Odds staked against us? | Page 2 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Odds and Edge - Probabilities in Day Trading 1 reply

The odds against us (retail traders)? what can you do? 16 replies

Beat the Odds in Forex by Igor Toshchakov 2 replies

Playing the odds: Options Trading with the 'House' Advantage 0 replies

anyone read Beat the Odds in Forex Trading? 0 replies

Attachments: Odds staked against us?
Exit Attachments

Odds staked against us?

  • #21
  • May 19, 2008 3:04am May 19, 2008 3:04am
Quoting fxtrader42
........No one to this point has discussed the number one reason why the odds are not in your favor. The number one reason should be leverage!......
Leverage doesn't have to be an enemy. Simply find a broker who allows mini or micro lots, and keep your position sizes small. Just because high leverage is available doesn't mean that you're forced to use it.

Leverage has nothing to do with either expectancy or risk:
-- Risk is the percentage of your account that you're willing to lose on a trade.
-- Leverage determines the amount of money you need in your account ("margin") to be able to take the trade.
See my post here for details: http://www.forexfactory.com/showthread.php?p=1582995

In my experience, timing entries and exits accurately enough, often enough to attain profit regularly, is the most difficult part. Expectancy is the product of your entries and exits, and operates completely independently of both position size (MM), and leverage.

David
  • #22
  • May 22, 2008 8:43pm May 22, 2008 8:43pm
The number one reason the odds are against us:

It takes more percent to recover from a drawdown than the actual drawdown itself.

drawdown % : % to recover
0 : 0
10 : 11.11
20 : 25
30 : 43
40 : 67
50 : 100
60 : 150
70 : 233
80 : 400
90 : 900

You are at a disadvantage because you have a finite amount of capital. You will have to replenish your account if your drawdown is too severe, or your "huge 400% winning streak" will only get you back to breakeven.

Combine this with spreads and rollover premiums, and you've got a real bully staring you down before you even start.
  • #23
  • Jul 5, 2008 4:30pm Jul 5, 2008 4:30pm
Quoting tdion
The number one reason the odds are against us:

It takes more percent to recover from a drawdown than the actual drawdown itself.

drawdown % : % to recover
0 : 0
10 : 11.11
20 : 25
30 : 43
40 : 67
50 : 100
60 : 150
70 : 233
80 : 400
90 : 900

You are at a disadvantage because you have a finite amount of capital. You will have to replenish your account if your drawdown is too severe, or your "huge 400% winning streak" will only get you back to breakeven.

Combine this with spreads and rollover premiums, and you've got a real bully staring you down before you even start.
Nice post , this is an excellent point that is often overlooked. It's the penalty that a trader must pay for attempting to compound his equity using fixed fractional sizing (FFS).

Some considerations:

With FFS, the order of the wins and losses is unimportant, e.g. at 2% risk per trade, 1.02 x 1.02 x 0.98 equals 1.02 x 0.98 x 1.02 equals 0.98 x 1.02 x 1.02 ...... and so forth.

The problem is that (again, using 2% risk as an example) 1.02 is not the reciprocal of 0.98. In fact 1 / 0.98 = 1.020408, i.e. we have a difference of 0.000408 to contend with. Or put another way, 1.02 x 0.98 = 0.9996, leaving us short of 1 by 0.0004. So that would mean, even with a 50% win rate and 1:1 RR, and assuming no spread or swap costs, we are losing 0.04% per pair of trades, i.e. an expectancy of –0.02% per trade. (Question for geniuses: who is pocketing this money? ).

Another way of looking at this is that, if we size our positions at 2% FFS, we need an edge of +0.02% from our entries and exits just to break even.

OK, let's tabulate this, for different values at risk (assume 50% win rate, 1:1 RR, no spread or swap costs):

At 1% risk, our expectancy is –0.005% per trade.
At 2% risk, our expectancy is –0.02% per trade.
At 3% risk, our expectancy is –0.045% per trade.
At 4% risk, our expectancy is –0.08% per trade.
At 5% risk, our expectancy is –0.125% per trade.
At 10% risk, our expectancy is –0.5% per trade.

This is a quadratic function, i.e. multiply the risk by N, and the negative expectancy increases by a factor of N squared.

Another very good reason to be conservative in one's position sizing.

David
  • #24
  • Jul 5, 2008 4:40pm Jul 5, 2008 4:40pm
yes david and the way around this is to bet the same lot size (or risk, depending on your style) repeatedly.... so that if you risk 2% per trade, you do not risk 2% of the new drawdown.... you risk 2% of the highest balance

this gives you essentially 50 combined losses from new high balance to complete wipeout... you may have some winners in the course of the 50 that will give you a few "extra lives" as they say.....

Quoting hanover
The problem is that (again, using 2% risk as an example) 1.02 is not the reciprocal of 0.98. In fact 1 / 0.98 = 1.020408, i.e. we have a difference of 0.000408 to contend with. Or put another way, 1.02 x 0.98 = 0.9996, leaving us short of 1 by 0.0004. So that would mean, even with a 50% win rate and 1:1 RR, and assuming no spread or swap costs, we are losing 0.04% per pair of trades, i.e. an expectancy of –0.02% per trade. (Question for geniuses: who is pocketing this money? ).
David
  • #25
  • Jul 5, 2008 5:04pm Jul 5, 2008 5:04pm
OK, let’s look at the positives of FFS, as against betting a fixed dollar amount (FDA).

Let’s assume we have a bankroll of $100, and a 10% ($10) bet size.
Using FDA, if our first 10 trades are losses, we are bankrupt.
If our first 10 trades are wins, we’ve doubled our equity to $200.

Using FFS, after 10 losses, our equity is $100 x 0.9^10 = $34.87
After 10 wins, our equity is $100 x 1.1^10 = $259.37

In other words, we’re better off compounding with FFS, in both winning and losing scenarios, at least from this standpoint. FFS self-adjusts, making the bet sizes larger dollar-wise, as equity becomes increasingly positive, and smaller, as equity becomes increasingly negative. We can never go completely broke using FFS, at least mathematically.

With FFS, the order of the wins and losses is irrelevant. With FDA, it’s irrelevant from the standpoint of X+X–X = X–X+X etc, but if (using 10% risk as an example) the losses exceed the wins by more than 10, we encounter ruin.

All this is because FFS is a geometric progression; FDA is an arithmetic progression.

With FDA, if our equity is negative by X%, it only requires X% to recover to breakeven (unlike FFS – see my previous post).

So which method is superior? Depends on your choice of viewpoint. I suspect the vast majority of folk choose FFS, for the simple reason that they want the opportunity to compound (say) $1k into $1m.

David

[Edit: Tom, I'd already typed the above before I read your post]
  • #26
  • Jul 5, 2008 5:10pm Jul 5, 2008 5:10pm
Quoting tdion
The number one reason the odds are against us:

It takes more percent to recover from a drawdown than the actual drawdown itself.

drawdown % : % to recover
0 : 0
10 : 11.11
20 : 25
30 : 43
40 : 67
50 : 100
60 : 150
70 : 233
80 : 400
90 : 900

You are at a disadvantage because you have a finite amount of capital. You will have to replenish your account if your drawdown is too severe, or your "huge 400% winning streak" will only get you back to breakeven.

Combine this with spreads and rollover premiums, and you've got a real bully staring you down before you even start.
Hi tdion,

I agree with your post, but isn't this only true if you went long? If you went short, I think the opposite would happen.

Meaning:
drawdown % : % to recover
0 : 0
10 : 9.09
20 : 16.7
30 : 23
...... so on
  • #27
  • Jul 5, 2008 5:11pm Jul 5, 2008 5:11pm
by recalculating the 10% after a new high balance is achieved, the compounding is equal to FFS. the danger is the dependency on the system to recover from drawdown within X trades....

Quoting hanover
Let’s assume we have a bankroll of $100, and a 10% ($10) bet size.
Using FDA, if our first 10 trades are losses, we are bankrupt.
If our first 10 trades are wins, we’ve doubled our equity to $200.

Using FFS, after 10 losses, our equity is $100 x 0.9^10 = $34.87
After 10 wins, our equity is $100 x 1.1^10 = $259.37
  • #28
  • Jul 5, 2008 5:12pm Jul 5, 2008 5:12pm
sorry, this is incorrect:

Quoting soltani
Hi tdion,

I agree with your post, but isn't this only true if you went long? If you went short, I think the opposite would happen.

Meaning:
drawdown % : % to recover
0 : 0
10 : 9.09
20 : 16.7
30 : 23
...... so on
  • #29
  • Jul 5, 2008 5:26pm Jul 5, 2008 5:26pm
Quoting tdion
yes david and the way around this is to bet the same lot size (or risk, depending on your style) repeatedly.... so that if you risk 2% per trade, you do not risk 2% of the new drawdown.... you risk 2% of the highest balance

this gives you essentially 50 combined losses from new high balance to complete wipeout... you may have some winners in the course of the 50 that will give you a few "extra lives" as they say.....
OK, I will test this. I'll create an XLS that plots equity over 6 trades (since there are only 2^6 = 64 possible permutations of wins and losses over 6 trades, that's manageable enough), and see if the average result exceeds that obtained by using "flat" betting. This will of course assume that the probability of a win and a loss is equally likely at any point, which will ensure that the MM is being tested independently of expectancy. I think it's easy enough to prove, by induction, that if the end result is the same for 4 trades, 5 trades, and 6 trades, that it will likeiwse hold good for 7 trades, 8 trades, ..... i.e. any number of trades.

We'll try using both FDA (lot-size based) and FFS (risk based) approaches to ensure that all bases are covered.

Give me a few hours and I'll post the XLS.

David
  • #30
  • Jul 6, 2008 4:34am Jul 6, 2008 4:34am
this was a joke to prove my thoughts that you people believe you can not do it, I do quite well as a trader. Maybe at the end of the day you guys need to look into the mirror.
  • #31
  • Edited 5:01am Jul 6, 2008 4:45am | Edited 5:01am
Attached is the promised XLS, which I believe further confirms that MM in itself can not improve overall expectancy.

The final P/L (expectancy across all possible permutations) is the value in cell W11, and should result in a value of zero, regardless of the parameters entered.

-- To implement "flat" betting (i.e. consistent position size) using fixed dollar amount (FDA, i.e. arithmetic growth), set up the parameters as follows:
$10,000; 2%; 1; 0%; 0
(multiplying bet size by 1 and adding 0 equates to consistent bet size)

-- To implement "flat" betting (i.e. consistent position size) using fixed fractional sizing (FFS, i.e. geometric growth), set up the parameters as follows:
$10,000; 2%; 1; 0%; 1

-- To implement a betting system where the position size is increased following a new equity high, change the "multiply bet size" and/or the "and then add" parameters.
For example, if you want to double the bet size every time equity reaches a new high, change the multiplier to 2. Or choose a less severe option by entering something like 1.2
Or, if you want to increase the bet size linearly, leave the multiplier at 1 and enter a positive value in the "and then add" parameter, e.g. an 0.5 increment would make the bet size increase thus: 2%, 2.5%, 3%, 3.5%, etc as each new equity high is reached.

-- You can likewise toggle between FDA and FFS for any non-flat betting method by changing the final parameter between 0 and 1, respectively.

-- Also, feel welcome to change the initial equity, and the starting bet size, to anything you wish.

-- Also, feel welcome to decrease bet size after a new equity high, if you wish, by entering a multiplier < 1 and/or an increment value < 0.

-- Extend the spreadsheet to 128 permutations to test 7 trades, or 256 permutations to test 8 trades, if you wish.

Convince yourself once and for all that, no matter how many trades, and no matter what betting system you use, the expectancy is always zero. Check what is happening to the component expectancies across each individual row (column W), and you will gain an understanding as to why this is so.

Conclusion: MM is irrelevant!

(The only exception is if you can somehow have your larger bets coincide with higher expectancy trades, and vice versa).

The idea that MM can provide an edge is one of the biggest trading myths.
I don't care if the best selling authors disagree: they are ALL WRONG.
I challenge anybody to post a legitimate mathematical proof that proves otherwise.

David
Attached File(s)
File Type: xls Test Tdion MM theory.xls   106 KB | 552 downloads
  • #32
  • Jul 6, 2008 4:59am Jul 6, 2008 4:59am
Quoting NoBread
this was a joke to prove my thoughts that you people believe you can not do it, I do quite well as a trader. Maybe at the end of the day you guys need to look into the mirror.
Here we go again........

___________________________________________________

Anyway........

Here's another XLS that allows you to experiment with even more types of betting system, again over 6 trades.

The result remains the same: MM can't improve expectancy, no matter what you do. It's a mathematical impossibility.

David
Attached File(s)
File Type: xls Martingale comparison testing.xls   157 KB | 465 downloads
  • #33
  • Jul 6, 2008 6:17am Jul 6, 2008 6:17am
Top 10 Mistakes Traders Make

Achieving success in futures trading requires avoiding numerous pitfalls as much, or more, than it does seeking out and executing winning trades. In fact, most professional traders will tell you that it's not any specific trading methodologies that make traders successful, but instead it's the overall rules to which those traders strictly adhere that keep them "in the game" long enough to achieve success.
Following are 10 of the more prevalent mistakes I believe traders make in futures trading. This list is in no particular order of importance.
1. Failure to have a trading plan in place before a trade is executed. A trader with no specific plan of action in place upon entry into a futures trade does not know, among other things, when or where he or she will exit the trade, or about how much money may be made or lost. Traders with no pre-determined trading plan are flying by the seat of their pants, and that's usually a recipe for a "crash and burn."
2. Inadequate trading assets or improper money management. It does not take a fortune to trade futures markets with success. Traders with less than $5,000 in their trading accounts can and do trade futures successfully. And, traders with $50,000 or more in their trading accounts can and do lose it all in a heartbeat. Part of trading success boils down to proper money management and not gunning for those highly risky "home-run" type trades that involve too much trading capital at one time.
3. Expectations that are too high, too soon. Beginning futures traders that expect to quit their "day job" and make a good living trading futures in their first few years of trading are usually disappointed. You don't become a successful doctor or lawyer or business owner in the first couple years of the practice. It takes hard work and perseverance to achieve success in any field of endeavor--and trading futures is no different. Futures trading is not the easy, "get-rich-quick" scheme that a few unsavory characters make it out to be.
4. Failure to use protective stops. Using protective buy stops or sell stops upon entering a trade provide a trader with a good idea of about how much money he or she is risking on that particular trade, should it turn out to be a loser. Protective stops are a good money-management tool, but are not perfect. There are no perfect money-management tools in futures trading.
5. Lack of "patience" and "discipline." While these two virtues are over-worked and very often mentioned when determining what unsuccessful traders lack, not many will argue with their merits. Indeed. Don't trade just for the sake of trading or just because you haven't traded for a while. Let those very good trading "set-ups" come to you, and then act upon them in a prudent way. The market will do what the market wants to do--and nobody can force the market's hand.
6. Trading against the trend--or trying to pick tops and bottoms in markets. It's human nature to want to buy low and sell high (or sell high and buy low for short-side traders). Unfortunately, that's not at all a proven means of making profits in futures trading. Top pickers and bottom-pickers usually are trading against the trend, which is a major mistake.
7. Letting losing positions ride too long. Most successful traders will not sit on a losing position very long at all. They'll set a tight protective stop, and if it's hit they'll take their losses (usually minimal) and then move on to the next potential trading set up. Traders who sit on a losing trade, "hoping" that the market will soon turn around in their favor, are usually doomed.
8. "Over-trading." Trading too many markets at one time is a mistake--especially if you are racking up losses. If trading losses are piling up, it's time to cut back on trading, even though there is the temptation to make more trades to recover the recently lost trading assets. It takes keen focus and concentration to be a successful futures trader. Having "too many irons in the fire" at one time is a mistake.
9. Failure to accept complete responsibility for your own actions. When you have a losing trade or are in a losing streak, don't blame your broker or someone else. You are the one who is responsible for your own success or failure in trading. You make the trading decisions. If you feel you are not in firm control of your own trading, then why do you feel that way? You should make immediate changes that put you in firm control of your own trading destiny.
10. Not getting a bigger-picture perspective on a market. One can look at a daily bar chart and get a shorter-term perspective on a market trend. But a look at the longer-term weekly or monthly chart for that same market can reveal a completely different perspective. It is prudent to examine longer-term charts, for that bigger-picture perspective, when contemplating a trade.
  • #34
  • Jul 6, 2008 1:40pm Jul 6, 2008 1:40pm
Great test Hanover !

This shows that no only MM you also need a good strategy.
  • #35
  • Edited 2:33pm Jul 6, 2008 2:15pm | Edited 2:33pm
david,

i admire your hard work.... and yes i disagree (is that really a surprise lol?)

i never argued that green pips aren't the top priority.... but money management can make your green pips go further.... ie) the same green pips that are pulling you out of drawdown with %R could be pushing new high balances w/ what i described (the price of admission: drawdown is inherently bigger than w/ standard %R)

your spreadsheet does not reproduce my mm. you are just looking at this from a different perspective than i am. it is amusing to see how we (the ff community) can attack at the same things with different approaches....

-tom

the question is, whether i am too lazy to create my own spreadsheet.... do i win a prize if i prove you wrong?
  • #36
  • Jul 6, 2008 3:03pm Jul 6, 2008 3:03pm
well the statistics reads that 5-10% only make it succesful in this biz. this is real and these set of people exist. this biz may seem so difficult. well if u ar not part of these 5-10% of traders who make money from d forex world on constant basis, then you just need to learn some few or more things, imbibe some few or more characteristics which you must have been hearing or reading. i can guess you are a new trader, following your time of registration on forexfactory or maybe u hav spent up to a year in d forex world, whatever way have it in mind that you can be succesful in this biz, u can make money on constant basis, its all about getting the right information, imbibing the right qualities which the market demands from u as a trader.
know thy edge and abide by it ! 4 in it shall thou be successful !
  • #37
  • Jul 6, 2008 3:04pm Jul 6, 2008 3:04pm
Quoting NoBread
Top 10 Mistakes Traders Make....
Thank you, Mr Wyckoff, for taking the time to post in the FF forum.

http://www.actionforex.com/articles-...2008012835796/

David
  • #38
  • Edited 3:57pm Jul 6, 2008 3:37pm | Edited 3:57pm
Quoting tdion
david,

i admire your hard work.... and yes i disagree (is that really a surprise lol?)

i never argued that green pips aren't the top priority.... but money management can make your green pips go further.... ie) the same green pips that are pulling you out of drawdown with %R could be pushing new high balances w/ what i described (the price of admission: drawdown is inherently bigger than w/ standard %R)
Look at each row of the XLS (the one attached to post # 31), and you'll see that wherever the MM improves expectancy on one permutation of wins and losses, it diminishes it on other row(s). By following the math, you can see exactly how. The end result is always zero. Unless you can somehow predict in advance whether the next trade is more likely to be a win or a loss, any one of the 64 permutations is equally probable.

your spreadsheet does not reproduce my mm.
Sorry if I misunderstood. I thought (from your comment in post #24) that you were increasing position size whenever a new equity high is reached, and retaining the current position size otherwise. If that's so, then the XLS reproduces that. If not, please explain how your MM differs, and I'll make the necessary changes.

you are just looking at this from a different perspective than i am.
In this instance I'm simply looking at every possible permutation of wins and losses, and applying an expectancy formula. For me, math and statistics are king.

it is amusing to see how we (the ff community) can attack at the same things with different approaches....

-tom

the question is, whether i am too lazy to create my own spreadsheet.... do i win a prize if i prove you wrong?
Your prize is that you will likely gain a better understanding of MM (I don't mean that condescendingly, we all learn from our own experiments), and, if you post your results here, you'll gain whatever kudos the community chooses to give you (for whatever that's worth - LOL), and know that you've helped everybody improve their own profitability. And I will be extremely grateful for being proved wrong, because my profitability will be improved also, and it will give me a new approach, or angle, to develop further.
.
  • #39
  • Jul 6, 2008 3:46pm Jul 6, 2008 3:46pm
Quoting melpheos
Great test Hanover !

This shows that no only MM you also need a good strategy.
Melpheos, thanks for the feedback.

I likewise believe that ultimately it's only by timing entries and exits accurately enough to overcome costs, on balance, that one can win at trading.

It goes without saying that if one doesn't have the discipline to implement one's method consistently, the method is effectively worthless.

The smaller we size our positions, the longer our equity will survive. Varying MM simply alters risk and return in exactly like proportion. MM in itself can not provide an edge.

David
  • #40
  • Edited 5:18pm Jul 6, 2008 4:54pm | Edited 5:18pm
p27 on your spreadsheet is where we diverge

you reached a new high balance (10,200) at o27.... so the next bet should have been (0.02 * 10200), and the win would have put at 10404.

it is a very minute point, but it is what it is.


okay so the idea (w/ a 2% risk) is to divide the highest balance you've attained by 50, and risk exactly that figure on each trade until a new high balance is achieved.

so after losing 25 times in a row, your account would be drawn down 50%.... wheras with %R, it would be 0.98^25.... which a 40% drawdown. Note that my method will draw down faster, but WILL RECOVER FASTER.

it is all based on your expectancy to make consistent green pips. You lose 50 times in a row (or any other stretch where your losers outweigh your winners by 50.... example) 60 losses and 10 wins over 70 trades) you bankrupt the account....

while with %R, it is impossible to ever completely bankrupt the account.

i have simply proposed a solution that is less risky than martigale, and eliminates the "make more to break even" problem.

EDIT) To further exemplify: in the example where you lost 25 trades in a row, you will be back to break even after 25 wins with my method. BUT.... with %R you will only recover back to 98%.....
0 traders viewing now