Some ways to define a Trend.

[sidm]

Doubts And Questions

1. How is Inside-Out Heuristic different from indicators like Bollinger Bands? A price that gets a lot of votes in terms of time points should be pretty close to the arithmetic mean. A price that gets a lot of votes in terms of transaction volume should be pretty close to volume-weighted average price (VWAP). Perhaps because price distribution is not Gaussian, the values obtained by averaging diverge from values obtained by counting discrete votes?
2. The Outside-In Heuristic assumes that trend channels based on straight lines are good estimators of support and resistance levels. This assumption in turn implies that support and resistance levels change linearly with time. Why should it be so?

I think I have the answer to #2. The straight-line approach is basically drawing a trend line through the last few swing lows (which indicate support). This is the demand line. The only simplification is to draw the supply line by drawing a parallel line on swing highs, presumably to ease the visualization process for someone staring at a chart on the screen. But from a purely analytical perspective, the supply line could also be drawn independently by passing a trend line through swing highs.

For some reason I am unable to edit the previous post, so had to make a separate reply.

 

[Turveyd]

I use a dynamic MA based method, here is my chart of my current setup.

Arrows drawn on for where I'm looking to enter with the current direction.

I had the last 1 live for +19pts.

The white envelope gives fast direction and also expected noise chop range.


It's simple, can pick up and trade and put away and it's fairly objective which makes back testing possible to a reasonible degree of accuracy.

 

[dbphoenix]

Thanks for linking the document. It was a very interesting read. It packs in a lot of info. I am jotting down my current understanding below. Would be great if you could answer the questions I raise.

How To Determine Support And Resistance Levels Using AMT

1. Inside-Out Heuristic (using market profile): Count the number of votes cast for each price candidate. The prices with highest vote count are the winner (fair value). Prices getting the bottom 30% of the votes can be considered as support and resistance regions. Look for reversals and breakouts in these regions.
2. Outside-In Heuristic (using trend channels): Just focus on extremes since that is where trading opportunities arise. Draw straight lines through swing lows and swing highs. These straight lines are good estimates of support and resistance levels. DBPhoenix (SLA-AMT document) simplifies this even more by insisting that only demand line (support trend line) need be drawn. The Resistance line and the mean line are assumed to be parallel to the demand line.

Doubts And Questions

1. How is Inside-Out Heuristic different from indicators like Bollinger Bands? A price that gets a lot of votes in terms of time points should be pretty close to the arithmetic mean. A price that gets a lot of votes in terms of transaction volume should be pretty close to volume-weighted average price (VWAP). Perhaps because price distribution is not Gaussian, the values obtained by averaging diverge from values obtained by counting discrete votes?
2. The Outside-In Heuristic assumes that trend channels based on straight lines are good estimators of support and resistance levels. This assumption in turn implies that support and resistance levels change linearly with time. Why should it be so?

I don't know where you're getting all of this from, but it's nothing that I pay any attention to, so I can't answer your question.

However, in short, indicators, MAs, trendlines, etc don't provide support or resistance. They can't. Any seeming S or R they appear to present is purely in the mind of the trader and therefore eventually useless. This is not to say that one can't develop a marginally profitable method based on any or all of this, but it's no different from constructing a building on sand.

To clarify some terminology (fair value, votes etc.) used above, I am re-summarizing AMT (like I did in previous post).

Auction Market Theory

1. Markets are price discovery mechanisms - they help in answering the question "what is the fair value of this asset?"
2. This fair value is discovered by interaction between buyers (demand) and sellers (supply). Buyers and Seller vote on what they consider as fair prices.
3. Among various price candidates, the ones getting most votes can be regarded as what the market considers the fair value of the asset at the current moment.
4. Votes can be measured by time or volume. The number of votes cast for a price candidate can be measured by the amount of time during which asset gets traded at that price candidate. Alternatively, number of votes cast for a price candidate can be measured by the number of transactions that take place at that price. If certain price candidates occur during a lot of time points, or if a lot of transactions take place at certain price candidates, then those price candidates are the winner and can be regarded as fair value of the asset.
5. Trading strategy: avoid trading at when asset is trading at fair values since it is properly priced. Look for opportunities when prices move away from fair values. When price moves away from fair value then look out for a reversal or a breakout.

Except for #4, this is essentially correct. As for #4, time is not all that important. What matters is the number of trades. These can take place quickly or they can drag themselves out over time, but the number of trades may remain the same.

What is more important than finding value is following the search for value, as it is here that profit opportunities are most easily found. This is why buying something that's in a base is less productive than buying it when or shortly after it leaves the base.

 

[Turveyd]

However, in short, indicators, MAs, trendlines, etc don't provide support or resistance. They can't. Any seeming S or R they appear to present is purely in the mind of the trader and therefore eventually useless. This is not to say that one can't develop a marginally profitable method based on any or all of this, but it's no different from constructing a building on sand.

MA's are better for seeing direction, but the 100ema on the DAX has called a fair few reversals of recently.

Ofcourse, if price goes to 100ema, I don't take a trade I wait for Heiken Ashi or my 10sma setup to give me a safer entry, you've got to remove all the time where it powers through all S/R which would obviously create losers.

Base, has advantages tighter SL, more upside, but lower odds on success.

 

[dbphoenix]

MA's are better for seeing direction, but the 100ema on the DAX has called a fair few reversals of recently.

Ofcourse, if price goes to 100ema, I don't take a trade I wait for Heiken Ashi or my 10sma setup to give me a safer entry, you've got to remove all the time where it powers through all S/R which would obviously create losers.

Base, has advantages tighter SL, more upside, but lower odds on success.

You don't say better than what, but one should not require a moving average to tell up from down.

As for "calling reversals", nothing that stems from the mind of the trader can do so. But then there can be a wide gulf between perception and reality.

 

[bmwhendrix]

You don't say better than what, but one should not require a moving average to tell up from down.

As for "calling reversals", nothing that stems from the mind of the trader can do so. But then there can be a wide gulf between perception and reality.

If I may throw in my 2c worth. Although some traders profess to using ma's profitable, (usually along with some PA) their value can be negated if one simply understands and accepts that price behavior morphs over time. Width of channels, time between peaks and troughs, trendiness, etc. Todays 5 min chart may look more like last weeks 30 min chart.

So, any given ma will have times it seems to work, depending on how the market is changing.
When a traders chosen ma seems not to be working, they may dismiss it as a noisy or choppy market. It may be, but it may also just be the normal variations that have and will continue to occur. I now bow out, sorry for the intrusion.

 

[Turveyd]

You don't say better than what, but one should not require a moving average to tell up from down.

As for "calling reversals", nothing that stems from the mind of the trader can do so. But then there can be a wide gulf between perception and reality.

Better than my bare brain, I read too much into the noise, so the over all direction in general works better, obviously directions can change, but not that often so it's all good.

I use Heiken Ashi for the immediate direction, I don't trade against this, but I also trade in the direction of the 10sma using my range, so if 10sma is going up, market pulls back to the dotted area odds are it will continue in the direction of the 10sma, so SL other side of the not dotted 10sma line and with M1 and DAX you'd be surprised 80%+ win rate, profits twice SL easy.

Staying out of no direction is key and taking advantage of noise / chop.

Does require experience ofcourse, never just that easy.

 

[dbphoenix]

Better than my bare brain, I read too much into the noise, so the over all direction in general works better, obviously directions can change, but not that often so it's all good.

I use Heiken Ashi for the immediate direction, I don't trade against this, but I also trade in the direction of the 10sma using my range, so if 10sma is going up, market pulls back to the dotted area odds are it will continue in the direction of the 10sma, so SL other side of the not dotted 10sma line and with M1 and DAX you'd be surprised 80%+ win rate, profits twice SL easy.

Staying out of no direction is key and taking advantage of noise / chop.

Does require experience ofcourse, never just that easy.

If you need all of this, there's no reason not to use it. However, it's all in your head, and price doesn't care about any of it.

As a reminder, the question was about Auction Market Theory.

 

[Turveyd]

If you need all of this, there's no reason not to use it. However, it's all in your head, and price doesn't care about any of it.

As a reminder, the question was about Auction Market Theory.

80%+ win rate, so you won't mind if I don't believe your opinion

Ofcourse price doesn't care, I'm using the current direction/trend in a logic way to me!

 

[dbphoenix]

80%+ win rate, so you won't mind if I don't believe your opinion

Ofcourse price doesn't care, I'm using the current direction/trend in a logic way to me!

The win rate is irrelevant to the subject, but it's good that you're happy.