You Must Play Every Hand

[tommcginnis]

There is truth there. One never knows if a trade will give an opportunity to be a winner or not.
How do you define "cherry picking"? When you think about it, your reason for entry is a "cherry pick".

It *is* true that retail traders can't expect 100% probability trades, fair enough.

But you *can* trade with greater and lesser probabilities of [desired] outcomes.
and you *can* work the expected trades for greater and lesser profit beforehand.

"Expectancy" of our trades is where our entries (and "expected" exits!!) and those now-"expected" profits meet. Positive expectancy means we have a long-term workable system -- and that is what the replies above refer to, when describing a back-tested system: these entry/exit points have been vetted, and have a positive expectancy.

Does one *know* a particular trade is a winner or not? No. But the trade *can*be* described with particular detail, in terms of loss, gain, and expectations over X-many trades. Not bad.

"Cherry-picking" is a relative term -- a loosely-defined, not-so-well-vetted system may have some trades that *barely* clear this ill-defined system, and some that *clear* qualify. The more robust the system, the less cherry-picking is even *possible*.

 

[Xela]

you *can* trade with greater and lesser probabilities of [desired] outcomes.

"Cherry-picking" is a relative term

Exactly so.

Every time I enter an individual trade, I don't know whether it will be a winner or a loser, don't pretend to, and don't need to. But with (well) over 95% statistical confidence, I can tell the collective outcome of my next 300 trades to within 3% accuracy - and that I do need to be able to do, otherwise I wouldn't be exposing my funds to risk.

 

[SimpleMeLike]

It *is* true that retail traders can't expect 100% probability trades, fair enough.

But you *can* trade with greater and lesser probabilities of [desired] outcomes.
and you *can* work the expected trades for greater and lesser profit beforehand.

"Expectancy" of our trades is where our entries (and "expected" exits!!) and those now-"expected" profits meet. Positive expectancy means we have a long-term workable system -- and that is what the replies above refer to, when describing a back-tested system: these entry/exit points have been vetted, and have a positive expectancy.

Does one *know* a particular trade is a winner or not? No. But the trade *can*be* described with particular detail, in terms of loss, gain, and expectations over X-many trades. Not bad.

"Cherry-picking" is a relative term -- a loosely-defined, not-so-well-vetted system may have some trades that *barely* clear this ill-defined system, and some that *clear* qualify. The more robust the system, the less cherry-picking is even *possible*.

Great comment.

Can you give me an example of Expectancy as it applis for long term profitability? Say for past trades or so.

Thanks

 

[SimpleMeLike]

But with (well) over 95% statistical confidence, I can tell the collective outcome of my next 300 trades to within 3% accuracy -

Good comment Xela.
Can you touch on this a bit more? I have collected data on about 300 trades?

How can I gain statistical confidence in this as well for my next 300 trades? A book reference will be appreciated if no time to explain to me.

Thank you so much.

 

[tommcginnis]

One word: Youtube.

(Often a dangerous place re trading subjects, on expectancy it's not too bad.)

 

[algofy]

Good comment Xela.
Can you touch on this a bit more? I have collected data on about 300 trades?

How can I gain statistical confidence in this as well for my next 300 trades? A book reference will be appreciated if no time to explain to me.

Thank you so much.

Law of large numbers. Look it up.

 

[Xela]

Good comment Xela.
Can you touch on this a bit more? I have collected data on about 300 trades?

How can I gain statistical confidence in this as well for my next 300 trades? A book reference will be appreciated if no time to explain to me.

Just by having a method with a pretty fixed/constant edge under whatever circumstance you trade it (and only trading it under those circumstances, of course). The closer your win-rate is to 50% (on either side of it), the smaller is the number of "samples" you need, to achieve "statistical significance".

Ralph Vince (who sometimes posts here): The Mathematics of Money Management: Risk Analysis Techniques for Traders

 

[systematictrader]

Rufus Mrehin,

Correct, if you have an idea, backtest it for X amount of trades.

But, there is soo many different variables, so keep it simple and pick a few options.

Or record every trade you take and the "what if, I had this.." scenario.

yes, thats what i boiled down to,,, after losing for so long, i boiled it down to the MACD indicator,,, i integrated extra variables, so far as 2016 made excellent percentage return on a sizeable account, iam working on backtesting further variables and more so further instruments to broaden the scope of trading since of course i cant expect what works to work forever,, the most important thing when iam backtesting, i test for my worse case scenario,,, i diversify by making sure position size is always equal, and i take every trade, regardless of what i think, and i always calculate what wouldve happened if i got caught on the wrong side of the market on the worse day on the chart for the last 20 years, what ever that is, i double it and make sure that if i suffer it iam still in the game

 

[systematictrader]

There is truth there. One never knows if a trade will give an opportunity to be a winner or not.
How do you define "cherry picking"? When you think about it, your reason for entry is a "cherry pick".

do u mind elaborating on this please? i believe i understand it as that if u have a reason of why u entered that in itself is a cherry pick,,, when ur entry should be more because the system told u to enter rather than u having a reason for it

 

[systematictrader]

It *is* true that retail traders can't expect 100% probability trades, fair enough.

But you *can* trade with greater and lesser probabilities of [desired] outcomes.
and you *can* work the expected trades for greater and lesser profit beforehand.

"Expectancy" of our trades is where our entries (and "expected" exits!!) and those now-"expected" profits meet. Positive expectancy means we have a long-term workable system -- and that is what the replies above refer to, when describing a back-tested system: these entry/exit points have been vetted, and have a positive expectancy.

Does one *know* a particular trade is a winner or not? No. But the trade *can*be* described with particular detail, in terms of loss, gain, and expectations over X-many trades. Not bad.

"Cherry-picking" is a relative term -- a loosely-defined, not-so-well-vetted system may have some trades that *barely* clear this ill-defined system, and some that *clear* qualify. The more robust the system, the less cherry-picking is even *possible*.

well worded