Back to Community
Allocation eligibility for successful contest algorithms

I was wondering if any more clarity/information could be provided regarding the correlation between doing well in the contest and being eligible for an allocation. If I'm not mistaken, currently, the contest is the only way for Quantopian to review algorithms eligible for allocation so intuitively it makes sense that the next batch of allocations is going to come from algorithms that perform very well in the contest. However, beyond this there is not much information that can be ascertained. I was therefore wondering whether any more information could be shared on the current allocation process e.g. How much out of sample data must an algorithm have before it is eligible to be invested in? Roughly how many algorithms from the contest will be invested in? Is position on the leaderboard showing current cumulative score more important when it comes to allocation eligibility than position in the all time winnings leaderboard?

13 responses

We apply a data-driven, user-blind approach to our selection process. We don’t look at your user background or your specific rank in the contest, but rather the performance of your algorithm within the set of construction constraints. The contest rules help guide you to create strategies that would be possible candidates for an allocation. Doing well in the contest does not guarantee an allocation, but it will improve your chances for selection. We aren’t capping on a specific number of algorithms that we’re looking for — if your algorithm passes the criteria and looks interesting, we’d love to speak with you!

We are looking for strategies that are beta-neutral, sector-neutral, and have their returns driven by the specific returns, instead of common returns. Generally, a larger universe is better to diversity away risk. As per the contest, you should use the QuantopianTradableUniverse (QTU) and optimize API for your universe selection and ordering mechanics.

We generally wait for strategies to accumulate 6 months of out-of-sample data before evaluating. This helps to corroborate the strategy’s in-sample performance.

It’s great to see your interest and I would suggest that you to focus on creating strategies that generally perform well in the contest and are consistent over time. Best of luck!


The material on this website is provided for informational purposes only and does not constitute an offer to sell, a solicitation to buy, or a recommendation or endorsement for any security or strategy, nor does it constitute an offer to provide investment advisory services by Quantopian. In addition, the material offers no opinion with respect to the suitability of any security or specific investment. No information contained herein should be regarded as a suggestion to engage in or refrain from any investment-related course of action as none of Quantopian nor any of its affiliates is undertaking to provide investment advice, act as an adviser to any plan or entity subject to the Employee Retirement Income Security Act of 1974, as amended, individual retirement account or individual retirement annuity, or give advice in a fiduciary capacity with respect to the materials presented herein. If you are an individual retirement or other investor, contact your financial advisor or other fiduciary unrelated to Quantopian about whether any given investment idea, strategy, product or service described herein may be appropriate for your circumstances. All investments involve risk, including loss of principal. Quantopian makes no guarantees as to the accuracy or completeness of the views expressed in the website. The views are subject to change, and may have become unreliable for various reasons, including changes in market conditions or economic circumstances.

Hi Alisa
Is it accurate to assume that strategies that do not include short positions are not eligible for an allocation?
Or by beta neutral you mean strategies that are not correlated to the S&P500 (perhaps a correlation less than 40%)?
Do let me know at your convenience.
Kind regards

Thank you Grant that is very helpful. Unless I am not understanding it there is no specific requirement to have a short position. The beta is pretty clear. Another thing not covered is if some data is imported through Fetcher. Have not fully ported over the code to the Optimize API as well. Let me see if I pass some of these hurdles eventually. Thanks very much for help. Kind regards and a pleasant evening Savio

I think you’ll struggle to create a long only market neutral strategy within the QTradableStocksUS universe.

Thank you Joakim I accept the gauntlet - will let you know how I make out.

Good luck! Going long negative market beta stocks as a ‘hedge’ perhaps? Note that ETFs that are shorting the market (SDS for example that shorts the S&P 500) are not included in the QTSUS universe.

Also, for the contest, you do need to be short as well:

Long/short: Contest entries must not have more than 10% net dollar
exposure. This means that the long and short arms of a Contest entry
can not be more than 10% different (measured as 10% of the total
capital base). For example, if the entry has 100% of its capital
invested, neither the sum value of the long investments nor the sum
value of the short investments may exceed 55% of the total capital.
This is measured at the end of each trading day. Entries may exceed
this limit and have up to a 20% net dollar exposure on up to 2% of
trading days in the backtest used to check criteria


Hello Joakim the way I am understanding the rules is that the 10% difference is measured on end of day holdings. Since my strategy is mainly intraday I might be able to get away with long only, you never know. In any event, it will give me satisfaction just to see the strategy being consistent - participating in the contest will be for fun and to claim bragging rights over a beer. Will be in touch - have a nice evening Savio PS I looked at short ETFs sometime back - 1x, 2x and 3x - their tracking error at the time did not make them suitable for my strategy - I understand they were synthetic shorts - I have not revisited them recently.

Fair enough Savio, and good luck! You may find that you're hitting the Turnover constraint then if you're doing mostly intraday trading, especially since you do need to close roughly half of your longs and open up short positions instead before the close. I think you'll find it difficult to meet all constraints but wish you best of luck. I'd be interested in hearing how you go.

Savio, I'll echo what's already been mentioned in this thread. Quantopian is seeking to make allocations to algorithms that have a particular risk profile, as outlined here. The contest is structured to make it easier to know if you're on the right path. Doing well in the contest is a step in the right direction.

Thank you Alisa - working on it feverishly now that Joakim has thrown down the gauntlet.

Savio -

I'd also recommend structuring your algo so that you can play around with multiple factors, and combine them. This can help smooth out returns. No time now, but if you are interested in an example, I can post one.

Thank you Grant - that is very kind of you. My current strategy is based on straight portfolio optimization although I have a pre-optimization filter. I have spent some time on flattening out the efficient frontier. I do not use Black Litterman, just plain old vanilla historical data without any of my own subjective bias. I put this together based on material I was studying for the CFA program. In theory it should work. Will let you know how I fare. Thanks very much again for your help and wish you a pleasant weekend - it is going to be a scorcher here.