I would think of this in terms of returns and risk, what are the alternatives, and what is the client's AUM.

What would I pay somebody to make me a guaranteed 25%? If there's no risk than I would probably be willing to let them keep 80% of the profits as I'd still be getting a no risk 5%, which is better than any other opportunity. However I would probably want a more 50-50 split as I'm providing the capital.

All strategies are only measured relative to alternatives. Absolute values of returns or sharpe ratios don't have much meaning. 25% / 25% risk would be a sharpe ratio of 1, so probably not that great compared to other strategies. If you can shrink the risk to 10-15% then we're talking as you're better than many other offerings.

Lastly, remember that every strategy is only as valuable as it can make money for the purchaser. If the purchaser has 100M AUM, then a 1% return is worth a whole lot more than a purchaser with 1M AUM. Likewise, strategies are capacity constrained and some cannot run a whole lot of capital, others need very favorable execution conditions only accessible to those who can do business with a prime broker.

Strategies have no intrinsic value, they are only as valuable as how much money a purchaser thinks they can make.

Personally, I'd consider where the 25% return would come from going forward, if the idea is to invest over the long-haul as part of an overall asset allocation. Traditionally, my understanding is that hedge funds were for hedging, not for absolute high-risk, high-reward. They were a form of insurance, in some sense. Institutional investors and high net worth individuals ("family offices") would add hedge funds, in the same fashion that individual retail investors would add bonds and other investments to a stock portfolio. The problem with the question, as stated, is that there is no risk information. How much variability in the return might one expect? What is the risk to the capital? Etc. The question is posed as if a bank were offering a government-backed CD at 25% annual return. Even if the track record were 25% return for the past 10 years, the question is, what return and risk can one reasonably expect in the future?

Standard hedge fund fees are 2% of AUM and 20% of returns.

There are very few with consistent returns over 25% over the long term as it is next to impossible. Medallion Fund (Renaissance) apparently does, but they are closed to outside investors. Before they closed the fund, they were charging something like 4% AUM and 40% of returns.

Standard rate for prop traders is ~30% net of high watermark, paid quarterly.

I know that Medallion fund and also Blue Crest wanted to take on more risk with their trading.They felt that "clients" would not be happy with the high levels of volatility and drawdowns. Medallion made about 21% in 2016 and Blue Crest about 40 %.So,it was obviously worth their while.

Following the question from this thread, I wrote a series of articles on the subject. Some might find them interesting, especially Part III.

A Price Tag on Alpha - Part I

A Price Tag on Alpha - Part II

A Price Tag on Alpha - Part III

Hope it can be helpful to some.

Hi Guy, thanks, I did find your articles interesting.

I do have one comment, and possibly, if you wish, a suggestion for you for a Part IV (?) as follows. Your discussion is based very much on the familiar concept of TIME VALUE OF MONEY. That's good but, for an individual trader, what about the MONEY VALUE OF OUR TIME?

Here are some questions to ponder:
- How much is my time really worth to me? (This may have little or no relationship at all to how much I got paid "in a job").
- How long will it take me to develop a trading strategy to license to someone else, as compared to writing a comparable strategy for my own use?
- What value do I place on managing my own money vs. trusting (i.e. risking) someone else to manage it for me?
- As a small trader / investor (well at least that's what I am anyway), how can I best leverage my skills to earn myself a better return and/or at lower risk than simply trading my own (little) account?
- If I use MAR = CAGR% / MaxDD% as my risk-reward metric for my own personal trading (or substitute Sharpe if you want, but then the calculation result is a little less obvious), how do I value a situation where I do not have to risk ANY of my own money, i.e. when the MaxDD% to my own money is zero?
- How much is information & good quality assistance worth to me in my trading efforts?

Obviously some of the numbers here are very personal and will almost certainly differ a lot for each of us, but I think even considering these questions may highlight some interesting points.

• I think your analysis is good, but one could argue that several important terms in the equation are still missing.
  
• At least in my opinion, there is a HUGE value to the good tools & the assistance that is provided by Quantopian AT NO COST.
  (Putting it a bit more bluntly, I would be lost with python were it not for the free help I receive that saves me hours, no DAYS, of my precious time).
• The Quantopian setup in general, and the allocation / licensing deal in particular actually looks pretty good when all relevant factors are considered.

Cheers, best regards, Tony

Hi Tony, thanks for the kind words. Probably a lot of the questions you asked could be answered with the epilogue of that series of articles which was just posted yesterday on my site. It started as Part IV but took a life of its own.

We can make money trading, no problem. The method is unimportant even if it is all that will matter. It is just that there are so many ways of doing it that after having chosen one, you are committed, until you find a better one that is. It is why you do all those backtests. It is to pick which one you will prefer, which one is more suited to your temperament and trading style.

The real question should be: how much will you have when you will stop trading? Will it be enough to compensate you for your time and involvement?

Definitely, your time is precious and valuable.

But there, like in the market, the rewards come from delayed gratification. Your time could be valued by the alpha you bring to the game.

Building your Retirement Fund goes from buying SPY where, over the long-term, you get about market average return, to building some alpha into your scenario.

The alpha is part of your compensation and it is all related to your skills as market participant. You should not view it as an instant gratification like some form of salary, but as a discounted and deserved reward for having spent all those years at playing the game.

Your objective is chart 5 in Building your Retirement Fund. You can pick column 1 which is technically no work and a measure of this long-term average market return, or you could go for the added alpha which will give you the equity line plus the cumulative withdrawals.

That is how much your time is worth! A real measure for your talent in extracting that alpha.

Should you find that it is not enough reward, then just jack up the alpha, find more capital, find better trading methods, and give it more time. Trading/investing is a long-term engagement. The biggest rewards are at the end of the line.

Here is chart 5 from that article which resumes my point of view.

Hope it answers your main questions.

As you say: "Should you find that it is not enough reward, then just jack up the alpha".

Well that certainly sounds easy enough!
Usually for me the problem is with that little word "... just..."

Cheers, Tony

Tony, yes, I can understand that.

You will find on my site over a dozen different trading strategies that could serve as basis to design your own, not counting the possible variations on those themes. They all provide that “just” that. I have also transformed some of the strategies found here on Quantopian to do the same.

The problem is not the choice, it is choosing one under uncertainty. I cannot say which trading strategy will be the best going forward. But I can make estimates based on some long-term backtests where the number of trades is so large that it will fuzzily respect the law of large numbers. In the sense that you can use averages as estimates based on what your trading strategy did over the years.

The following are all expressions saying the same thing and arriving at the same number.

F(t) = F(0)∙(1+ r_m + α)^t = F(0) + Σ(H.∙ΔP) = F(0) + n∙APPT

where APPT is the average net profit per trade, and n the number of trades. For a trader, to get his alpha, especially if APPT is relatively small, he will definitely have to trade a lot.

These equalities do not speak about which trading method was used. The first part express the problem as one of added skills (α), the second as a payoff matrix, the outcome of your chosen trading strategy (H), and the third saying that, whatever the skills, whatever the strategy, the number of trades will matter a lot more than you think since trading is usually for short time periods resulting in low net APPT.

It is what you have to juggle with in a way that is satisfactory to you.

Note that at the rate Leo Williams is going, being on pace for a 5,000% CAGR, I don't even come close. Leo, bravo. Love what you do.

Guy, thanks, yes, I read and understand what you write. I will certainly take a closer look at your site and see if I can pick up some new ideas that are also in keeping with my own personal trading styles (plural because what I do in my own account is quite different from my efforts here on Quantopian).

As for Leo Williams, yes amazing indeed!

At the Singapore Quantcon recently, Delaney Mackenzie was urging people to look at something other than price data, which of course is very good advice,
but here was I, wondering just how much benefit could REALLY be derived from using "alternative data"? Incredible!!!!!!!