Hey-My 2 cents on this

If your algo can really kill it, borrow and leverage the funds for growth, slower but better.

I think most algos designed and are being sold to investors are one-sided(own opinions). As each algo is based on a traders bias- it will do the number crunching but tailor made to operate in one way. Does your algo incorporate machine learning and Big data, can it be applied in multiple products, how has it performed on huge volatility days in terms of position taking and risk management. What is your risk of ruin and how do you handle extended days of loss.
Most important question- trend/scalp and what percentage are exchange fees as a ratio of profitability per trade.
Before you demand $3M put it in a live settup- see the cost structure go wild- trading costs take over a third of profits and magnify losses.

Who is your clearing firm- Any rebates on volume trades and market liquidity provision?
If your Algo ticks all those boxes- dont manage money- lease it out for a fee per month.

Thinking Aloud

At first someone might dismiss this as just another of the many newbie’s dream you would come across however one thing sets it apart: the 23% return is reasonable target. Usually a marketing message with words like ‘Algorithm’ ‘international markets’ would be accompanied with promises of 100-1000% . Also the effort to qualify the 23% through a performance comparative analysis is a nice try even though the figures are inaccurate and maybe ambiguous (which asset managers ?)

Wanting $3M with some backrest results is what sounds so so newbie. Hehe lakini I understand ni vile tumezoea pesa ya corruption ikitajwa tajwa Hadi $3M inakaa pocket change. August 2018 I raised $5000 (peanuts in comparison to $3M)and it was hectic to convince the person. This how it went.

First I put $450 to a brokerage account because we know very well no one will invest by showing demo or backtest results.Except our mothers of course. Started fine and hit 30%+ first 6 months.

https://www.myfxbook.com.../avantgarde2018/2405231 . Closed due to ESMA

With 6 months proof of concept, registered in forums to try my luck in getting partners. At www.bogleheads.org, three people got interested. Also in www.aussiestockforums.com got some interest. Hapa Wazua sijawahi jaribu hata. But shock on me, the drilling people will do b4 trusting you with their money. Kwanza drop bracket words like algorithm trading when Interacting with potentials. Simply state what your code is doing. Badala ya international markets sema tu specific currency or those specific metals in your portfolio. Second they care more about the potential downside before the upside. How much can we lose in a day or month or overall? How do you know it is time to quit the system? What methods does the system use to forecast ? MAs, Regressions? Ukienda kutisha na ‘Machine learning’ be ready to explain the rationales in detail. Any interesting dataset ama ni the usual fitting historical data ? Note these are just some complex questions that require modelling to answer. Na by modelling I don’t mean pale MQL4, that can work only to a certain extent. You can easily get pissed off knowing the effort both time and emotional you put in only for more to be demanded.

So I recommend you try interact at Bogleheads.org but you really have to do your homework to attract the kind of money you mention. Hopefully smaller amounts from many people. I doubt hapa Wazua utapata pesa. Utapata the typical Kenyan mistrust ama the cat race mentality '23% iko down, me I can do 100%’ hehe

Good luck to us.

Hi Wachira

Thanks for the info. Well for the 23%...thats before fees but after fees the investor still compounds at around 19%. Guys may dismiss but kenyan asset managers compound at 10% and yet they manage collectively 1 trillion (look at the zamara yearly actuary survey it lists all asset managers with their specific returns)...so am sure we cant do worse.

The algo doesnt do intraday trading so am not worried about actual returns not being in sync with historical returns. The algo holds positions for weeks and even months!! average holding period is around 7 weeks per trade...and yes it is a machine learning algo, so it adapts as market keeps on changing so stop losses and exit positions are dynamic and not static

Well thats impressive if you put $450 and got a 30% return within 6 months. Personally I am well aware nobody will invest with you on historical data, so one must do proof of concept.

Well, I put in $2,000 with a broker 3 months ago and the algo recommended we "short" the eurodollar contract in anticipation of the fed raising interest rates in december 18th. Well, we were closed out on December 7th (friday) at a profit of $587 because the risk sentiment has been changing and equities are starting to have a sell off..and the code mined this info and recommended we close our short position..That comes to around 30% in 3 months since the trade was initiated in September 3rd...

Yes am aware of all the issues you raised such as drawdowns, markets traded, hedging mechanisms, average loss, average profit...etc... Obviously you cant raise all this issues in a blog but we have been talking to some investors and at that point we go into detail on what is the system logic, how performance comes about, how we hedge our trades etc. All in all the key thing for any investor is risk of loss and our algo is such that out of every KES 1, 70cts is invested in t-bills...so at no point can the investor BLOW UP!!Never...

Thanks for the recommendations...I will check on them.

P.S. dont worry i always get people saying they can even do 100% in a year.. ..If it was so simple Warren Buffet wouldnt be the richest person in the world...He is the greatest investor out there and guess what he attributes his success to:
1. Being born in America
2. having some lucky genes
3. Compound interest. basically his long term compounded return is 20% (50+ years)... He started with $10 million in 1965 and that is what has been compounded at 20% to get the $80+ billion he is worth now. Any serious investor will tell u compounding at 20% in the long term is no joke...So the guys who say they can do 100% dont really understand how money work

come on, can we stick to the topic of how one raise money or connect with a potential client? That will be helpful instead of ooh this or that performance

The issue is not the 100% or 270%....The thing is doing it consistently...That is STATISTICALLY IMPOSSIBLE..Markets are so RANDOM...Well even our algo in one year it made 77% and another year 60%...but also there is a year it made 8%...The key thing is dont make a loss in a given year; 0 > -ve. But all those returns when you do a compounded return comes to 23% or an average return of 25%....

Even bernie madoff ponzi scheme promised clients a consistent return of only 16% P.A (year in, year out)... Well, we all know that turned out to be a hoaz when the chickens came home to roost...