We are Forward Thinkers.
Our thesis is simple. Let's dive in.
We believe computers are critical to finding sustainable edge in the ever evolving financial markets. Investors are inundated with data and opinions on a daily basis, making it hard to know when to take action. DropShot's core technology uses Machine Learning to search through mountains of data, finding signals that stand the test of time. We strive to take the guess work out of portfolio rebalancing and leave our clients with highly favorable outcomes. Our totally systematic approach is multi-faceted, relying on both diversified alpha and dynamic market exposure.
Family Offices • Foundations • Pensions • Endowments • Fund of Funds • Public Institutions • RIA's
Them
Firms that rely on human intuition for trading are completely unable to ingest and harness the vast amount of data that exists for financial markets. Even if they were able to extract some intuition, they are often unable to translate this to a successful trading strategy in a responsive manner.
DropShot
This is where DropShot’s algorithmic data engines help us a lot. We ingest raw data from a range of sources. The data includes global financial exchange data, news, economic data and other sources. This data is systematically cleaned, processed and deployed for our algorithm so that the algorithm can use the most up to date and comprehensive information possible.
Them
Due to a lack of quantitative abilities within many financial management firms, the idea of testing is often overlooked. Particularly in a human discretionary case, the success of a trading strategy cannot often be replicated in a dynamic market that is always popping regimes.
DropShot
Our testing is what sets our process apart. Use of advanced machine learning tools comes with the price that the process can turn into a black box. Our algorithm building framework lights the way, thoroughly vetting strategies and throwing out unstable signals. Diligently applying the scientific method to ensure the best chance of future trading success is the name of our game.
Them
Many algo hedge funds operate in the high frequency space, taking rapid short and long positions in rapid succession. These kinds of strategies tend to focus on very expensive, low latency equipment to execute orders, and rely on very expensive data at the tick-by-tick level. Conversely, RIAs and many other money managers take a passive approach, placing money for very long periods of time into relatively static allocations. This lumbering and slow approach to financial markets leave them often in the dust of more nimble firms.
DropShot
DropShot is between these two extremes. We trade medium frequency, rebalancing daily. We approach investing the way a retail investor does, taking long only positions in highly liquid, publicly traded ETFs. Doing so allows the Fund exposure to a wide variety of non correlated assets.
See the Model In Action
Our portfolio performance speaks for itself, take a look.
The Theory
And Supporting Evidence
Hypothesis
Active, automated strategies based on machine intelligence can achieve market outperformance over time, especially relative to passive or human discretionary trading methods.
Context
Of public equity assets in the US, worth $31 Trillion, only 2.4% of the trading is accounted for by quantitatively focused funds. (Economist, Oct 2019) This stands in contrast to many reports that algorithmic trading dominates the markets. In truth, many orders are executed electronically, and thus get classified as ‘algo trading’ but this is not the same as using quantitative methods to trade systematically.
Evidence
Our past success (#7 in Barclay’s Balanced Index, past 3 years) has made us believe even more in our systematic approach. We are really trying not to have any trace of human complacency in our process, letting science drive the decision making. Past performance does not guarantee future success after all, but hard work doesn’t hurt.
Conclusion
Ultimately, our past trading performance is the litmus test for this idea. We have always operated systematically, using machine learning algorithms. Therefore, our nearly 5-yr track record is the perfect way to test the hypothesis.