Maximum sensible bet size for great setups

My general heuristic is to plan for 12 losses in a row. This will be very rare but can happen. You have to factor in the probability of the loss and the size of the loss to really know the answer to this. They are not the same thing!!!

Using my simple heuristic, take 20% and dividing by 12 gives approximately 1.7%.

Based on this you would assume you could risk 1.7% max on this trade. However, let's say you set a stop of size X and that stop has a 1% chance of historically being hit. You will see that this estimate may be quite conservative.

You could be more aggressive then by taking some small but higher probability of a risk of ruin and bet more aggressively. I.e, perhaps 80% of your losers have a 20 point stop, you could set your 1.7% at the 20% probability. These are all just rough estimates, you understand. Another even more aggressive posture would be to make the 1.7% the loss of your average loss which would make your max risk higher.

A 20% max drawdown tolerance isn't a very aggressive risk tolerance, I might add.

I think what most people don't understand is that there is a probability (unknown) of various losses and gains. It will benefit you to start to think in terms of probabilities instead of definites. okay I'm willing to take 20% drawdown at the X% confidence level.

In reality, most trading methods that work have a small probability of a quite large loss. This is because the uncertainty factor of the rare event is rather high. And, this is something that most beginners (not saying you are) do not really comprehend.


Quote from Ghost of Cutten:

Let's say you see a very favourable trade setup. For example, something with a 75% win rate and a 5:1 payout. Obviously, the Kelly Rule recommends betting way too much to be sane.

My question is what should be the absolute maximum bet size for really favourable conviction plays. If we assume that no trade can ever be known to have a win rate of over 75%, due to the inherent uncertainty of markets, what is the maximum optimal bet size for these slam dunk trades with high conviction and excellent potential payouts when they work? Assume a maximum drawdown tolerance of 20%.
 
Quote from Ghost of Cutten:

Let's say you see a very favourable trade setup. For example, something with a 75% win rate and a 5:1 payout.
Go all in. Everything you have and can borrow.
 
20-25% return per month is done via 5x leverage and taking 100 pips per week

I think this might translate into 10 ES points per week? Sorry, I don't trade futures so I'm not sure if this correlates correctly

5x leverage means that it will take you 20 horrible trades at 100 pip losses to wipe out your account -- if you have any trading skills, this should not be an easy thing to do
 
Most of those percents can only be achieved if you're on the business end of a pump or fraud. So the % bet is not much of an issue, it's who your cellmate will be.
 
I have never been so fortunate to have such a high profit factor, but here's how I would approach it in the absence of past trade by trade details. I like my strategies to have a max drawdown of no more than 20%, and my overall trading accounts no more than 10%. There is a 25% chance of 1 loss, a 6.25% chance of 2 losses in a row, a 1.56% chance of 3 in a row, and a .39% chance of 4 in a row. Assuming that there are limited opportunities for such a trade I think 3 in a row is likely enough and 4 in a row rare enough that I could consider 3 in a row the worst case. So to stay within my drawdown limits I would trade no more than 3.4% of my trading funds and no more than 7.1% of the funds allocated to this particular strategy. These are the amounts that would hit my drawdown limits. If these were hit I would re-evaluate the position size.
 
Quote from Kevin Schmit:

Go all in. Everything you have and can borrow.

You're joking right? :eek:


The only time I consider using max lev is if the probability is AT LEAST 90% ....with a TIGHT STOP
 
Quote from u21c3f6:

There is no problem with Kelly. The problems usually arise because of faulty inputs (such as believing you have a 75% chance of receiving 5-1).

Joe.
+1
 
Depending on your timeframe, what was max number of losing trades in a row past 10 or 20 years, How long to new equity highs, largest drawdown per 12 months, I prefer to lean more on what can go wrong, will. Also, keep in mind one's age, account size. If one is in there 20's, you have ages to recover, but in your 50's, better to be more conservative.

I once had 23 losses in a row on a method that tested no worse than 8 in a row over 10 years in currencies. I lost 23% of that account, took two years to get back to even. And has never lost that amount again, that was 17 years ago.

And something else you always want to consider, if you were long and the markets just shut down, for a week, like 9-11, when the markets open again, can you still trade? Never want to bet the farm, pigs get slaughtered.
 
Quote from Handle123:


And something else you always want to consider, if you were long and the markets just shut down, for a week, like 9-11, when the markets open again, can you still trade? Never want to bet the farm, pigs get slaughtered. [/B]

This point is often overlooked. What happens if you hedge ES with weekly index options and the market is closed for the week after a gap down? You have a big problem that's what b/c your options will expire worthless but you will still be on the hook for the loses on the futures.
 
Back
Top