spanish89...

[zupcon]

It will only be 50/50 if you detrend it, otherwise it will be very close to 50/50, no?

There is an argument for deterending data, although its a practice I tend not to employ, and yes, a slight bias will be seen due to trends in longer timeframes (but its not that great a bias) and of course if your taking a sample of long AND short trades the 60/40 bias in the trend direction is offset by the 40/60 countertrend so over the sample itstill comes out at 50/50 (or as you say very close tending to oscillate around this level)

 

[new_trader]

This is just getting silly now, it IS RANDOM !

This is what you need to do, take 100 trades each and every day across a variety of timeframes and instruments for the next year, NO stop loss and NO targets. Leave em to run for 50, or a 100, or 200 bars in your chosen timeframe, then close em.

After youve collected a few hundred for each timeframe, determine the win / lose rate then come back and tell me it aint 50% (+/- an element of noise which will be a function of the square root of the number of trades in the sample)

The moment you start timing entries, or closing trades due to a stop are target is the time the probability deviates from 0.5. (The probability of course is a bit irrelevant) but in the absense of fiddling about, over a sufficient sample size the probability of an up or down move from a random entry in any selected timeframe is 50/50, and 5 minutes backtesting will confirm that... spending a couple of years actually watching price move in real time will confirm the same, albeit far more educational and useful

Zupcon,

You seem far more knowledgeable on statistics and probability than me, but let me see if I understand this correctly:

If I went around the world and selected at random a tennis player just before a match starts, and I did this for every game being played around the world at any moment in time, then my success rate for choosing the winner should be around 50%. Correct?

Would this then mean that if a 5 year old child was to play against Serena Williams, the 5 year old child would have a 50% chance of winning?

Would you bet money on the 5 year old?

 

[rathcoole_exile]

He has been asking me about the interviews and for tips, seems like he's got his heart set on applying.

if he does get the job there, i publicly promise here and now to cease and retire from trading for good.

 

[Skill Leverage]

Zupcon,

You seem far more knowledgeable on statistics and probability than me, but let me see if I understand this correctly:

If I went around the world and selected at random a tennis player just before a match starts, and I did this for every game being played around the world at any moment in time, then my success rate for choosing the winner should be around 50%. Correct?

Would this then mean that if a 5 year old child was to play against Serena Williams, the 5 year old child would have a 50% chance of winning?

Would you bet money on the 5 year old?

This isn't what he is saying at all.

 

[Hoggums]

It'll be all OUR fault when he doesn't get it. We publicly tarnished his clean wholesome image.

 

[poborsky]

Spanish 89 seems to have made some bad trading decisions resulting in losses. Sounds familiar!!!

 

[JRP2891]

I also read on his thread a while ago that he was planning to use the thread as something to present to prop firms as to why they should employ him. Spanish i wouldn't do that if i were you, you might be able to talk your way in if you hold back on some of your bull, but if you let them see your thread they're going to shut the door in your face.

 

[Skill Leverage]

new_trader if you were looking for a little more than that I'll give you a hint; the probability of a 5 year-old returning a Serena Williams serve isn't the same as the other way around.

Also the key to Zupcon's statement is RANDOM.

 

[Mr. Charts]

if he does get the job there, i publicly promise here and now to cease and retire from trading for good.

Bit risky that, as he may con them into giving him a job. OK I know they are not stupid, but who knows what deceptions he comes up with............just look at his record !
Richard

 

[KillPhil08]

I hope he gets the job.

 

[Claudia123]

Change the record

lol you don't like my music ???

 

[Skill Leverage]

lol you don't like my music ???

No he doesn't like your chat

 

[arabianights]

Bit risky that, as he may con them into giving him a job. OK I know they are not stupid, but who knows what deceptions he comes up with............just look at his record !
Richard

Futex are well aware of who he is and his track record I should have thought... but I would still think he's potentially worth a punt, you'd need to see how he would cope under good risk management.

 

[Claudia123]

No he doesn't like your chat

PMSL then why does he talk to me??? Does any one here have a sense of humour???? Guess not. Oh well bye bye

 

[Skill Leverage]

Plenty of people do. FYI if all someone ever said to me was 'change the record' I wouldn't consider us BFFs.

 

[zupcon]

If I went around the world and selected at random a tennis player just before a match starts, and I did this for every game being played around the world at any moment in time, then my success rate for choosing the winner should be around 50%. Correct?

Over a large enough sample size, Yes, 50%

Would this then mean that if a 5 year old child was to play against Serena Williams, the 5 year old child would have a 50% chance of winning?

No not at all, the odds would be stacked very firmly with Serena. This is the point that I'm trying to make, the 50/50 probability starts to get skewed when you start meddling with the selection process. For example, if you where knowledgable about tennis, and picked players based on form, you'll hopefully achieve a higer win rate. Same thing goes for trading, experienced trades achieve better than average performance, non experienced achieve worse than average performance (which is the interesting point, as they'd do far better by submitting to random chance). If your knowledge of tennis was non existant, sometime's you'd be "lucky" enough to pick Serena, sometimes you'd pick the 5 year old, its irelevant, you'd still get the 50% track record

Would you bet money on the 5 year old?

No I wouldnt, but only because Im aware of Ms Williams prowess on the courts. My position on this is that markets are NOT random. However if you enter and exit the market on a completely random basis (which many do), it should not come as a surprise to get a random distribution of gains and losses, and back or forward testing will very easily prove that this is exactly the case.

Apologies to S89 fans for going a bit off topic,

 

[new_trader]

Plenty of people do. FYI if all someone ever said to me was 'change the record' I wouldn't consider us BFFs.

Most would consider them as old. Surely the saying today is change the MP3.

 

[Claudia123]

Most would consider them as old. Surely the saying today is change the MP3.

Thank you new_trader - 50/50 chance I put the wrong MP3 on though

 

[Skill Leverage]

Someone shoot me.

 

[Skill Leverage]

I would say 'there's a 50/50 chance it did', but I have a sense of humour.