Well for one I wouldn't be interested in doing such a big amount as soon as we start.
Why not? The system should work regardless of the amount right?
Instead of using escrows which completely complicate the whole service (I'll explain how so later) I would rather do a smaller amount and build up the trust between us both instead.
Why would I rather do this?
Well, for one, an escrow completely complicates the whole service and this is why:
You send the coins to the Escrow which comes out of your pocket
I send a collateral to the escrow which comes out of my pocket
I fulfill the service with my own or others coins which will complicate other customers investments
I then am stuck when it comes to sending out other peoples investments as then I have to attend to yours
I go to look for something I can make up for via Bitcoins but cannot since I used them as a collateral for you
Then we will both be angry and confused with each other.
You misunderstood the process.
1. You send amount X to the escrow, post the tx-id here. Escrow confirms receipt.
2. I send amount X directly to you, post the tx-id here.
3. You do what you do for 48 hours.
4. You return 1.25 * X to me, post the tx-id here.
5. Escrow confirms that I received the payout and sends your collateral back to you.
You hold the entire investment worth of coins while you do the gambling. The amount deposited at the escrow is to guarantee that I receive my investment back (but no profit), if you fail to deliver.
as for this:
Expectations / expectation value is a fundamental concept in the field of statistics. Any strategy that, in the long run, is expected to return more than what was put in has a positive expectation value. This value and its associated variance can be computed using statistics. Please don't try to bullshit me with math-terminology, it won't work.
Well in the extremely rare case that has not yet occurred in my many tests, the expectations of the statistics have returned fine. And as I said if in the extremely odd occasion something goes wrong, all coins will be returned via backup sourcing.
In your many tests? So you did tests to confirm your strategy? Rather than a fundamental analysis of the probabilities involved? Also, the sentence "the expectations of the statistics have returned fine" may sound impressive, but it doesn't actually mean anything.