Coin tossing and random entries | Page 2 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

50% win ratio, 1:1 sl tp, and random entries 127 replies

The Art of Coin Tossing 45 replies

Pure Risk to reward trade System - 1 to 3- Totally random entries 178 replies

Expert Advisor with random entries 3 replies

Random entries discussion... 7 replies

Attachments: Coin tossing and random entries
Exit Attachments

Coin tossing and random entries

  • #21
  • Dec 28, 2009 4:58pm Dec 28, 2009 4:58pm
Quoting hanover
How well does it pass the 7-point plan?
How about adding point #8 for confirmation: Flip a coin.

Just kidding!

Wanted to thank you for the link to the spaghetti indicator.
Comparing relative strength between currencies?
That's what I call better than coins, darts, astrology or even
my magic eight ball!
  • #22
  • Dec 31, 2009 10:27pm Dec 31, 2009 10:27pm
Quoting hanover
It's possible that our concept of black swans differs. The only book of Taleb's that I've read is Fooled By Randomness, and it was a long time ago. I see a black swan as something so unlikely that it defies any kind of testing, and wouldn't be worth tampering with my setups to try to factor it in. If I miss an out-of-the-blue, once-in-a-decade move, then so be it. But of course no two people approach trading in exactly the same way.

I use at least three different entry setups that give me a significant positive bias. I can't say conclusively...
We are on the same wavelength concerning BLACK SWANS. But if you look at gbpusd price action this week, you may understand. If on Tuesday, you went short, there was a nice profit. If on Wednesday, you went short, you get stopped out. Now using a coin toss we have 4 possible out comes, HH, HT, TH, TT. Heads go short, tails do nothing.

HH, gives you a big profit and a small loss

HT, gives you a big profit and no entry

TH, gives you no entry and a small loss

TT, gives you no entry and no entry

The worse case is a small loss.

Attached Image
  • #23
  • Jan 1, 2010 4:03am Jan 1, 2010 4:03am
Quoting steveshelby
We are on the same wavelength concerning BLACK SWANS....
You haven't specifically stated where you enter or exit, so I've assumed that you enter at the day's open, and exit (if not stopped out) at the day's close.

Depending on your SL placement, you could have been stopped out on Tuesday also, and thus incurred losses on both days.

IMHO, the flaw in your logic is that you're ascribing a 50% probability to a 'small loss' or a 'big profit'. That is not necessarily so. Everything else being equal, RR and win rate operate in inverse proportion to each other. Trying to cut losses and let profits run will lower win rate commensurately.
  • #24
  • Jan 1, 2010 9:26pm Jan 1, 2010 9:26pm
Quoting hanover
You haven't specifically stated where you enter or exit, so I've assumed that you enter at the day's open, and exit (if not stopped out) at the day's close.

Depending on your SL placement, you could have been stopped out on Tuesday also, and thus incurred losses on both days.

IMHO, the flaw in your logic is that you're ascribing a 50% probability to a 'small loss' or a 'big profit'. That is not necessarily so. Everything else being equal, RR and win rate operate in inverse proportion to each other. Trying to cut losses and let profits run will...
I am not ascribing any probability. If price moves in my favor, more often than not, based on historical data, it will move 50 pips or more when trading the major pairs. Many times it is over 100 pips. To me, that's a big win. Cutting losses quickly gives the ability to try the trade again and again, if an entry is triggered. I am more concerned with expectancy than with risk/reward. The average loss size is under trader control. The rest is up to the market.

Let's stay I got stopped out once, twice or even three times, the final winning trade more than makes up for the losses.
  • #25
  • Jan 3, 2010 6:53am Jan 3, 2010 6:53am
Quoting steveshelby
I am not ascribing any probability....
Apparently I misunderstood. Your mention of a coin toss (HT, TT, etc) led me to believe you were assigning a 50% probability to each possible outcome.

If the price moves 50+ pips then fine, so long as it moves the 50 pips before it hits your SL.

Anyway, no matter, I'm not trying to prove you wrong. The original topic of the thread was coin toss entries, and that was the direction of my discussion. If you have a profitable trading method, then nobody is in a position to criticize, least of all me.

At the end of the day, profit is the only real goal. Given that each of us starts with a blank canvas, there are doubtless several different ways of achieving this. In that sense, trading is a subjective process, and nobody is necessarily right or wrong. If I may speak generally, one of the tendencies in this forum is for folk to believe that their own way is the only profitable way to trade. Which is nonsense, of course. I hope that I haven't given that impression.
  • #26
  • Jan 3, 2010 8:03am Jan 3, 2010 8:03am
Random coin-toss entries have an academic interest and are a valuable learning exercise. However, I agree with the assessment that actually using such an entry is in fact "giving up".

Why give up when there are very simple entry techniques with a proven statistical edge (i.e. hit rate > 50%)? These edges are everywhere, you just have to find them.

For instance, if price is above some moving average then you have a statistical edge by only taking long entries. If you buy at the open of every 4 hour bar that's above the EMA 20 and close the trade at the bar's close, you have a 60% chance of a winning trade. It's as simple as that.
You are in a maze of twisty little passages, all alike.
  • #27
  • Jan 3, 2010 11:12am Jan 3, 2010 11:12am
Quoting steveshelby
I am not ascribing any probability. If price moves in my favor, more often than not, based on historical data, it will move 50 pips or more when trading the major pairs.
More comedy gold from the forum that keeps on giving.
The breaking of a wave cannot explain the whole sea.
  • #28
  • Jan 3, 2010 11:27am Jan 3, 2010 11:27am
hi,

i want to share my opinion about 50/50 or coin toss. having 50/50 win/loss doesn't mean we are breakeven. there is spread that affect our overall p/l figure.
for example, if we use random entry and close order after price move exactly 10 pips in either direction, at spread = 2, we get profit = 8 or loss = -12.
if total trade = 100, 50/50 means profit = 40, loss = -60
i did simple math analyzes and came to this formula:

win percentage needed to be breakeven = 50*(spread/distance+1), result in percentage.
distance is the required distance for price to move to hit a fixed tp/sl
for spread = 2 pips, and distance = 10 pips, we need 60% win / 40% loss to be breakeven
  • #29
  • Jan 3, 2010 11:34am Jan 3, 2010 11:34am
Quoting hanover
At the end of the day, profit is the only real goal. Given that each of us starts with a blank canvas, there are doubtless several different ways of achieving this. In that sense, trading is a subjective process, and nobody is necessarily right or wrong. If I may speak generally, one of the tendencies in this forum is for folk to believe that their own way is the only profitable way to trade. Which is nonsense, of course. I hope that I haven't given that impression.
Not at all hanover. You are one of the few individuals that I have debated that hasn't stooped to name calling. Profit is the only goal of trading and if I learn a thing or 2 from you or anyone else along the way, that will just make it a bit easier. If you are profitable, I am sure you make your profit using a different method than I do and that's how it should be. Each trader finding what works for them. Unfortunately, when one shares on a public forum, there are those who would rather tear down the method and flame the poster than learn. It is always amusing to me when a Johnny Come Lately tells me they back-tested my method and it doesn't work even though I have been using it profitably for years.
  • #30
  • Jan 3, 2010 11:36am Jan 3, 2010 11:36am
Quoting stevegee58
For instance, if price is above some moving average then you have a statistical edge by only taking long entries. If you buy at the open of every 4 hour bar that's above the EMA 20 and close the trade at the bar's close, you have a 60% chance of a winning trade. It's as simple as that.
Can you show that? I mean, can you show this with EA printout or by other means?
  • #31
  • Edited 11:57am Jan 3, 2010 11:46am | Edited 11:57am
Quoting stevegee58
Random coin-toss entries have an academic interest and are a valuable learning exercise....
i just made an ea to test your theory. attaced pics are a balance curve of trade buy above ema 20 /sell below ema 20 and the reverse logic (buy below ema, sell above ema). the first strats produce 45% profit trade, the second produce... not 55%... but 46%. that caused by spread..
Attached Image(s) (click to enlarge)
Click to Enlarge Name: ema20_cntr_trend.gif Size: 6 KB
Click to Enlarge Name: ema20_trend2.gif Size: 6 KB
  • #32
  • Jan 3, 2010 11:51am Jan 3, 2010 11:51am
Quoting steveshelby
Johnny Come Lately tells me they back-tested my method and it doesn't work even though I have been using it profitably for years.
http://www.forexfactory.com/showthread.php?t=213145
Quote: "After studying for most of 2009, I am about to embark on real live trading."
I'm sure it only gets more complicated from here...
The breaking of a wave cannot explain the whole sea.
  • #33
  • Jan 3, 2010 11:54am Jan 3, 2010 11:54am
Quoting Craig
http://www.forexfactory.com/showthread.php?t=213145
Quote: "After studying for most of 2009, I am about to embark on real live trading."
You made an assumption. That quote is for forex trading. It doesn't imply I have not traded other instruments. The following may have been clearer but since this is Forex Factory I made an assumption:

"After studying FOREX for most of 2009, I am about to embark on real live FOREX trading."
  • #34
  • Jan 3, 2010 11:57am Jan 3, 2010 11:57am
Quoting sangmane
i just made an ea to test your theory. attaced pics are a balance curve of trade buy above ema 20 /sell below ema 20 and the reverse logic (buy below ema, sell above ema). the first strats produce 45% profit trade, the second produce... not 55%... but 46%. that caused by spread..
Quote
If you buy at the open of every 4 hour bar that's above the EMA 20 and close the trade at the bar's close, you have a 60% chance of a winning trade.

That did not say to buy above / sell below.

Buy at bar open if bar opens above the ema(20) and sell at bar's close. Could you code that EA?
  • #35
  • Jan 3, 2010 12:00pm Jan 3, 2010 12:00pm
hmmm, not sure if i interprets your rules correctly. but this is ea that i made to test
Attached File(s)
File Type: mq4 ema20 EA.mq4   < 1 KB | 463 downloads
  • #36
  • Jan 3, 2010 12:34pm Jan 3, 2010 12:34pm
Quoting sangmane
hmmm, not sure if i interprets your rules correctly. but this is ea that i made to test
Nope, you did it wrong. Buy at the new bar's open and sell at that bar's close above the EMA 20. Sell at the new bar's open and buy at that bar's close below the EMA 20. No stop loss, no take profit.

But that wasn't the point of my comments anyway. I didn't propose a trading system. I was pointing out a statistical behavior that could be incorporated into a trading system.
You are in a maze of twisty little passages, all alike.
  • #37
  • Jan 3, 2010 12:55pm Jan 3, 2010 12:55pm
well, i based on this rule:
If you buy at the open of every 4 hour bar that's above the EMA 20 and close the trade at the bar's close, you have a 60% chance of a winning trade.

what ea did is at the open of a new bar, it check if price (in this case, the open price) above ema20 => buy, reverse for sell.
then at the next new bar, which is, imo, the only indication that the previous bars has been closed, close the order.
  • #38
  • Jan 4, 2010 9:43am Jan 4, 2010 9:43am
Quoting sangmane
hmmm, not sure if i interprets your rules correctly. but this is ea that i made to test
Hey sangmane, I said before that you did it wrong but I looked more closely at your EA and it's correct LOL. It's impossible to concentrate on anything when there's a 10 week old puppy running around.

My point about the statistical edge is still true. 60% of the bars above the EMA 20 close higher as well as the bars below the EMA 20 close lower. I'd post my Excel spreadsheet but it's probably too big (4 megs). Is it a trade-able system by itself? Clearly no.
You are in a maze of twisty little passages, all alike.
  • #39
  • Jan 5, 2010 6:47pm Jan 5, 2010 6:47pm
Quoting stevegee58
Nope, you did it wrong. Buy at the new bar's open and sell at that bar's close above the EMA 20. Sell at the new bar's open and buy at that bar's close below the EMA 20. No stop loss, no take profit.

But that wasn't the point of my comments anyway. I didn't propose a trading system. I was pointing out a statistical behavior that could be incorporated into a trading system.
http://i773.photobucket.com/albums/y...550/audusd.gif

Looks like a winner.
  • #40
  • Jan 6, 2010 6:22pm Jan 6, 2010 6:22pm
Hanover,

I think you know my feeling about predicting short term market moves... they are well laid out in one of my previous threads. What it boils down to is this: "When you place a trade, what evidence, assurance, or hope do you have that you will recover the spread and be able to close in profit?" The answer usually is, "I believe the market will do this, or I hope the market will do that." That is all we have. No one has a quantifiable measuarble edge... at least no one that is a small, retail daytrader. So go ahead and flip your coin. That method is a good as any and better than most. In the end, you have entry, exit, money management, and market volitility. These are your tools. Forget trying to predict what today's close will be, or if the price will be higher or lower in the next 5 minutes, because it just cannot be done.

p.s. I know this makes a lot of people mad, so please just pm me with all of your rants about how stupid I am and how simple it is to make money daytrading forex. Yes, I've heard it all before, but you may feel better venting to someone. Please do not hijak Hanover's thread.

regards,
0 traders viewing now