Welcome to NexusFi: the best trading community on the planet, with over 150,000 members Sign Up Now for Free
Genuine reviews from real traders, not fake reviews from stealth vendors
Quality education from leading professional traders
We are a friendly, helpful, and positive community
We do not tolerate rude behavior, trolling, or vendors advertising in posts
We are here to help, just let us know what you need
You'll need to register in order to view the content of the threads and start contributing to our community. It's free for basic access, or support us by becoming an Elite Member -- see if you qualify for a discount below.
-- Big Mike, Site Administrator
(If you already have an account, login at the top of the page)
a couple tips (related to daytrading) that I think will help newbies
Summary: I recommend using range charts, and not using "candle patterns".
I know alot of people use non-price-based charts & candles and they're successful with that, and they'll disagree with my opinions, and that's perfectly okay, and their comments are always welcome, but what I wrote below is what I honestly believe, so I'm honestly just trying to help people. Obviously I could be wrong though, so make your own decisions of course.
I think price-based charts are best (namely point & figure, renko, & range). But point & figure charts look ugly (due to how most/all software paints them) and they're basically swing charts that don't show the price churn in between the swings, and IMO renko isn't calculated correctly (plus it also consolidates some price churn that I personally like), so I use range charts. Those three charts are basically the same, in that they are price-based, and they result in plotting the same swings. By swings I mean the most basic swing consisting of one bar on either side of the swing bar, but once you start using swings that have 2 bars on either side of the swing bar, then there may be slight differences in the swings created by those three different types of charts.
The basic reason I think price-based charts are best is because the market is fundamentally tied to price. Price moves due to what the market participants think about the price, and also what the participants randomly want to do. Time frames less than a day or so don't affect the market, nor does volume. So by mixing time or volume into the price, you're adding noise.
So I use range charts with 'HLC' bars -- with no Open since it's always the same as the previous bar's Close.
A bonus benefit of range charts is that when picking what scale to set your charts at, there's fewer choices to choose from -- from a small 4-tick range chart to a fairly large 24-tick range chart, there's only 20 choices. Whereas volume charts & tick charts have alot more choices. Time charts also have more choices, although lesser so. So range charts will make it alot easier for you to choose your larger scale(s) for the 'bigger picture' of what's going on (support/resistance). But even if you only use 2 or 3 scales, it's still easier.
Volume charts are plotting price, and price is good, but they're not just plotting price -- they're also plotting volume which is bad. Volume doesn't show you direction or amplitude, so it's noise. All market participants base their decisions on price, so the volume is noise. So charts should plot price alone (range, point & figure, or renko charts). In general, you don't need to plot volume, because in these large markets (futures or large cap stocks), all market participants are always participating at all times, IOW they're watching the price all the time, so even if they didn't put their orders (volume) in yet, they're still participating, so you don't need to look at volume to determine if they're participating or not -- because they're always participating.
"Tick charts" are showing 'activity' by plotting how many transactions there are. But the number of transactions doesn't tell you any useful information. Alot of the transactions are small traders/companies and you don't want to see what they're doing anyway -- you want to see what the big companies/banks are doing because they have the big volume which makes trends. Big volume is what moves the price, not the number of transactions, so you want a volume chart, not a transaction chart. If you compare tick charts to volume charts, they're very similar because tick charts are just showing you almost the same thing as volume charts, because the average volume of each transaction over time is about the same. So the people that use tick charts should use volume charts instead because that would be more accurate information, (although I don't recommend volume anyways), so my point is that the people using tick charts don't understand that the only useful thing that tick charts do is show you volume, so they should be using volume charts, but they shouldn't even use that either. So the tick chart users are wrong not just for one reason, but for two reasons.
People get away with using volume/tick charts because on average over time it's still plotting price so the swings average out to be similar to range charts. But the larger the scale, the more different the swing data will be from price-based charts. So people get away with it only because they're using very small scales.
As for time charts, noone decides to buy or sell based on what time of day it is, or how much time elapsed since the last big level, so time charts distort the true swings. If you were plotting outdoor temperature, you could use time charts because time is a vital component of the laws of physics of temperature. But time isn't a major component of market prices. Unlike tick/volume charts, I don't think people can get away with using time charts because they can be alot different than price-based charts so they're often giving you wrong swing information. But if the time frame is short enough, I suppose people get away with it, but it still distorts the data.
The significance of a swing is based on how large a price (y-axis amplitude) it is. So swings are determined by y-axis, price difference. The x-axis doesn't really matter, although there is an x-axis just so data isn't plotted on top of itself. Swings are based on the current bar passing the prior bar's High or Low so swings are based on the amplitude/y-axis. But with non-price-based charts, the amplitude varies so it randomizes your swings. Anything with a varying bar size (amplitude) is adding randomness to the swings that the chart plots. Since swings are defined by y-axis movement, for swings to make sense, the current bar must be able to be compared to previous/historical bars on the y-axis, but with non-price-based charts, you can't compare one bar to any other bar basis the y-axis because they're all different sizes. For example, a 15-minute time chart has one bar with very low activity, so it's a very small amplitude bar, but then the next bar will make a swing because the previous bar was so small, so it's making swings based on 15-minutes of low activity, but that's not what you want -- instead you want the swings to be based on a minimum price movement. Conversely, if a 15-minute time chart has a big amount of price movement within 15 minutes (a big bar), then the next bar won't be able to make a swing since the previous big bar's High or Low is so far away -- again, that's not what you want.
About candles:
Range charts always Close on either the High or Low, so they don't really have candles. But I think that's a good thing, because I don't think candles actually give you any useful information. With a range chart, your "candle pattern" for a signal is simply a range bar that reversed direction. Whereas with your tick/volume chart, you're looking for a "good" candle pattern, but that is just an illusion. What a "good" candle pattern is really showing you is only simply that price moved a certain minimum amount. The candle patterns actually change if you phase shift the bars (if you start the chart on a different tick), which proves that they are an illusion. You can prove this yourself -- just look at a smaller tick/volume/time chart that has twice as many bars as your normal chart has, and build your normal chart's candles from the smaller chart ... you'll get two different normal charts with two different candle patterns -- based on which phase of the smaller chart's bars you start on. So you'll see a "good" candle pattern, and also a "bad" candle pattern from the same exact data! When you see a "bad" candle pattern and the price ends up moving in your trading direction against that "bad" candle pattern, what happened was that the "bad" candle pattern was actually a good pattern -- all the "bad" candle pattern means is that there was more price churn (& volume) before the price started moving/trending, which is a good thing, so the "bad" patterns are actually good. So the "good" patterns are good, and the "bad" patterns are good, so they're all good. Looking at candle patterns is sometimes giving you wrong information -- when you see a good pattern then it doesn't hurt you, but when you see a bad pattern it's wrong information. All that matters is the market reached some kind of starting price (like it's making a potential top/bottom), and then moved a significant amount away from that price in the direction you're looking to trade. If you still want to look at what happened in the time span between those prices, to see some kind of smaller scale signal/pattern, then look at the bars or swings on a smaller scale chart (for that time span).
Candles only show you one thing -- what happened inside of that bar, and only crudely since they only show you 2 data points (the Open & Close) besides the High & Low. They don't give you any information about what happened outside of that bar. IOW if you use multiple candle bars to show you a "pattern", the candles aren't showing you anything different than what range bars are showing you -- that price moved from here to there. When it comes to a bar, there's the High, the Low, and the middle (average), there's really nothing else to use for data analysis, although the Close is the latest price of course but that's really only applicable for the current realtime bar.
I'm not familiar with Bookmap, but I just looked it up a little online. For me personally, at least so far, I haven't been into volume. So far, I just can't seem to find much value in it. Obviously the people with the largest volume move the price, and we traders want to follow that trend. But the price can move on high volume or low volume, plus volume doesn't tell you which direction the price will move -- I'm sure the big volume uses limit orders so the price could be moving up with high volume, but then the price stops and trends the other way because the big volume was just accumulating with limit orders going up in preparation for market-ordering the price down. So how can anyone read what the trend direction is from that?
I've looked up some ideas about footprint charts, and I haven't used it myself yet, but it seems like it might have some value, but I'm not really sure yet.
As far as volume profile, I don't know what to make of it, it seems to me like it's just going to show you high volume profile peaks near big price levels (always 'inside' of the level though which makes sense obviously).
I guess I'm just not very sophisticated in my analysis techniques, but I have yet to see much value in volume analysis. But if I ever see anything very useful about it, I'd be the first one to check it out, and the first to admit that I've been ignorant about it. But so far, I just don't understand why it's so great.
The only useful thing I've been able to figure out so far about volume, is to run a simple "volume speed" indicator. It's a histogram format. It's just volume divided by time, and then you have to multiply that by volume in order for all the bars to be comparable to each other. IOW it's volume squared, divided by time. That seems to be much more usable than just simply plotting volume by itself. It shows alot of volume speed (or general market "activity") at common exit points, and also for the most part also at major tops/bottoms. But, it's not always as good a predictor of tops/bottoms as it may sound, sometimes it doesn't help so much, sometimes it does, so it's fairly unreliable. So I don't really use it.
I wish there was more data besides price to get information out of, but so far I just use price. But that's not so bad I guess, the big volume players seem to want to buy low and sell high, IOW they try to get in at good value prices, and all the smaller players always do that too of course, so that's the market's main "law of physics", so I just try to ascertain what a good value price is via a simple proprietary channel indicator, much like Bollinger Bands but not that because I don't like it for a couple of reasons, and that's my "support/resistance"/"setup" where I look for a price action signal to get in. My price action signal is probably the same exact techniques that most others are doing (I use two types of signals, one of them is used by DayTradingAcademy and the other one is used by the PATS system), but for the final "signal bar", instead of using candles I just use one single range bar that reversed direction.
I guess my analysis is rather boring and possibly 'old fashioned', although modern day computers & software are able to do wonders that could never be done decades ago -- drawing many charts on the fly, real time data, running indicators, etc -- the computer does all the work for you.
If it works for you, that's great. In my 15+ years trading futures I've tried em' all and came back to time-based charts again and again.
Just to let you (as range-bar fan) know, the open of a range-bar isn't the close of the previous bar, the open is always 1 tick above or below the previous bar high/low that's why some call them also break-out bars.
Every business wants/needs to know how many products are bought/sold (volume) at what price and in which period (time). Time is most important in life as well as in trading. It's all we have (in life).
Concepts like absorption, ease of movement, effort versus result, supply/demand (im)balance, price acceptance/fair value are all too important to understand and cannot be derived from price alone IMO.
Like you said, if it works for you, that's great. But so you know, Not all range style bars are the same. There are range bars available with close previous bar/open new bar match.
I have developed and traded many good strategies using time based bars, though I have shelved most of those and now use Range bars systems that perform much much more consistent.
Although I add one should be very careful of any exotic bar styles when backtesting. Make sure you know what your doing?
In trading, Time sometimes is not your friend. JMHO.
Range bars is kind of new thing for me, in fact it had kind of flew under the radar so far. But since you mention comparative performance, can you please elaborate? I mean in which areas it was more consistent? Or what benefit was prominent over time bar?
//Excuse if question is too silly or basic, I really don't know about it or anyone who uses it
I am pretty sure that candles are based on price also. They might be locked into a time structure but they are still based on price. Just my humble opinion though.
I prefer range bars as well, especially when combined with order flow tools.
here's a good example why:
the high (3,695) had a significant seller. when looking at a 5 min chart, we can see 11 bars with that high:
when adding a footprint to that 5 min chart, we can see a lot of numbers and kind of a mess:
now the same with a 10 range bar. only 1 bar to analyze and already looking very suspicious:
and when looking at a 10 range footprint bar, we can clearly see that huge seller. again only need to look at a single bar:
of course both, the 5 min and the 10 range chart show the same thing. it's just a lot easier to see the details. should also mention, that kind of set up does not work all the time. but it's a good start to pay attention
that's why some call them also break-out bars.
