BHReach

1st let me say that Tom Williams' original book, 'The Undeclared Secrets That Drive the Stock Market' seems to be based on sound principles; however, there are issues with some explanations and their videos have problems. 1. In one video, Gavin Holmes displays a 4 hour chart of AAPL. He states in the video that it creates 2 bars per day. That is nonsense. The market is open 6-1/2 hours per day, if you start the 1st 4 hour bar when the market opens and end the last bar at the end of the day (the traditional approach), you will get 4 hours worth of data on the 1st bar and 2-1/2 hours worth of data on the second bar. If you want 2 bars per day, you have to put 3-1/4 hours worth of data on each bar. Gavin's actual chart was even worse than that, there were 3 bars per day. He started the 1st 4 hour bar 3-1/2 hours before the market opened so the 1st bar had 1/2 hour of data, the 2nd bar had 4 hours of data and the 3rd bar had 2 hours of data. If you compare the volume for these bars you are comparing apples to oranges because all the bars do not contain the same amount of data. If he put up a chart with 3 minute bars, 5 minute bars and 9 minute bars mixed together, he would be laughed out of the business. What he is doing with the AAPL chart is mixing 1/2 hour bars, 4 hour bars and 2 hour bars on the same chart but he is so clueless he calls them all 4 hours bars and doesn't realize he is doing anything wrong! If these guys cannot even display data correctly on a chart, you have to wonder how useful their software is and how credible anything they say is. 2. They put too much emphasis on individual bars. If the Professionals test for supply, how and why would they make the test fit neatly into one bar? The answer is they wouldn't. The test may last for only a portion of a bar, it might fit on 1 bar or it might span several bars. They will vary how they execute the test to disguise what they are up to. 3. The TG guys put too much emphasis on the close. There is some justification for that on daily bars because the market actually does close and there is unique behavior associated with that; however, on intraday charts, the close is just the last price for the bar and if you look at different data feeds, the intraday closes are different. 3a. The speakers in their videos consider a bar to be an up bar if the close is even 1 tic higher than the previous bar. That is ridiculous. There is enough noise in the market that very small differences in closes are random. 4. Their explanations in some cases are incomplete. For example, nowhere is it explained why a No Demand bar has to close higher than the previous bar's close. When testing for Demand, Smart Money rapidly marks prices up to draw out any buyers. If no buyers come in, i.e., No Demand, it is by no means obvious why price then has to remain above the previous bar's close.

Others have already answered your question, I use market fractal instead of time frame because time frame does not seem appropriate for tick, volume or range bar charts. Bill

Apparently I stepped on some toes by expressing my opinion about floor trader pivots and other calculated ('magic number') price levels. I apologize for that. Let me summarize my views so they are clearly understood: 1. FT pivots et al are crude measures of regression to the mean (referring to them as gravitational levels is OK too). Mean reversion is a very important market principle. That means you should be able to construct profitable trading methods using the calculated levels. 2. My observation (note the term observation not scientific study) is that in the vast majority of cases, when the 'magic number' price levels work as S/R, they are coincident with previous S/R based on PA and in most cases when they don't work, there is no previous S/R based on PA at that level. That means to me that they don't add much in the way of new S/R levels that are not already available from S/R based on PA and they do add some false S/R levels. 3. There are many variations of FT pivots. Sierra Charts has 10 to choose from. Which to use, which to use? At the risk stepping on more toes, I don't use indicators much either. Most algorithmic indicators are some kind of digital filter derived from price, that makes them lagging indicators. If you use them to generate trading signals, you will enter trades later than using PA most of the time. For example, a 10 bar MA lags prices by 5 bars. It tells you on average what prices were doing 5 bars ago. In addition, most indicators add no new information. The information in price derived indicators can be seen directly in PA (in most cases). Also, most of these indicators have a fixed look back period (like the 10 bar MA). That means the digital filter is tuned to a specific frequency. As long as the dominant market cycle is compatible with the tuned frequency the indicator works well. When it's not, the indicator works poorly. E.g., a MA crossover system works well in a trending market (low frequency dominant cycle) but gets whip sawed in congestion (high frequency dominant cycle). Note: a MA is a low pass filter. Trendlines/channels are a good way to sense changes in the pace of the market. If you take the trouble to learn how to use Andrews Pitch Forks plus variations, they can be a very effective trading tool. I don't use them, I can see changes in pace directly from PA. You need to be aware that changes in the price scale can effect your perception of market pace. This is most noticeable when prices go parabolic. To see what I mean, take any chart and double the price scale, now double it again. Compare the 3 charts and notice on the 3rd chart how much the trends are squashed wrt the 1st chart and how difficult it is to compare the pace of the trends. I am not exactly sure what VWAP is but from the flyover hint Walter put on his chart, I would guess it's a MP thing. I looked at MP briefly and decided it was too complex to learn in a short period of time so I put it aside for later investigation. From the short exposure I had to it, it looked like the important price levels occurred where prices spent a lot of time (like in a congestion period). I already have those marked as important S/R levels based on PA so I don't know how much new information it will add. Also, since it is based on volume, it cannot be used for the Forex market. I like things that are simple and work in all markets on all market fractals. PA with PA based S/R fit the bill for me. Every trader has to create their own trading plan that is compatible with their temperament. Bill

I don't use Fibonacci, Gann, or Pivot levels for the reason previously stated, I consider them redundant. The information overlap between those levels that work and Price Action based S/R is extremely high. When considering trade entry, I want confirmation from independent sources of information. For example, I like to trade reversals at S/R when it is combined with trend exhaustion. S/R levels are really zones. The 'size' of the zones depends on several things, the market fractal the level is visible on, how the level was created (swing hi/low, area of congestion, gap), and price pace as it approaches the level. If price is approaching S/R with high momentum it is likely to at least overshoot if not blow right through. OTOH, if prices are struggling to make progress, they are likely to under shoot the S/R level. It is amazing how accurate the S/R levels are respected on the smaller fractals with average price volatility. I don't think you can draw that conclusion. He has demonstrated sufficient skill in mathematics and programming to be able to do the pivot research. Not showing the details of what he did does not mean he simply eyeballed some charts. I do admit that the lack of detail about what he did reduces the value of his conclusions (i.e., it's wise to be skeptical). Bill

3 x 3 x 3 = 27

They may not have been created based on solid market logic but they are a very crude way to measure "regression to the mean" type turning points. From my own observations, when they appear to reverse the market's direction, I look to the left on the chart and in the vast majority of cases they coincide with a price action based S/R area and that S/R is usually the more accurate of the 2. Doug Tucker has done some research on Floor Trader Pivots and he could find no difference between them and random horizontal lines as far as reversing the market goes (they do have other uses). Bill