One can abstain from pyramiding.
I tried it but not rigorously enough,
To know if it's worth the increased risk.
However, it has been far from being bad.
It's especially appealing Theoretically.
However, it can be fatal in practice.
To PYRAMID into STRENGTH
- Thread starter K-Pia
- Start date
I tried it but not rigorously enough,
To know if it's worth the increased risk.
However, it has been far from being bad.
It's especially appealing Theoretically.
However, it can be fatal in practice.
Ok, there are 2 scenarios where you MUST pyramid if you want big gains and not stay an average trader.
1 classic trend, with pullbacks : you add positions at every pullback.
But you need to manage your average price too so you can't do it indefinetly.
2 euphoria or panic : these are volatil days when nothing can stop the squeeze.
You are not sure at the opening if that be could a big one. But every time you pyramid, there is just no opposition. Add positions regularly and place your stops to protect. Be a pig.
The first case is easy and natural, but you need to know what is a trend of course.
The second one : you need experience, focus, and solid risk management.
At last but not at least, there is one scenario where you MUST avoid pyramid :
in a trading range
CM
The thread is called "To Pyramid into Strenght".
So we agree about the condition this scheme requires.
... One always needs to be responsible, toying with leverage.
I can't argue with success, but you're saying to pyramid when you know the probabilities are in your favor. If this is based on large sample testing of a quantifiable scenario, I say go for it, but otherwise you simply don't know if the odds are in your favor. What's your basis here?
My basis are called "experience", "price action" and "technical analysis", not quant, backtest or sample testing or whatever.
I can assure you when you trade more than 10 000 times, you will begin to understand when odds are in your favor or not. Of course, there ain't no certainties, but patterns repeat and repeat again.
Let s give one short example. You can backtest it if you want. But better take it as granted, you will gain time and money.
Assume we were in a tight trading range for several days. Then S&P is down 1%. The day after, S&P is down 2%. And the day after, S&P tries to rebound 1%, then is negative.
At this moment, odds are in my favor to short heavily S&P, and increase my shorts all the way down.
Why? Because we probably face a squeeze, and bulls begin to panic. Indeed, when you are in a trading range for a long time, it means there were a balance between bulls and bears. The longer the trading range, the better. Imbalance creates panic, volatility and big money for the astute trader.
CM
Fair enough. I spent years honing the skills you mention above, and I think every trader needs to do that. Like you I also had the impression that all the backtesting, sampling, and quantitative stuff was a joke.
I eventually got to a point where I just got tired of thinking and analyzing the charts on the screen. I also realized the value of diversified non-correlated bets. I was able to translate my experiences into a mathematical model. It increased the quality of my life and allowed me to manage multiple instruments and test multiple scenarios all at once. The value is all in the implementation, and unfortunately without experience it's difficult to get it right. Surprisingly, it improved my performance even though I expected the opposite.
Happy trading
Last edited:
Yes and this is perfectly prudent plus a great idea in general. However it's not mutually exclusive with pyramidding - which is purely a function of just how well said bets are working for you.
I think one of the things you're implicitly focusing on is the psychological aspect of big positions and how instead trading multiple smaller positions works better for you from a psych standpoint. Adding to a position does not mean one forgets about managing risk on that position. Let your gains work for you while keep risk <= to what the original position initially had (or close enough).
Actually trading smaller multiple diversified positions is proven to work better from a statistical standpoint. It smoothes out the equity curve, which is definitely more comforting from a psych standpoint
Yeah I'm not disagreeing with it at all and I do it myself. But when you have a great trade, and it's highly evident it'll continue, you *should* add.
Why not do it mathematically?
For example, there are those who feel strongly that volatility should play a role in position sizing. Not me, I'm just saying. I believe the usual assumption is that the position size is proportional to 1/volatility, however you chose to measure volatility. Then finding the coefficient of proportionality is a straightforward optimization problem. Easy peasy.
If you have a measure of trend strength (let's call it X), you find the coefficient of proportionality that applies to X. Then you pyramid when X significantly increases from your trade entry point. The derived relationship between position size and X tells you mathematically how much to increase. No guesswork.
For example, there are those who feel strongly that volatility should play a role in position sizing. Not me, I'm just saying. I believe the usual assumption is that the position size is proportional to 1/volatility, however you chose to measure volatility. Then finding the coefficient of proportionality is a straightforward optimization problem. Easy peasy.
If you have a measure of trend strength (let's call it X), you find the coefficient of proportionality that applies to X. Then you pyramid when X significantly increases from your trade entry point. The derived relationship between position size and X tells you mathematically how much to increase. No guesswork.