Scalping and coin tossing

[Euler]

I would find it quite hard to lose 10 ticks 10% of the time.

[hgodden]

If the mad bomber was around then you'd use your skill and judgment (or just bloody common sense!) to realise that you shouldn't be offering money to the lay side

Most scalpers probably only lose 1 or 2 ticks most of the time when it goes against them. A 10 tick loss sounds pretty drastic to me! I remember Peter saying once how people always overestimate the chance of the market flying off in one direction and burning them for lots, doesn't seem to happen that much, at least if you're quick on the draw

[Ferru123]

Well that's not high probability then is it?!

Since when is a trade with a 90% strike rate not a high probability trade?!?

An analogy. Let's say you lay horses at 10.0 on Betfair. Ninety of the time, if the market is right, you'll make money. 10% of the time, you'll lose money. When you take commision into account, it's a losing proposition overall.

I know a highly successful trader who bets on reversals. He has just a 25% strike rate, but still makes very good profits.

Your success or failure is determined by your edge, not your strike rate. You could have a 1 in 10 strike rate, but if you get 20 ticks on average when your your trade does succeed, it doesn't matter...

Jeff

[Euler]

If the mad bomber was around then you'd use your skill and judgment (or just bloody common sense!) to realise that you shouldn't be offering money to the lay side

Most scalpers probably only lose 1 or 2 ticks most of the time when it goes against them. A 10 tick loss sounds pretty drastic to me! I remember Peter saying once how people always overestimate the chance of the market flying off in one direction and burning them for lots, doesn't seem to happen that much, at least if you're quick on the draw

Spot on!

[superfrank]

You're stating the bl00dy obvious with all these examples Jeff.

When I refer to high probability it is in terms of overall profitability, not a simple win/loss strike rate.

[LeTiss 4pm]

If I lose 10 ticks in 10% of my trades, then my trading has suffered a drastic reverse in fortunes.

However, the point you're making is valid, in the sense that novice scalpers will often tread water by winning 7 consecutive trades, but then having that profit wiped out by a bad one.

[Ferru123]

If the mad bomber was around then you'd use your skill and judgment (or just bloody common sense!) to realise that you shouldn't be offering money to the lay side

All sorts of things can cause the market to spike, and I'd say that spikes aren't really predictable unless the Mad Bomber is clearly on a mission!

Most scalpers probably only lose 1 or 2 ticks most of the time when it goes against them.

I'm approaching this from a theoretical basis, and I'm assuming a market that's random. It may be that the market spikes less than a random model would predict. But if we're establishing whether there's a mathematical edge over a billion trades, don't we need more to go on than a general impression that you lose 1 or 2 ticks most of the time?

An analogy. There used to be a hedge fund called Long Term Capital Management. They had sophisticated mathematical models that left them in little doubt that they had an edge. Investment banks lined up to be their clients. A few years after starting, they lost over $4 billion. Before risking my money, I want to be confident I have an edge, as I've been fooled by randomness too often in the past!

Here's another consideration. Let's say that, when we arrive at the market, most of the money has been matched at 4 prices. I used to think that that meant that the market had decided that the horse's true price was represented by those prices. But now I'm not so sure. It could be that, due to randomness, the market happened to have settled on those prices rather than going wandering off. If that is the case, then surely we can't assume that the market won't go careering off at any moment.

Jeff

[Ferru123]

When I refer to high probability it is in terms of overall profitability, not a simple win/loss strike rate.

Me too - although that wasn't clear from your earlier message, hence by 'bloody obvious' comments.

A 90% strike rate with an edge would be great. But to get back to the original question, I don't see how you can be confident of achieving that over a billion trades using reverse book scalping...

Jeff

[Ferru123]

You're probably made enough trades over the years to be confident beyond reasonable doubt that you have an edge. I say 'probably' because I'm not a statistician - that's not a dig.

If I'm simply offering money to the market and taking the available price when the market has run its course, then I have a limited downside and an unlimited upside. The market will spike against me sometimes, but overall it will spike just as often in my favour. But when you have an unlimited upside and an unlimited downside, you're relying having a high enough strike rate to compensate for the times when the market goes against you. And I think that whether you do have that edge can be established only through a large number of trades - I think the market is too random for it to be modelled theoretically (at least not without lots of empirical data).

Jeff

If I lose 10 ticks in 10% of my trades, then my trading has suffered a drastic reverse in fortunes.

[LeTiss 4pm]

Jeff, you are full of enough hot air to blow a balloon to France!

You keep coming out with various trading analysis, even offering advice to newbies looking for help, yet by all accounts you still play around in training mode

Listen to your own advice, stop making a lot of noise, and start making money instead!