What is "consistency"?

[M.A 10]

"To be a successful trader u need to be consistently profitable".

I have read similar statements countless times on forums. Consistent profitability assures that a strategy is good enough to beat market in the time period in which a strategy is tested. Even most successful traders loose a percentage of trades. So consistent profitability is usually measured in different ways by different traders. Most popular, IMO, is to check monthly results. But does everyone take end-of-the-month results to measure consistent profitability?

 

Does consistency mean to be profitable every day? Every week? or every month? Or do u measure it by analyzing making "n" sets of trading history and then check if each set is profitable or not? Do u think it depends on trading style?

[Enigmatics 10]

  M.A said:

"To be a successful trader u need to be consistently profitable".

I have read similar statements countless times on forums. Consistent profitability assures that a strategy is good enough to beat market in the time period in which a strategy is tested. Even most successful traders loose a percentage of trades. So consistent profitability is usually measured in different ways by different traders. Most popular, IMO, is to check monthly results. But does everyone take end-of-the-month results to measure consistent profitability?

 

Does consistency mean to be profitable every day? Every week? or every month? Or do u measure it by analyzing making "n" sets of trading history and then check if each set is profitable or not? Do u think it depends on trading style?

 

I'll tell you one thing. I placed upon myself the paradigm that I need to be consistently profitable "X" amount each day to cover what I need every month. I think it's worked against me. There is almost a symphony of ebb and flow to doing this. There are moments when a quick trade for easy profits is necessary and those times when it's best to sit back and let the position breath, even if it means you don't have a big week.

Edited July 1, 2013 by Enigmatics

[zdo 20]

M.A.

 

re "What is 'consistency'?"

Historically the favored way for measuring managers is adequate slope and sufficient smoothness of equity curve.

 

:nowgoingofftopik???: I have found that it’s best for me to ‘measure’ consistency not on outcomes, but on the consistency of my own moment to moment and day to day perceptual acuity … (fwiw, that's individualized info…learned through naturally being a little bit more mutable than average, and across many years of often trading multiple systems at once, realizing the long hard way that returns are simply not consistent across time… strive for inner consistency … outer / returns consistency – forgetaboutit )

 

Steady, ‘consistent’ returns across a long sample often means the trader is running mm (too) far below the optimum… yet (somehow) finding a way to remain ‘exposed’ anyways…

[M.A 10]

  Quote

adequate slope and sufficient smoothness of equity curve

Gud one. I think having a look at equity curve will be enough to check profit consistency.

 

  Quote

Steady, ‘consistent’ returns across a long sample often means the trader is running mm (too) far below the optimum

 

There is no optimum level and we can always do better. I understand that even consisten returns doesnt mean optimum level. But does it mean "far below optimum"?

[zdo 20]

  M.A said:

There is no optimum level and we can always do better. I understand that even consisten returns doesnt mean optimum level. But does it mean "far below optimum"?

 

When I said "Steady, ‘consistent’ returns across a long sample often means the trader is running mm (too) far below the optimum…" I was talking about sizing mostly... not personal performance...

 

MM math has one sizing schema that approximates optimum-f... a 'sweet spot' sizing for the expectancy (has to be +) of a particular system. Setting aside the practical part that optimal-f is still to high and just focusing on the 'theoretical' - what I was saying is if you run at about a third of optimal - f (for most 'typical' systems) you will produce a little bit more of the smooth part of 'consistency' (as defined) (... and a lower slope too... btw.) ... very generally speaking though...

It's all really system specific.

...and one of the best 'professional' traders I've ever known, Dinesh D., routinely 'outperformed' and routinely had a very jagged eCurve and routinely was in (up to) 40% drawdowns... to talk with him you would project that he was risk averse and seeking consistency... but somehow in real life trading he was able to forego 'consistency'...

 

A properly correlated and dynamically sized 'portfolio of systems' is the penultimate 'consistency' machine...

 

hth

[M.A 10]

It is hard to undestand u or maybe I am the only one not getting u.

I have read about optimal-f mm, not in detail thoug. Even if we use 1/3 of optimal-f, curve will not change because Profit/Loss ratio doesnt depend on optimal-f. The only change will be in amounts and thus in individual percentages for profit and loss.

 

Also, with Dinesh example, are u saying that a risk averse trader routinely has 40% drawdown and he is a professional trader? Sounds weird. Isnt it a bad definition of a Pro trader?

[tradingwizzard 10]

I kind of agree with you here.....

[zdo 20]

M.A.

 

You are joining those who realize that I am probably the worst communicator ever… but for the occasional soul who does actually benefit from my communications, I wouldn’t even bother...

 

Re: “Even if we use 1/3 of optimal-f,” … that’s correct. The red runs and accomp. drawdowns will still occur. However if you run the tests on most typical systems at circa .33 optimal-f the EC will look SMOOTHER – which is one of the ways of metricking ‘consistency’ we were discussing. It is system dependent. If you bitfiddle and slide testing up and down around that .33 of optimal-f, you will find each system’s smoothest %... never will be all smooth… except for the occasional chance and temporary anomaly

 

with Dinesh example, I was briefly discussing the broader side of the question

What is "consistency"?. …and I didn’t say he was a risk averse trader. I meant that in my initial personal contacts with him my impressions belied the actual amount ‘unconsistency’ he could tolerate.

And btw, 40% drawdowns do not ‘define’ a professional trader as bad at all. (These days he wouldn't get hired as a manager... but..).

It is system dependent. For example, at the returns per instrument level (not portf level), one of the most ‘inconsistent’ methods you could ever apply is true and near true trend trading. Each trader needs to ‘study’ him or herself AND ‘study’ the available systems and then align himself to the systems most compatible to himself… the varying 'need' for ‘consistency’ is an aspect of that …we all would love ‘consistency’ in the ideal… but in actual practice some of us can tolerate more inconsistency than others can.

Anecdotally… coming up on 30 years trading and when I look back and remember the times I felt the need and tried to figure out how to be more ‘consistent’ in my trading per se that was typically just a symptom of much deeper issues… just sayin…

 

All the best,

 

zdo

[M.A 10]

  Quote

You are joining those who realize that I am probably the worst communicator ever

Nope. I do not plan on joining any "group" . But this is fact that ur writing style is a bit difficult for me to understand. Maybe due to too many abbreviations or maybe due to some other reason.

 

About 0.33 of optimal-f...maybe an example would be best because I am still confused.