Which strategy would you trade?

[Eldredge]

calculate the sharpe ratio, the best ones have high sharpe and most trades

Thanks for the suggestion. I'm a little rusty on my sharpe ratio calculation, but I will take a look at it.

 

[Van_der_Voort_4]

Can you kindly explain your logic?

Looked to me OP's strategies trade about once a day for every day in a five year period? Total trades from 784 to 1264 depending on strategy. Each year ~ 250 trading days, so trading signals are either once every trading day or once every couple of trading days?

Worst case scenario: Only 4 negative months out of 5 years, not bad at all? How many of us can claim such results?

there is no logic to the explanation,it comes from experience

I only looked only at 4th scenario,at first i was sure OP is talking about 4 to 8 months test,

time to recover to break even is too short to have that many losing trades without this being curve fitted and effected somehow by stop loss implemented not being realistic.If net positive trades are that good why there is so many losing trades.
If these presented results were truly robust the amount of trades to produce these results in given time period would need to be X times higher.I can't give an estimate by how much without doing simulated tests.
The give away is time spend in a draw down,when you break this down zero negative months is not what you want in 5 yrs period

 

[Eldredge]

What was the AUM in the simulation?
What asset class is this?

WIth this information, I'd pick 4 to run with real money.

I'm not sure what AUM is. I ran this back test in a spreadsheet with imported daily data. Trading NQ YM EMD & RTY. Thanks for the opinion.

 

[Eldredge]

Easier would be to tell you which strategy not to trade. For me that would be the first.
Can't you trade all the other strategies combined or do you need more capital for that ?

Thanks for the suggestion. I can't trade all of them at once because they are already combinations of the same instruments.

 

[Eldredge]

You have diligently carried out back-testing over 5 years. But can you say which you would be able, confident, and happy to trade continuously - and to scale up - over the next 5 years?

Excellent question, and the reason for my post. I like to be in the market, so I prefer more trades (I don't think this is necessarily a good trait, but it is reality). Like everyone, I don't like draw-downs - the draw-downs are close enough that this isn't a factor. I hate periods of no profit, so the shortest breakeven is appealing - five days' difference doesn't sound like much, but I know an extra week to get back to even will feel like a long time. I would like to scale up, and the most scalable option is the least profitable. Naturally, it's all about the profit.

So, each strategy has some good and some bad - the answer is not clear to me, so I thought this thread would help. Thanks again for the input.

 

[Eldredge]

They are all profitable and since no strategy that back tested successfully guarantees future successes, I would trade all four going forward and use real money and real trades to down select.

I agree. Thanks for the suggestion. I am inclined to choose one and scale it up with profits. I like the idea of trying them all, but that would take many months. Since they all use the same instruments, I can't really trade them at the same time. Maybe I will choose one and paper trade the others to see what happens...

 

[Eldredge]

4

Thanks!

 

[Robert Morse]

I'm not sure what AUM is. I ran this back test in a spreadsheet with imported daily data. Trading NQ YM EMD & RTY. Thanks for the opinion.

So futures. That is good but you need some reference to know how much capital you needed to create that profit, how much margin was used and what your returns were. It make a big difference if it required $1mm or $25K.

 

[Eldredge]

4th,time in a draw down as a measure of robustness

I would not trade any of them with real money

At first i was sure this was over 5-8 months period,then i read your post again

12 to 14 trades on average over 22 trading days period with 1% draw down from the final account balance.

I would prefer more "rusty" strategy if i intended to trade it LIVE

Your stops are just not realistic,my guess is you buy the dip in strong trending market.

imho ,if logic behind the strategy was robust to my comfort level i would like to see higher draw down amount and more trades in longer time period tested.

Thanks for your thoughts. What do you mean by "rusty"?

If anything, my stops are much larger than many would use. The only stops are: flat at the end of the day, and whatever the balance in my account is. Intraday draw downs may be slightly larger than what is shown - but not enough to trigger a margin call. I was trading a version of these strategies on 9/11 (real money) and survived without significant damage (I had open positions that I did not close on 9/10 - that wouldn't happen with the strategies under consideration). A day like that teaches you something about the false security provided by stops...

 

[Eldredge]

What is the "theory" behind these strategies? For example, before I try to develop a strategy I first have a theory (i.e. after extreme selling, a bounce is likely which can be profited from).

I do the same. I don't really want to discuss the particular strategies at this point, but thanks for your input.