Risk Analysis - Critique

[Palindrome]

If anyone cares...

I'm trying to understand my performance over the past 11 months. I know how to google, and I apologize if I appear lazy. New system I designed.

I understand some of the below, but I'm looking for a critique and to help understand my performance better:

I know that my sharpe ratio is good, I read anything over 3 is "excellent"
But what about Sortino and Calmar ratio? Am I taking too little risk in my trading system? Any thoughts or input positive or negative would be great.

There are some egg heads on ET that love this crap, I'm just a street smart trader with basic finance background having a tough time spelling things correctly let alone analyzing this stuff.

Any help would be welcomed.

 

[truetype]

Summary stats are meaningless without an understanding of the underlying returns-generating process.

 

[Palindrome]

Its a futures system, directional based (not spreads).

I place bets on where the market will go... Take profits and take losses...long short (ZN, ES, Currencies, Metals, US Energy complex etc)

Average hold 2 hours to 3 days, no hold over the weekend.

What else are you looking for?

 

[traider]

Its a futures system, directional based (not spreads).

I place bets on where the market will go... Take profits and take losses...long short (ZN, ES, Currencies, Metals, US Energy complex etc)

Average hold 2 hours to 3 days, no hold over the weekend.

What else are you looking for?

What type of optimization did you do to arrive at your results?

Is it a single factor model, ie just looking at price?

 

[Palindrome]

"Is it a single factor model, ie just looking at price?"

I believe what I do is multivariate time series analysis, I believe that is what it is called. I understand how markets move, in basic terms.

The figures I posted are derived from IB, my brokerage account that clears my futures transactions.

Not sure what you mean.

Are you talking about my decision making? It's 80% systematic and 20% discretionary. The Discretionary decisions are basically not taking certain trades. So basically 100% systematic.

I look for trend continuation and trend reversals. Trading 101 I guess.


So I don't have any fancy response for you other than I trade.

 

[Palindrome]

I figured this out I believe:

Thats the average daily return. Makes sense, i've done 66% over the past 11 months (.20 *330 days )

The one factor that does not matter, but kind of does... I was trading half size for the first 7 months of the systems existence. Just getting used to the mechanics of it.

I think Mean return should be more like 0.30% * 365 = 109% per year. But to be conservative we will just analyze it as is..

Again, If anyone cares to help me analyze this and all those "Risk Analysis" Terms

 

[Gotcha]

I figured this out I believe:
View attachment 186885

Thats the average daily return. Makes sense, i've done 66% over the past 11 months (.20 *330 days )

The one factor that does not matter, but kind of does... I was trading half size for the first 7 months of the systems existence. Just getting used to the mechanics of it.

I think Mean return should be more like 0.30% * 365 = 109% per year. But to be conservative we will just analyze it as is..

Again, If anyone cares to help me analyze this and all those "Risk Analysis" Terms

I feel like this is a small number for a futures trader. In fact, % return for a futures trader has never meant sense to me.

Assume a 100k account, and what you're doing is trading 5 contracts. (hence, 20k margin per contract, more than adequate). If your ROI is 75% (rounding up from 66% for 11 months), this means you made 75k in a year, and hence only $300 per day. (assuming 250 trading days) But this is trading 5 contracts, as per my assumption, hence you're averaging lets say 1 ES point or so per day.

Now we could plug in other numbers, like 500k account, and 75% profit means 375k profit for the year, so now you're making more like 1.5k per day, which is of course fabulous, but who is trading 5 contracts with a 500k account?

Of course I have no idea how big your account is, or how many contracts you trade, but I assume most futures traders keep an amount of money in their account to sufficiently trade the number of contracts they trade with enough of a drawdown margin to cover 10 or 20 losses in a row or whatever they expect.

Its not like investing where you scale up, for most traders. If you can trade 10 contracts with a 100k account, its not like you would be trading 100 contracts with a 1mil account. But certainly, if you're trading/investing stocks, then these percentages scale up nicely and ROI is a more meaningful number.

In summary, I expect a profitable futures trader to make hundreds of percent per year. If you're trading 5 contracts, you more than likely do not need more than 100k in your account, and if you're trading 5 contracts and extracting anywhere from 3-5 ES points daily, then you're making several hundred thousand, and hence your ROI on your 100k account is easily 200% and higher.

In your specific case, I assume that you just have far too much money sitting in the account and hence why the % is so small. For a futures trader, a more meaningful number really is how much income you're generating either in terms of points or dollar value, and however big your account is is just a number that is the cost of doing business by having that margin there to initiate trades and cover several losses in a row.

 

[traider]

"Is it a single factor model, ie just looking at price?"

I believe what I do is multivariate time series analysis, I believe that is what it is called. I understand how markets move, in basic terms.

The figures I posted are derived from IB, my brokerage account that clears my futures transactions.

Not sure what you mean.

Are you talking about my decision making? It's 80% systematic and 20% discretionary. The Discretionary decisions are basically not taking certain trades. So basically 100% systematic.

I look for trend continuation and trend reversals. Trading 101 I guess.


So I don't have any fancy response for you other than I trade.

Your sharpe ratio looks way too high for what you are doing. If you can actually achieve that, you will be a billionaire in no time since futures scale quite easily.

 

[Handle123]

If anyone cares...

I'm trying to understand my performance over the past 11 months. I know how to google, and I apologize if I appear lazy. New system I designed.

I understand some of the below, but I'm looking for a critique and to help understand my performance better:

View attachment 186883


I know that my sharpe ratio is good, I read anything over 3 is "excellent"
But what about Sortino and Calmar ratio? Am I taking too little risk in my trading system? Any thoughts or input positive or negative would be great.

There are some egg heads on ET that love this crap, I'm just a street smart trader with basic finance background having a tough time spelling things correctly let alone analyzing this stuff.

Any help would be welcomed.

Wouldn't concern yourself with Calmar as that more of comparing hedge funds, like min of 3-4 years, and higher is better, means that they not had huge drawdowns.

Sortino, some like it more than Sharpe dealing with drawdowns, I have not had to use up to now, perhaps in the future I would.

Your sample size is way too low, 14 days is like nothing, I use 14 years, as I have gone from 300 to 3,000 to now of 15,000 plus sample sizes to know what the past offered for drawdowns, I always do separate comparisons as I average down on each scalp/day trade, because days where I lose, most often I lost cause of the averaging down and weeks it takes to recover, but drawdowns based on one lot now are more intra-day, but bottom line, averaging down for me makes more sense as most systems were developed to have many "BE+1 tick" based on original entry.

Am sure you have a bell curve done which represents best way to enter, you want to see 1-2 away from optimum to show they are profitable as well, meaning if a breakout, 2 beyond normal entry and 2 below normal entry shows no worse than breakeven, there is no magic number working all the time.

Really can not judge is you not risking enough since sample size way too low, and if you thinking of becoming a CTA, most people want to see stats for last 3 years.

And futures scalability does have limits, just about any market let you in with min. slippage but it is the getting out that will hurt you like at lunch in Indexes and even crude, not much in financials. I think most hedge funds don't do day trading is cause of the limits and commissions even when owning seats. Never heard of any "Dark Pools" dealing with futures.

Have you ran tests on how you would do if you kept over weekend? Perhaps hedge over weekend? What would it take to drop losing percentages on losses?

Good luck.

 

[bln]

I understand some of the below, but I'm looking for a critique and to help understand my performance better:

View attachment 186883

Any help would be welcomed.

Looks insanely good to me.

You are making an average of 0.20% per trading day, which pans out to ~50% for a year in linear return.

And you are doing this for a annual draw down of only -3.74%, which is the cool thing.

That gives you a ratio of 13.36x which is extremely good.

It is over a very short period of time of course, only 11 months. I'd prefer a minimum of 3 years to measure a system.

CALMAR measures accual real world returns to real world risk. It's my favorite metric.

Sharpe and Sortino is teoretical metrics and uses variance.