Hello. New user is here. Quick background: two friends (one is a CEO of a large telco Company and the other is a Ph.D. in Physics) are about to start a small trading shop. Have algorithms (developed/ tested/ deployed) that are currently running on TS (plan to move to NInja) utilizing 1 sec data for futures trading (YM currently). Algos are pretty stable and several ppl are willing to invest with us (mostly HNW). Question: what kind of mistakes we can make if we start that trading as an LLC, owned by us (the majority stakeholders) and several other ppl (minority stakeholders). We plan to rise / deploy anywhere between 200k and 1M during the first year. All suggestions are welcome as we are new to this "world" and eager to continue learning the proverbial "lay of the land". Thank you. Cheers!
Trending Content
Career Resources
Your first margin call will be exciting.
Thought about that, this is why the average holding time for a trade is ~1 min, with the majority of trades closed under 1 minute.
It’s hft pitch or close to it. Margin calls don’t exists. The calls they will get on that capital base is not being able to afford the tech they need. Or figuring out the algo loses money and slowly bleeding to death in small losses/fees.
It’s a tough business now.15 years ago they would have done well.
Not quite correct. The algos are making money (already responded to your latter comment below) and we can afford pretty much the tech we need. Especially given that a significant part of our calculations is based on certain analytical frameworks that significantly reduces the hardware requirements. Without going into details, I can offer a very simple analogy. You can either use computer power to numerically solve a system of differential equations, or you may construct an analytical framework to obtain an analytical solution. In the latter case you will be able to apply the results to the real life situation more efficiently and using significantly less computing power.
just FYI....this is not a "futures trading shop"....this is a hedge fund...you are starting a hedge fund, that trades futures.
Right, this is where we want to go, but with couple hundred k's it would be very presumptuous to call this a HF )) We have just a very short track record, this is why we rely only on our friends, while raising the money.
Disagree this is basically the definition of Chicago prop. It’s not a hedge fund.
Though the lines blur a ton.
I am confused - are you trading live already?
Yessir
What’s your realized SR and PNL per trade value? Cause if I had to guess with 1 min holding time, latency and transaction costs will be your main proble. PS. Please don’t use Globex tickers when describing your strategy, it’s annoying , just say “I trade 10 year notes” or whatever
double-posted, deleted the second comment. Apologies
Not my area, and I’m by no means an expert in any area, but the main worry id have if i were in your shoes is that the alpha will go away (keep hearing about how fast alpha disappears in hft and the fact that you’re explicitly dependent on the behavior of other hft participants would seem to be a little worrisome from this perspective), though depending on how long you’ve been trading it/length and quality of backtests this may be somewhat mitigated (and assuming the goal is to use the money to partially invest in better systems so you can come up with more ideas then this is probably more of a reason to go for it while the gettin is good rather than to not do it at all). I might also be misunderstanding the context under which you’re talking about the time quant, but if you’re basing your models off this as well then this would actually increase my confidence in the potential durability of your signal since it seems like there’s a plausible story about how this is roughly the correct abstraction for your system, and some of the other players who you’d expect might already exploiting sub anomalies are too caught up in the dynamics of the tick by tick trades and the order book to be looking for it. (Ps if anything I’ve said is particularly irrelevant or stupid please let me know as I’m still developing my understanding of how to think about these things/how they work).
Thank you. may I ask to reformulate your question / suggestion in a more precise way, so I can better understand you concerns/questions/suggestions. Regarding the "dependency" on other algo - I do not see any issues here, as this is a natural state of affairs (I would say) since you always play against an opponent as every "buy" requires a "sell" and vice versa. I realize that my answer is too general, as the nature of your statement is not quite precise.
Not sure how strong this point is, but here goes:
I guess what i was trying to say is that everything I’ve heard (which could be wrong again I’m no expert ) indicates that higher frequency signals have a short life span relative to lower frequency signals, so you should have strong priors regardless of what your strategy is that it will probably go away sooner rather than later. If it is the case that this arises mainly because of competition (e.g. other participants discovering the alpha and competing it away, which would be consistent with the observation that high frequency signals having shorter life spans than lower frequency signals since the fact that your average profit per trade is on the level of single digit ticks, means that fewer people have to discover it for it to go away), then the fact that your signal is based on modeling the behavior of other such signals means that you are exposed not just to the normal reasons for a signal to break down (e.g. the fact that other people also find your alpha and do it faster/as fast, but also the breakdown of the alpha of the participants you’re detecting. Now as I put it that way I realize how dependent I am on this notion of signals breaking down mainly because they get competed away and not something like changing market regime, which I really have no evidence for, (I’m no expert), but it seems somewhat plausible and therefore a little worrisome (though again the whole thing could be mitigated by a long enough track record, and/or long and reliable enough backtests and unfortunately I don’t know enough to say how long that would be, and it could be the case that everybody’s algos obviously rely heavily on modeling the behavior of other hft participants, in which case this would be one of those irrelevant and stupid points I mentioned above).
What is the business plan here? Build up a track then market to institutional investors? Grow to strategy capacity? Or stay small managing your HNW-networks wealth?
The first suggestion is correct. We plan to develop a track record (would be interested to work with an established entity as well), expand the set of instruments that we trade (to include other futures and ETF's) and grow big organically.
Ok, well best of luck with the growth! Obviously the track milestones are 1 year and 3 year, and the AUM milestone is 100mm. For each one of these boxes you tick you will get more investors looking your way. If you already have a track getting it audited is important to investors.
On the business side, I'd just mention that you should be careful giving away equity to investors, as it can become an issue at a later date when they may want to redeem and you need to buy them out or if they start to put pressure on the business side to force your hand. A compromise may be to do a revenue share or reduced management fees, and to sunset this over a few years
what economics would you be willing to give up for access to more capital?
We are still contemplating this. What is the "industry standard" here, if there is a standard, of course. Or what are the possible scenarios that work in the real world? Thank you.
if you are looking to join one of the large multi-manager hedge funds, they payout between 15-18% (but you must be able to put to work 50mm-100mm minimum).
Chicago style prop shops pay 40-50%.
If you are willing to put up 1st loss capital, than you can keep 70-80% at a prop firm.
Here is the rub. You are playing in the time frames and the alpha sizes that are mainly exploited by the low latency guys (maybe not UHFT, but low latency non-the less). Your Sharpe ratios appear to be as good as that of HFT players (did I understand you correctly, is it 0.8-0.9 daily Sharpe ratio, so 12-14 Sharpe annualized?). Yet you are claiming to have negligible infrastructure overhead and no latency sensitivity. Frankly, given what you are describing, most people would either say that your strategy is either severely capacity constrained or that you are lying about the performance. I assume the former but a lot of people will naturally assume the latter.
** I know, HR on line 1
Sort of ot but would like to ask to you guys. In your opinion how many money it occurs to start a smalll HF/family office? (just curious)
Observed...
HF break even depends on strategy, could be 100m could be 300m. Lots of institutional type structure needed as you take external capital.
To be a "proper" FO would be about 30-40m due to cost
Prop shop could be 1-2m
big1 got a question for you...but i can't PM. could you msg me?
Thanks, gotcha. But apologies in advance for mu ignorance, what’s a prop shop?
Thank you, This is exactly why we are thinking about a prop shop as LLC.
I'm not giving you legal advice here, solely my own opinion - but from a regulatory/corporate governance perspective, sounds like you guys are pretty loose. Maybe too loose.
Structural Considerations
Where are you forming your entities? Some states (Wyoming for example) offer incentives to form an LLC there which may seem good to a fledgling manager (saving a couple hundred/thousand a year in formation/taxation costs), however when investors learn you are domiciled in Wyoming they will laugh you out of the room.
Where are your investors coming from? I have helped set up more funds than I can count and almost none of them utilized a single LLC as their "structure". Even if you are only taking in HNW money, you want to have a structure that is flexible, in case you end up having success and wanting to raise more money down the line, especially if that money is coming from ERISA, US tax-exempt, foreign institutions, etc. A Cayman master/feeder structure is extremely popular for that reason - it allows you to raise money from most of the main sources and invest it most easily. Flexibility is key, because it is expensive to set up a fund properly, but way more expensive and perilous to the business to "fix the structure" midstream.
You should likely set up a management company that you collect your profits through rather than participating pro-rata with the investors through an LLC. There are a host of reasons for this that you can google.
I implore you to get professional legal help to set up your structure and operating agreements. It is all too common for new managers to take shortcuts and "save money" on the structuring of their fund entities and the negotiation of their governing documents, and then end up a couple years down the line spending way, way more money to litigate/restructure their way out of tussles with investors and counter parties. Private funds investment law is very complex and requires a deep bench of securities law and tax law talent to navigate successfully. Do not go it alone.
This is super important, so it bears repetition: Do not skimp out on setup costs now, because the money you save will not be worth the headache it causes down the line if/when you want to scale your fund. There are a million ways to get in trouble with the law, your investors, or your counter parties, the vast majority of which can be avoided entirely through proper fund setup.
Other Considerations
Securities law does not look favorably on verbal agreements and fast-and-loose fundraising. To do this right, you need to fashion legitimate partnership agreements, private placement memoranda, subscription agreements, and other governing/disclosure documents. You want to be sure that at any point, you can say "my investors were clear on the risk, and everything was handled transparently and in detailed writing".
There are a host of regulatory filings that you have to make, depending on where you fit into the '40 Act. I would spend time and money figuring out what type of manager you are, and therefore what is legal and what is not as far as your fundraising practices and regulatory filing obligations are concerned.
TL;DR - a lot of fledgling managers think that managing money is all it takes to get into running hedge funds. They fail to realize that a hedge fund is a business, and separate from just putting up nice returns. The suggestion that all you need is a solid investment thesis is as ridiculous as saying that all a restaurateur needs is a good recipe. Without proper structuring, regulatory compliance, corporate governance, etc., you are essentially setting yourself up to be the target of (warranted) concerns about your legitimacy. The world of private funds is very insular and niche, and so it behooves you to get in touch with a law firm that has a track record to help you set it up.
Great comment! I very much appreciate the scale of the comment as well as the details you provided. Exactly what I wanted to know. Is there any possibility to communicate with you off-line?
Sure thing - PM me.
I recall Nevada as an attractive place to domicile an LLC...
DE LLC's are the way to go, period. It's just "what's market".
2) LLC is usually the way to go for this business. 2-b) you guys need to figure out if this is a hedge fund or CTA. What products you’re trading? This will determine if you’re going to be a hedge fund or a CTA. Their basically the same thing but different regulations are applied to them... if you’re trade futures... setup a CTA... these are basically backdoors for a hedge fund structure without having the hedge fund regulatory requirements 3) consider yourself lucky to be able to deploy 1M. I started my company and grew it from 10k to about 800k due to trading performance in 18 months... smooth equity curve.. low drawdown. How many investors did I get or can I market myself to? ZERO. Apparently investors only invest money with managers that have bad performance and have been in the business for 10+ year. So it’s a tough road... but again.. if you were able to get 1mm just to start this venture.. good for you.. must be nice! 4) many many more things to think about There’s a lot more.. feel free to message me your email.. maybe we can setup a phone call and learn from each other.
Good points. One correction though. We will not be just "taking" money. The first investors will get a piece of equity in our Company. Therefore, the LLC will be trading on its own account with its own funds. The allocation and the strategy will be governed by the LLC's BoD. I see your point about investors, we understand that an "agreement in theory" and actual "opening the purse" are too "big differences". But hey, we still have not lost our drive to get it done. Cheers!
One more thing.. I see your point that you’re not “taking” anyone’s money but rather the first investor is taking a piece of the LLC... the LLC because it’s in the business of buying and selling futures HAS to be registered with the NFA and the principal or principals of that LLC will also have to register
The other discussion people are having whether attribution this to hedge fund/cta/prop shop doesn’t really matter. I know plently of CTA running HFT strategies and vise verse. Main difference occurs due to instrument being traded rather than the strategy being traded. Hedge funds usually trade futures AND equities And have to comply with FINRA (equities) and NFA (futures). CTA only trade futures and have to comply with NFA. Prop shops can trade both if they’d like.. so really the only question you need to ask your self and the team is.. do we want to register with NFA and FINRA? If you’re only trading futures... CTA will do.
I think you are on point with that investors are only willing to invest with established players as to they don’t care about performance as they do about their bias that a well know fund is better than a smaller more lean fund. It also means everything to have institutional money in the fund. Confirmation bias, ladies and gentlemen. Also an exemption can be filed with the NFA in most cases for a small fund.
As the type of strategy is kind of HFT, the hedge fund structure is the one way to go, due to the execution lag of placing orders across different managed accounts.
They really need to look at prop firms.
It’s realy a prop strategy - not hft not hedge fund not cta. Cta really aren’t trading on minute time frames. Also no one investing in a cta or hedge fund would understand giving up 60% of profits. A prop firm wouldn’t have a problem with that.
This thing really only scales to needing 10-30 million in capital as explained. And that might be more than necessary. It doesn’t take a ton of margin to back positions when not carrying overnight. And on these type of products I doubt it could scale to over 200 e-mini contracts in total (total risks across equity futures)
Haha, I literally just asked this question but in regards to a real estate investment fund. I've not done anywhere near the due diligence yet but an LLC is definitely the way to go as it protects you from bankruptcy and lawsuits in a worst case scenario. Bankrupting a company you invested your own money into SUCKS, but wont ruin your life financially and legally. I also think in the medium-long run it'll be a lot easier to raise capital from outside connections if you're an LLC. Trying to convince people to just transfer their money straight to you seems a bit shady.
The cons of starting a company are mainly that you'll have to shell out some cash to lawyers and accountants, but for every other purpose it's the way to go.
Have you guys had any success so far?
In one word - yes: 1) we've got a minuscule investment (first step); 2) updated our platform by fine tuning our algorithms; 3)moved to NinjaTrader and went on-line on 11/20/2019. 4) currently in 15th week of live trading, with last two weeks being the best weeks for us so far; 5) got people interested and even one client, who invested (so far) a small amount with us. In summary - moving forward, but slower than we thought we would. Well, we were (and still are the outsiders), so we learn this "world" while working, likesome sort of OJT :) Would be happy to discuss the details, if it seems interesting.
Voluptatem totam itaque facere facere repudiandae repellendus quasi porro. Sit qui quasi eveniet nostrum. Quae ut laborum aut commodi nemo praesentium in.
Voluptatem libero velit tempora maxime asperiores sit modi delectus. Aut deleniti veniam asperiores modi accusantium cupiditate. Iste est optio unde sed et quo. Asperiores aut sit hic ut. Earum quasi ut tempora corrupti incidunt. Possimus ad aut aut repellat perspiciatis.
Libero esse inventore quaerat non tempora. Laborum corrupti deleniti alias ea neque quos. Et est nihil nesciunt excepturi.
Nihil facilis nisi cum porro sunt suscipit. Aut exercitationem repellat dolorem libero. Corporis animi provident odio laboriosam in. Sed aliquid quo eius asperiores voluptatem.
See All Comments - 100% Free
WSO depends on everyone being able to pitch in when they know something. Unlock with your email and get bonus: 6 financial modeling lessons free ($199 value)
or Unlock with your social account...
Consequatur veniam non mollitia veritatis. Dolorem explicabo vel dolorem debitis possimus eaque. Ratione omnis natus porro assumenda omnis ipsum saepe. Dolor sapiente commodi dolore temporibus. Ut sapiente amet doloribus.
Dolorem nam laudantium sed officiis ipsam. Totam nobis eveniet eos et nostrum iste ducimus. Sequi sint in sint quidem odio. Itaque exercitationem laboriosam et ut unde cumque.
Doloremque id dolores expedita ab modi temporibus in nihil. Voluptatem enim iste id sed. Possimus et qui voluptatem reiciendis distinctio voluptatem. Cum velit suscipit numquam dolorem et. Laboriosam magnam et molestiae amet quo praesentium corrupti. Et quibusdam tenetur repudiandae quas. Culpa non inventore consequatur voluptatem et.
Fugiat earum dolores laudantium. Sit quae molestiae id dolores. Quos commodi hic ut odit non est. Voluptatem qui sint quae fuga ex repellendus dolores.