviews, as well as our final grades.
That amounts to 200,000 total models a day across those four segments.
That’s a ton of data, and that’s just for one single cryptocurrency.
We replicate this for ALL other cryptocurrencies out there!
That means we run 20 million models a day combined for the 100 cryptocurrencies we have in our system at the moment, and those numbers are growing with each new cryptocurrency we add.
We don’t think anyone in crypto has done anything like this, at this scale, and made it available to the public!
Our platform never stops trying to optimize towards a better performing model.
The very next day, we continue the process, with the best performing model from yesterday.
Imagine how smart these models will get after every day.
We do not believe in a one-size-fits-all machine learning model.
That’s why we create unique machine learning models for every cryptocurrency because we examine every individualized data point, no stone left unturned.
This is similar to Charles Darwin’s natural selection.
Over several generations mother nature selects the strongest traits in a species.
This leads to drastic physical changes and gives species things like quills on a porcupine or thumbs on humans.
This could take hundreds of years, our algorithms allow us to recognize and achieve these drastic changes to our models overnight.
Humans cannot compete with this firepower and this is why AI is the future.
Humans have no way to learn from that many mistakes.
Humans make 5,000 trades over the course of 5 years and adjust and learn based on these.
If your phone can recognize your face and understand who you are, why can’t machines recognize the price movement in charts or the data points that matter in selecting a cryptocurrency?
AI is inevitably a better problem solver than us.
Our job is going to be to define the problem at Token Metrics then our machine learning models will solve it.
So with Token Metrics, you get the benefits of quant trading and the constant improvement of machine learning, all in one platform. Without having to be a developer!
That’s why Token Metrics is the only platform you’ll ever need to make profits in crypto.
It will always be working hard 24/7 for your assets, every second of every minute of every day.
But it’s not just about trading.
As we're sure you know, over the long term, asset selection also plays a significant role.
Buy and hold for the right time frame can produce truly momentous returns.
With Token Metrics, you get the fast profits of trading with the slow, oversized profits of buy and hold.
Actively day trading is not suitable for most people.
Passive investing or buy and hold is not the solution because crypto is a very volatile market.
Passive investing with active portfolio management is a much better option, in our opinion.
In other words, buy and hold with regular portfolio rebalancing done based on bullish and bearish signals of cryptocurrency.
This can be automated with a crypto investment platform that leverages human capital and experience with data and machine learning.
In the future, we want to build functionality into Token Metrics that lets you choose how often you rebalance your portfolio based on your investor profile and investment goals.
We currently create and back-test individual models for token ratings, using the following:
Our back-testing algorithm uses 74 data points and is continually improving, thanks to machine learning.
We'll tell you when a particular crypto has changed from bullish to bearish, and you can choose whether to trade in or out of it for yourself.
Or down the line, in future versions, you could have our algorithms do all the rebalancing for you on a weekly, monthly, quarterly, or annual basis.
And of course, you’ll have access to all the reports and analysis for yourself, too.
Token Metrics is far more than a “black box” trading system.
We know how important it is to be able to see all the data for yourself—that’s half the fun of crypto investing.
We built our reputation on transparency, and Token Metrics is no different.
All our machine learning model back-tests are available for users to view, keeping all of us honest and continuously in search of better returns.
And speaking of data…
You’ll get access to our IEO ratings, rankings, and reviews long before we release them to the public.
ICOs were what got many of us into crypto investing in the first place.
Just like with the evolution of the spreadsheet, the way we rank ICOs and IEOs has evolved, too.
We brought Sam Monac on as Token Metrics’ Head of Research.
He used to work at Liberty Mutual which is an American diversified global insurer.
It´s the third largest casualty and property insurer in the United States.
Even if you haven’t heard of Sam Monac before, you’ve likely seen his results.
We believe that our team’s research has been used by the world biggest cryptocurrency exchange Binance and several other exchanges to pick many of their own IEOs.
For reference we discovered and shared with our private investor network, Matic Network, Harmony, and Elrond, over a year before their public IEOs.
Sam Monac, Head of Research
Imagine if you had an opportunity to invest in all of those projects before they had their IEOs on Binance?
And as a Token Metrics subscriber, you can get exclusive access to our IEO ratings, rankings, and reviews—long before we release them to the public.
We’ll rank IEOs on a fundamental and a technical level, giving both scores from zero to a hundred.
Our fundamental analysis has many metrics, including:
The technology and Github source code reviews go even deeper into the workings of the project.
Paresh Masani, our CTO and former IT Executive at Goldman Sachs and JP Morgan, leads our technology research.
All our results made consumable as exclusive written reports and videos that only our subscribers can access.
It’s no secret that getting this kind of crucial information early-on can give you a huge advantage when it comes to investing in an IEO.
That’s why you’re investing in an IEO in the first place—getting in early before the price possibly shoots up.
Being a successful investor comes down to having excellent information and connections!
That’s why, when you subscribe to Token Metrics, you’ll also get exclusive access to our private investor Telegram group.
In here, you’ll be able to talk with our analysts and researchers directly, in real-time. Our CEO, Ian Balina, will also be a regular on the group, so you’ll have a direct line to him, as well.
We’ll also hold bi-weekly webinars, that’s two webinars a month, which we don’t intend to release to the public.
These webinars are where you’ll get the best information on our upcoming analysis, research, and token reviews.
You’ll also get a more in-depth look at our crypto market analysis and have a far better chance of connecting with the right people to invest early.
And then you’ll be able to discuss the webinars in our private Telegram group as well.
If you buy our product now, you will get access to Token Metrics at a huge lifetime discount because we’re still actively building the platform, and expect to add even more machine learning and robo-adviser functionality next year.
Harness the power of machine learning without being a developer to become a better crypto investor.
Copyright © 2019 - Token Metrics - All Rights Reserved
Token Metrics Media LLC is a regular publication of information, analysis, and commentary focused primarily on blockchain technology and business, cryptocurrency, blockchain-based tokens, market trends, and trading strategies.
Token Metrics Media LLC does not provide individually tailored investment advice and does not take a subscriber’s or anyone’s circumstances into consideration when discussing investments, nor is Token Metrics Media LLC registered as an investment adviser or broker-dealer in any jurisdiction.
Information contained herein is not an offer or solicitation to buy, hold, or sell any security. The Token Metrics team has advised and invested in many blockchain companies. A complete list of their advisory roles and current holdings can be viewed here: http://tokenmetrics.com/disclosures