Do you use Martingale position sizing?
- Thread starter Sponger
- Start date
i myself believe in scaling in losers, and often partially scaling out at breakeven or near to keep leverage under control and manage entry price on trades.
ie:
1. buy 100 stock at 100
2. buy 100 stock at 90
3. sell 100 stock on hesitant recurse to 95
4. buy 100 stock at 85
5. sell 100 stock at 90
(avg is now 90 instead of 100 on 100 shares)
6. stock takes off to 110. take profits.
very simplified, but works on scalping model as well.
i don't buy whitster's commodity argument versus stocks concerning value. try doing that with natural gas. sure it won't go to 0, but implied minimum leverage on commodities (or all futures) versus stocks invalidates the argument. you'll lose a lot more scaling in a losing position on corn from 4.00 to 2.80 than trading a perfectly sized stock position that has decent fundamentals.
of course, you have to make sure you are trading stocks that have solid fundamentals... some don't, and for those, thats a different game. the pure TA momentum game might be better to play with tighter stops and not too much scaling.
i wouldn't want to be around a major earnings disappointment on a high PE stock with no cash (or low book, and high debt) and try the scaling strategy. when valuations are in stratosphere, safety and 'value' is likely too far from entry price.
ie:
1. buy 100 stock at 100
2. buy 100 stock at 90
3. sell 100 stock on hesitant recurse to 95
4. buy 100 stock at 85
5. sell 100 stock at 90
(avg is now 90 instead of 100 on 100 shares)
6. stock takes off to 110. take profits.
very simplified, but works on scalping model as well.
i don't buy whitster's commodity argument versus stocks concerning value. try doing that with natural gas. sure it won't go to 0, but implied minimum leverage on commodities (or all futures) versus stocks invalidates the argument. you'll lose a lot more scaling in a losing position on corn from 4.00 to 2.80 than trading a perfectly sized stock position that has decent fundamentals.
of course, you have to make sure you are trading stocks that have solid fundamentals... some don't, and for those, thats a different game. the pure TA momentum game might be better to play with tighter stops and not too much scaling.
i wouldn't want to be around a major earnings disappointment on a high PE stock with no cash (or low book, and high debt) and try the scaling strategy. when valuations are in stratosphere, safety and 'value' is likely too far from entry price.
Also important, not many realize this, the profit you gain with this method is nowhere near to possible drawdowns.
Suppose you bet $10 at roulette RED. If it is RED you will have won $10. But if it is BLACK BLACK BLACK BLACK BLACK BLACK BLACK BLACK and then RED, you will have won.... $10. In the meantime you last bet would have been $1280 (if i counted correctly) and you would already have lost $10+$20+$40+$80+$160+$320+$640=$1270.
Of course trading has no finite bets, it is a continuous price-movement and we ourselves determine when a bet starts and ends.
Still, buying 1000 @ 100, then 2000 @ 50 gives you the joyful feeling of having payed only $66 per share, but of course the price is at $50 so you loss is already 3000*16=48.000, of the 200.000 at stake. Go figure.
Ursa..
Lets imagine a $1000 hypothetical bankroll managed according to some sizing strategies, with an initial risk of: $10 (1% risk, which is conservative).
1. Martingale
risks 10, 20, 40, 80, 160, 320, 640 ->BUST
it lasted 7 rounds
winnings are fixed $10 amounts
2. Fixed Amount
risks 10, 10, 10... ->BUST
it lasted 100 rounds
winnings are fixed $10 amounts
3. Fixed Percentage
risks 10, 9.9...8...7...
it never ends
winnings are 1% of bankroll, so it grows geometrically.
4. Kelly and other adjustable sizing strategies
risks 10...9...8...7...
it never ends
winnings are hyper geometrical since the position sizing adapts its size to the historical track record which effectively puts more weight when the equity curve is raising.
5. Scaling in and out based on adjustable sizing strategies.
risks 10...9...8...7...
it never ends
winnings are hyper geometrical and in addition there's an higher W% at the expense of a smaller W amount (which is typical result of entries and exits being averaged). Since the sizing strategy builds into a winning track record and this method increases the W%, the equity curve slope will increase as the system maintains itself more time in the higher range of sizing.
Most pro traders I know use some sort of methodology that embodies the principles of no.5. There is no need calling this or that or writing a book about it. IMO it is written on the DNA of successful traders and comes out in all sort of flavors. I still waiting to see a pro trader telling me that he martingales....
1. Martingale
risks 10, 20, 40, 80, 160, 320, 640 ->BUST
it lasted 7 rounds
winnings are fixed $10 amounts
2. Fixed Amount
risks 10, 10, 10... ->BUST
it lasted 100 rounds
winnings are fixed $10 amounts
3. Fixed Percentage
risks 10, 9.9...8...7...
it never ends
winnings are 1% of bankroll, so it grows geometrically.
4. Kelly and other adjustable sizing strategies
risks 10...9...8...7...
it never ends
winnings are hyper geometrical since the position sizing adapts its size to the historical track record which effectively puts more weight when the equity curve is raising.
5. Scaling in and out based on adjustable sizing strategies.
risks 10...9...8...7...
it never ends
winnings are hyper geometrical and in addition there's an higher W% at the expense of a smaller W amount (which is typical result of entries and exits being averaged). Since the sizing strategy builds into a winning track record and this method increases the W%, the equity curve slope will increase as the system maintains itself more time in the higher range of sizing.
Most pro traders I know use some sort of methodology that embodies the principles of no.5. There is no need calling this or that or writing a book about it. IMO it is written on the DNA of successful traders and comes out in all sort of flavors. I still waiting to see a pro trader telling me that he martingales....
I was sitting in a pub with someone at work and his theory was instead of doubling you triple your betsize instead. This way you win about the equivalent of the drawdown instead of just your original bet.
He was using roulette as an example.
He couldnt understand the fact that roulette is a negative expectancy game regardless of the betsize.
People assume that they have an infinite bankroll, when in reality that is nonsense. So even with a postive expecation game, if your bet size is too big then even a small run of losses will wipe you out.
This is why casinos set maximum position sizes on punters.
Even though the casino has the edge, if the bets it accepts are too big it could get wiped out by a losing streak.
scriab, you are 100% off base.
ursa otoh, knows this stuff
scale trading on a STOCK can work 100X in a row. until it doesn't. and you get wiped out. cause a stock CAN go to zero, or close to it and never return.
check out a chart of VERT and tell me how well scale trading would work on that.
most people here are opining on scale trading without doing even a modicum of research on it. why does that not surprise me.
commodities ARE different. i make a living trading dow futures, but i also trade a fair amount of corn, beans, oil, gold etc.
i am not going to get into a long primer on scale trading. there are plenty of resources out there. but for those who think that scale trading stocks is a good idea, you are incredibly ignorant of risk management, risk of ruin, and other aspects of game theory, etc.
note that scale trading is not MERELY buying on weakness. it is not that simple. heck, in my B&H account i often buy on weakness and build positions over months or years. that is not the same thing as scale trading.
again, one thing i can count on in ET is people forming opinions FIRST without actually researching.
ursa otoh, knows this stuff
scale trading on a STOCK can work 100X in a row. until it doesn't. and you get wiped out. cause a stock CAN go to zero, or close to it and never return.
check out a chart of VERT and tell me how well scale trading would work on that.
most people here are opining on scale trading without doing even a modicum of research on it. why does that not surprise me.
commodities ARE different. i make a living trading dow futures, but i also trade a fair amount of corn, beans, oil, gold etc.
i am not going to get into a long primer on scale trading. there are plenty of resources out there. but for those who think that scale trading stocks is a good idea, you are incredibly ignorant of risk management, risk of ruin, and other aspects of game theory, etc.
note that scale trading is not MERELY buying on weakness. it is not that simple. heck, in my B&H account i often buy on weakness and build positions over months or years. that is not the same thing as scale trading.
again, one thing i can count on in ET is people forming opinions FIRST without actually researching.
1. Who would scale into any 1 position risking 100% capital? Probably some sort of idiot. Assuming max risk is set @ 2% 100+ & 1- is simply amazing;
2. What sort of trader would scale into VERT?
3. You can be just as wrong on commodity cycles as you could be on stock performance evaluation.
Also, as far as I'm aware one can scale into an unrealised losing position and into a winning one, which one do you refer to?
BTW I still would like to get an answer to my previous question if possible, if scaling is an "absurd" practise for stock trading, then why would people actually buy them in the first place? In the end of the day I can see your comment being related more to risk management concerning adding size disregarding maximum loss per position, I understand that.
Folks...I would like to post that scaling requires the ability to take losses. It can be a "floating" thing.
Martingale system is based on "Gambler's Fallacy" that if one event happens it is less likely to happen again.
This is stupid enough on a roulette wheel but in a trending market it is about as mad as you can get.
That said, I have witnessed many bund traders who doubled into losing positions all the time and were very successful.
This is stupid enough on a roulette wheel but in a trending market it is about as mad as you can get.
That said, I have witnessed many bund traders who doubled into losing positions all the time and were very successful.