OddThinking

Almost two years ago, I wrote an FAQ on my Job Status at the time – which is to say, my lack of job status at the time.

It is time for a new one. This is based on the questions I have been getting asked about my “new role”.

Julian’s Job Status

So you’ve got a real job again?

Yes and no. Mainly no, but with a significant amount of yes.

When did that happen?

Well, I managed to last 11 1/2 months job free, until I stumbled.

That’s almost 12 months ago? Why didn’t you tell us?

Yeah, sorry about that. It was partly because there was a very strong chance the project wouldn’t proceed, and mainly because the project has a veil of secrecy over it to protect us from competition.

Oh? So it is a start-up?

I guess you could call it that, yes. But not like a typical start-up.

No VCs. No funding rounds. No second mortgages. No working really hard for long hours.

No funding? How do you pay the staff?

Well, for the most part, it has been a partnership, so there’re no salaries to pay. It is a software project, so low expenses.

Actually, we did hire a contractor for a few months, to speed things along, and we covered that from our own pockets.

So what’s the project then?

Err… we monitor, um, financial markets in specific sectors, err… looking for opportunities for mis-priced securities, based on a, uh, model that incorporates fundamentals and historical outcomes, and automatically negotiates contracts on an exchange… err… yeah.

You work for a bank? You!? Really?

No! No! That’s not it at all.

Why are you being coy then?

Well, because the financial markets we analyse are gambling markets. I have become a punter.

You’re a bookie?

No. I’m a punter. I bet against bookies.

Actually, perhaps that’s not such a strong distinction. Bookies don’t work they way they used to. These days, you can bet on exchanges, like BetFair. Arguably, you are no longer betting against bookies, but directly against other mug punters. Of course, if the market is liquid enough, market forces should quickly make the two options equivalent.

In any case, I don’t consider myself a bookie.

Don’t be ridiculous! You can’t make money against a bookie. They always win.

Yes, bookies always do (excluding very particular circumstances where they haven’t got time to lay off a large bet).

It is also true that bookies have a very good track record of predicting the chances of a selection winning. (I know; I have done the figures. They are generally really good.)

That doesn’t mean you can’t also win when playing against a bookie. You don’t have to be better than them all the time – you only have to be better than them sometimes and be aware of when those occasions are.

What’s your web-site?

We don’t have a web-site.

I am writing a bot, to play online.

A bot? Oh, it plays poker for you?

No, not poker. Sports-betting. There are dozens of sports you can bet on these days – from the staples, like horse-racing, soccer, basketball and cricket, to more esoteric sports like bandy and fishing.

You can’t win. Sports matches are rigged!

Yes, that is a concern. It happens. We have to take that into consideration. We try to detect dodgy markets. If we see suspicious characteristics, we get out. (Perhaps, one day, we will try to bet alongside the criminals, but not at this point.)

So, what sports do you bet on?

No comment. I don’t want to attract the more sophisticated professionals to the market we are in, and I don’t want to scare off the less sophisticated professionals that are already there.

So, how do you work out what to bet on?

Obviously, that’s a trade secret. Our model is built by using subject-matter expertise (not mine!) to guide the construction of mathematical models (mine!) that are then trained and tested on large databases of historical data.

And you can really work out who will win?

Nah, we can only work out probabilities. Then we monitor the markets, and look for people who think we are wrong, and bet against them. We don’t have to be super accurate – merely more accurate than the market.

So, you are going to sell tips?

Only if they are no good! 🙂

The intention is that they will be good tips, and therefore we will use them ourselves to make money.

Is this legal?

100% ridgey-didge.

Is it moral?

That’s a question I gave some thought to. Some people don’t approve of this industry. I set about to find out why, to see if I would agree with them.

Objections seem to come in several flavours.

It is bad, because, in the bible, God says gambling is bad.

I dismiss this for two reasons.

The first is that the Christian bible says a lot of things that we, as a society and as individuals, choose to ignore as a source of ethics; we need more than a mere mention in the bible to determine how we behave.

The second is that I don’t believe it says that in the bible. I read some anti-gambling screeds that referenced parts of the bible, and the quotes were all so generic (e.g. against greed in general) that they could be equally applied to any source of income.

It is bad, because a small minority of people are problem gamblers.

I had an interesting chat with a friend with a PhD in psychology. She had studied problem gamblers, and was able to explain the how there were several categories in which they generally belonged.

I convinced myself, from this discussion that (a) I wasn’t a problem gambler myself (which wasn’t really in doubt; I don’t get a thrill from gambling, and generally avoid it,) and (b) I wasn’t doing anything to encourage problem gamblers.

Let me make this clear: I do not encourage people to gamble. In fact, let’s keep it simple. Don’t gamble! The odds are against you (excluding very particular circumstances…) and you are going to lose. Go find a more productive hobby.

It is bad, because it isn’t contributing to society.

Guilty as charged. I am as bad as a day-trader or real-estate agent. (I hope you’ll consider me, perhaps, a little better than a real-estate agent.)

My only mitigation – my entire career has been filled with regular software development roles that have contributed nothing to society, but have paid me well. I have written software to help the sex-industry, arms-manufacturers, and (may God forgive me!) those damn phone systems that ask you to “press 1 for sales”. I’ve worked on Digital Rights Management systems, against the customer’s interests. For six months, I was paid by the Australia tax-payers to work on a project that I left in exactly the same position that the management thought they were in when I started. I have been paid to worked for months on a project that got canned… and then I have done it again… and again…

I may not be contributing much to society, but for once, I am not demanding a regular paycheck from someone else for the privilege.

That’s not much of a mitigation, really, is it?

It is bad, because gamblers have little respect for the rights and dignity of the animals involved in horse-racing, greyhound racing and fishing.

I am not betting on any sports involving animals (excluding some very peculiar sports-players, who act like animals.)

Do the owners of the web-sites make it hard to scrape the data?

No, they make it really easy – and they charge for it. Professional gamblers are a significant source of their revenue, and they generally offer high-quality APIs.

Is it working?

On paper, yes. In practice, there are large fluctuations, so it has to run for a long time before you can be sure. It hasn’t been running long enough; our percentage returns still jump by large amounts each day – as do my moods!

How much are you betting? How much are you making? What is your rate of return on turnover?

No comment.

Are you working hard?

No. Working my own hours, which vary a lot in intensity, and vary a lot in what time of the day or night they are.

There’s been a few times where the system has been down and there’s been a rush to get it working again where there has been a bit of a push, but overall the hours have been good.

Is it finished?

It is running. It is risking my money. There are lots more opportunities to improve it, which I continue to chip away at. I still a long way from the point of diminishing returns to my improvement efforts.

So, once it is finished, you’ll be set for life?

No, as the market gets more sophisticated, the returns will dimish. This opportunity has a limited (and hard-to-predict) lifetime.

Finding an unsophisticated market with sufficient liquidity is tricky. This is the main reason for being a little secretive about the details.

What if there is a bug?

Shuddup, shuddup, shuddup! I could lose money.

More than once I have stared at my malfunctioning code and reminisced about having a civilised job where you get paid every month even if your code doesn’t work.