Start a prop firm with cTrader today
Recently I have been listening to a couple of old podcast episodes with a guy who works as a risk manager at a sports betting company in Las Vegas.
There is definitely a subset of this newsletter’s readership that enjoys the dealing desk nerd stuff and, if you are one of those people, I recommend listening to this as there will be a lot of parallels that you find interesting. It may even bring back old memories for those of you that used to have a sports book (or still do).
The basics of sports book management seem to be…
- ‘Move the line’ – ie. change their odds to reflect risk on the book
- They limit bets due to risk limits
This to me is exactly like what most brokers do. You skew pricing to attract more flow on one side to balance risk. You limit order sizes to stop people from giving you huge potential exposure. There is even a futures market for them to hedge in.
And this brings us to props.
One of the facets of props that we’ve looked at before is the fact that props have an ‘interesting’ leverage structure.
In theory you have something like a 100:1 account. In practice, you cannot lose more than, for example, 10% of the margin in your account.
So let’s say you have a $100k notional value account. At 100:1, that means your margin is $1k but you can only ‘lose’ 10% of that – or $100.
If you assume a ‘normal’ margin close out is at 50% of account equity, that would mean you are effectively trading with something like 500:1 leverage with one of these accounts.
Note that some props close you out at 5% of account margin, which would mean 1000:1 leverage.
This is obviously a lot of leverage to use. From what I see, a CTA – probably the most comparable type of trading activity – might use something like 10:1 leverage. And this would be considered a lot. Most closed-ended funds are capped at 20% of net assets or 1:5!
But if you are a prop, your challenge conditions are theoretically a means of determining whether or not someone that’s taken a challenge is a good trader or not.
However, given the ‘real’ level of leverage on the accounts is so high, it must be extremely difficult to actually work out if that’s the case.
You might be able to tell if someone is cheating you (moves into profit fast) but distinguishing between a gambler and a ‘trader’ would be much more difficult.
This then has problems for managing risk.
Brokers can use two way flow to offset their exposure. They can skew their price to take more flow on one side. They can also use the underlying market (or an LP) to offset their risk.
If you are a prop firm then there is no two way flow and you cannot skew to offset. This leaves you with hedging.
But if you have basically found people who were profiting at 500:1 or even 1000:1 leverage, there is no means by which to say who is a gambler and who is a ‘good trader’. Given that the average challenge passer crashes out in about three weeks, they are probably nearly all in the former camp.
This leaves you with two problems.
Firstly, the obvious one, is that you can’t know who to hedge.
The second is easy to understand if you just imagine hedging everyone that passes a challenge. The account sizes are often huge. For example, a prop offering a $100k account is typically referring to the account’s margin, not its notional value. So a $100k account at 100:1 leverage is $10m notional.
You have no margin from the client to fund these positions with the LP because…you are offering a funded account?
Finally, if these people blow up, which they very likely will, your account with the LP is cooked and you just lost many, many times what the funded trader put down as margin.
Which leads us back to the discussion of the risk manager.
Sports books aside, in the gambling world, there is – for the most part – no market you can go to in order to hedge risk. There is no underlying roulette table market or slot machine exchange which you can use to hedge out your client’s exposure.
The result is that you have to ‘tweak’ the settings to limit your liabilities. This is why roulette tables have two non-coloured spots. Otherwise you could, although you’d still lose eventually, much more easily keep doubling up your bets each round on a specific colour, until you won. Similarly, slot machines do not offer uncapped potential payouts.
For props, the most likely method is to make challenges more challenging. But this has its own challenges. The more difficult the challenge is, the more random the outcome, and the more impossible it becomes to figure out who is good and who is not.
Maybe the simplest solution is to just give them a live account equal to the amount they paid for the challenge?
I don’t know.
Ask your risk manager.