gut;589724 wrote:$100 means nothing to me. You have to talk in terms of VaR and ROI. There's a cost of capital.
Herein lies the beauty of selling a bot in downloadable form. The denominator in your ROI becomes mostly fixed, and the overhead is ALMOST nonexistent, as the bulk of the investment in creating such a software is a one-time investment on the front-end. You don't incur any default per-unit costs (once it is created, it can be downloaded as many times as needed, so long as there is adequate bandwidth. Thus, you are able to invest more into marketing (often measured in CPV, CPC, or CPA) and the monthly costs, such as hosting the download.
Here's why it makes sense to the person selling it to the masses, even if it has shown to be able to make a nice return: Suppose in tests, he's seen a positive return in four of six months, and an overall return of $1.05 on the dollar. This would be a very valuable system, yes?
Ultimately, the model to the consumer is very similar to the model of the producer. Let's use the hypothetical we've already got.
Consumer model:
w = the ROI of the software to date
x = the fixed cost of the FX bot/software (for our purposes, we'll say $100)
y = the investment
z = the overall return
w(y) - x = z
In our example:
1.05y - 100 = z
Now, with the producer of the software, the model is again similar.
a = the fixed cost of creating the software for the first time
b = the gross profit from selling each unit
c = the chosen cost from selling each unit (if one chooses to create a CPA model for an affiliate program)
d = number of units sold
e = the number of months the sales model has been running
f = the fixed monthly overhead cost of hosting such a software
g = the overall return
(b - c)d - e(f) - a = g
Now, let's flesh that out a bit. You can find adequate hosting for, say $5 a month, and a domain name can be gotten for about $12 a year ($1 a month), so monthly overhead sits at about $6. Suppose you want to pay affiliates (ie contracted sales people) 60% commissions on the $100 software. And suppose it cost $2K to build the software and the site where it is available. Then the model would look like this:
(100 - 60)d - e($6) - $2000 = g
40d - 6e - 2000 = g
Now if a person markets properly, d would only need to equal 50 to cover the startup cost (we could factor in taxes, but you get the gist), and then every additional sale on top of that covers e up to 6.67.
So the possibility for selling a software that shows to be profitable in the FX market, if done digitally, could indeed be VERY profitable.
I've never been involved with such a software launch, but I know a couple people who have launched their own to the general public and made as many as 4000 sales in a week's time. That's pretty profitable, I'd say.
gut;589724 wrote:And I disagree with you. With leverage virtually any GOOD system has significant value, even moreso when you consider funds and banks have access to more leverage, better borrows, and faster execution than joe average. Any system that has an expected value greater than its transaction costs is worth big bucks to hedge funds and investment banks. Such a system is never available for purchase to the average schmuck. Yeah, I suppose there are some with very marginal profitability that funds stay away from because of tail risk. Again, goes back to luck that you can make money with such a system if you are lucky to get out before it inevitability fails - that tail risk can't be ignored, you make money for many months and then it's all wiped out in a few days. You say the pattern changes, of course, but it's almost impossible to see that pattern change before you're wiped out.
In short, buy a bot if he needs help with execution and risk management. If he wants to make money, though (assuming he has a talent), he has to learn some fundamentals so he can express his macro ideas in a trade.
I do agree with much of this. The ones that are hugely successful are indeed the ones that are never made available to the public, and it IS important not to rely on a bot for years of income. However, even many of the bots made available to the public are profitable in the short term. Just as with any trading market, it's always about pulling out of one at the right time and investing in another at the right time.