The Precedence of Price

AD: Before I begin this post, I’d like to briefly mention Bitcoin.Live, who are sponsoring my blog.

Bitcoin.Live offers regular, detailed content on their free-to-access blog, created by a panel of analysts (including Peter Brandt), and covering all manner of market-related topics. I found both the video material and the blog posts to be genuinely insightful, with many differing analytical perspectives available for viewers and readers. The platform also offers premium content for paying subscribers who find value in the free material, with daily videos, alerts and support provided. Check it out and bookmark the blog.


I’d like to preface this post by mentioning that this, unlike the vast majority of blog posts I publish, is expressly aimed at newer speculators, with little to no familiarity with markets.


The price paid for any asset is the single most important aspect of speculation, and one that largely determines the profitability of said speculation.

This idea is drummed into our heads by the most successful traders, investors and speculators of all time; and justifiably so.

Whilst investors like Warren Buffett refuse to acknowledge the long-term utility of technical analysis, relying solely on fundamentals in their decision-making process, they all – going all the way back to Benjamin Graham – make the concession that price is paramount. Price is both fundamental and technical in its nature, with all technical analysis deriving from two data-points: price and volume.

For the purposes of this post, we will be focusing on price, as it is the most critical component to your decision-making process.

Now, the problem with a sole reliance on fundamental analysis is that it often disregards the price-history of an asset. Whilst there are many valuation models in equity research that allow for a relative measure of cheapness to be determined, this is not the case with cryptocurrencies, at present.

In this space, we rely heavily on past prices for an indication as to the relative value of current prices. This is because fundamental analysis is very much in its infancy within the cryptosphere, and those valuation models do not yet exist, thus technical analysis is critical to prevent from overpaying for any given altcoin; an act that renders profitability highly unlikely.

A project might be impeccable in its fundamental quality, comprising of an engaged and growing community, consistent development, a working product and unique use-cases; but if you enter a position in that project at or near all-time highs, there is a strong chance that you spend a long time underwater or even realise a loss. Indeed, gravity is far greater a force within this particular asset class, and no amount of structural soundness will prevent the eventual (and cyclical) collapse.

As such, a study of price-history is indispensable when considering an entry for a new position. In fact, if you do nothing else in your research except study price-history, you will likely be doing enough to find some degree of success.

Now, this is all exceptionally simple; intuitive, one might argue. And so, I’ll refrain from rambling on any more. Rather, I believe it will prove useful to provide some examples of this concept of relative cheapness, if only to illustrate its simplicity.

First, consider all that has been said thus far. All we are looking for are periods of cheapness and dearness across price-history – in essence, we are reframing the concepts of support and resistance and simplifying them.

Below are printed the before-and-after charts of several coins. These are just a sample of the dozens and dozens of examples that are available in the market:

DOGE:

Dogecoin is perhaps the best possible example to illustrate this concept, as it has experienced numerous market cycles in its price-history. Looking at this first chart, intuitively, which prices look as though they are cheap and which look expensive?

Here, I have depicted my thoughts on this, where there is a price range of extreme cheapness, which has only been traded 4 times; a price range of cheapness, which has been traded 8 times; and a range of expensiveness, which has been traded 6 times.

Those that have bought within the extremely cheap range have been rewarded every single time with a possibility to sell within the expensive range. Those that have bought in the cheap range have been rewarded 5 out of 7 times with the same opportunity (excluding the current, 8th touch of this range) without experiencing price trade inside this range a second time prior to reaching the expensive range. To clarify this, look at the 2nd and 6th periods of price trading inside the cheap range: subsequent price-action took price up and out of the range but fell short of the expensive range, moving back inside cheap before making another (this time successful) run at expensive.

In all of Dogecoin’s price-history, buyers below 50 satoshis have been rewarded with an opportunity to sell above 100 satoshis, no matter when they decided to buy below this point.


STEEM:

Now, above is printed a STEEM chart, but I have excluded the first few months of price-history simply because it was an anomaly that massively distorts the chart. This is often the case when coins first begin trading, as circulating supplies are extremely low, drastically affecting early pricing.

Where, on this chart, would you consider price to be cheap and where is it expensive?

Clearly, there is one price-range that indicates relative cheapness across STEEM’s price-history. This range has been traded inside on 3 separate occasions. Each entry within this range has been reward with an opportunity to sell inside the expensive range except the most recent one, from which price is only just exiting. If we look between the 2nd and 3rd entries into the expensive range, we can see a failure for price to return to cheap, and thus bargain hunters would have missed this opportunity; a risk always present for those who adhere to the approach of buying at the very extremes of relative cheapness.


BCN:

BCN is an interesting case, as it spent a long time in a relatively tight price-range before experiencing several cycles in quick succession. Where would you consider price to be cheap for BCN, and where is it expensive?

Now, whilst this entire exercise is largely subjective and open to interpretation, BCN in particular is a trickier case. Whilst I doubt anyone would argue against there being a cheap range below 20 satoshis, I have marked out the area above it (between 30-40 satoshis) as relatively expensive rather than, say, less cheap or relatively cheap. That is because this range between 30-40 satoshis has only acted as a catalyst for a move towards the expensive and extremely expensive ranges once without price also moving into the cheap range. It has also acted as a range from which price has actually declined back into cheap without moving any further up on two occasions.

You might feel that this is actually still a relatively cheap range, and that is your call to make. For me, there is only one range that I’d be buying, and that is below 20 satoshis. Now, every single time price has traded within this range, it has been followed by at least a move up to 40 satoshis, with the more common pattern being a move up towards the expensive and extremely expensive ranges.

Price is currently trading inside the cheap range for the 5th time in BCN’s price-history.


SC:

Siacoin has a beautiful price-history, with very clear areas of relative cheapness and dearness. Looking at the above chart, where would you expect these to be?

Clearly, the range below 50 satoshis is extremely cheap; this range has only been traded within on 2 separate occasions. There is, however, a more recent cheap range that has formed just above it since around November 2017. Why is this range relatively cheap? Because it – like the extremely cheap range – acted as a catalyst for extremely expensive prices.

We have three gradations of dearness depicted, also. There is a relatively expensive range that has been hit 4 times in price-history; 3 of which have occurred after price has traded inside the cheap or extremely cheap ranges.


That concludes the examples, but I would advise you to go and seek out your own examples. The more you study and observe the entire price-histories of altcoins, the more you’ll spot periods of relative cheapness that can be capitalised on. Don’t make things difficult for yourself by seeking out opportunities in altcoins with obscure and clumsy charts; it is unneccesary when there are thousands of coins out there.

I hope this post has provided some insight for less experienced speculators.

As ever, feel free to leave a comment or question below and I’ll get back to you!

The Application Of Risk

Risk is an elusive concept to cover, and certainly a much misunderstood one. It is defined in different ways for different purposes but it is critical to fully understand what constitutes risk in order to find sustained success in any speculative venture.

Depending on the context, risk can mean the expectation of volatility and illiquidity: This market is one of great risk. It can also be an albeit abstract measurement of the likelihood of success or failure: I believe this is a low-risk proposition. Lastly, risk can be a calculation related to exposure and downside versus upside potential: I have £5,000 at risk here, though my returns could be as great as £15,000. This latter definition is the one most commonly used by traders, but I believe an understanding of them all is particularly useful for profitable speculation. Only by seeing the full picture of market volatility, exposure and risk versus reward can we then come to some sort of conclusion on the second definition; whether our current speculations will prove successful. In fact, part of the full picture of risk is illuminated by the price-history of the market, depicting where, in the past, similar scenarios to those we are presently expecting have seen success or failure.

This post will, I hope, serve as foundational material for those who are unfamiliar with applying the many aspects of risk to their speculative positions. I will run through the process for each of the three relevant definitions of risk and how they each relate to the full picture.

Firstly, however, I’d like to emphasise the most essential point concerning risk: Particularly when speculating in the cryptosphere, the thing that determines whether one masters risk management or not is whether one invests money they cannot afford to lose. If you start out with non-discretionary income, you’ve already lost the game. The other components of risk management are only relevant if your speculations are comprised of money you can afford to lose. If this is not the case, the likelihood is that no amount of searching out low-risk, high-reward opportunities is going to save you, as your emotions are inextricable from your positions.

For those that the above applies to, stop reading and reorganise your portfolio until it resembles something that you could lose the entirety of tomorrow and it would not affect your quality of life. For the rest of you, let’s crack on.


Risk, as in volatility and illiquidity:

When a market experiences high levels of volatility (as is the case with all cryptocurrencies), it is said to be risky. Similarly, when a market is highly illiquid, there is inherent risk in exposing your capital to said market, as there is a possibility that, once a position is entered, it would be very difficult or costly to exit until market conditions improve sufficiently.

These aspects of risk management are critical to one’s speculative positions, as they are very much linked to personality and thus the quality of the decisions one makes. If you are highly risk-averse by nature, entering a position in a more volatile and illiquid market is perhaps not the brightest idea; if you are risk-tolerant, an illiquid market may not bother you and high volatility may not affect your trading decisions. In either case, having a clear understanding of these aspects of risk prior to entering a new position is a contributing factor to the likelihood of success.

But how can one determine volatility and illiquidity as a component of risk management? Most commonly, these are abstract terms for the general market participant, relative to the more concrete calculations one makes for exposure and reward-to-risk. However, simple calculations can be made to make things clearer.

For volatility:

  1. I tend to first determine the duration that I’m expecting to hold a position for. In my case, this almost always tends to be over a month. We’ll use a month for the purposes of clarification.
  2. Given this trade duration, I collate (in a spreadsheet) 30 days of historical price data for the coin I’m interested in using Coinmarketcap’s Historical Data tab.
  3. I delete everything except the Close Price data.
  4. I then calculate the average Close Price for the 30 days. This is my benchmark figure.
  5. Using this average Close Price, I calculate the percentage change from it to the highest Close Price during the month.
  6. I do the same for the lowest Close Price, also.
  7. For example, if the average price was $1, the highest price was $1.50 and the lowest price was $0.30, this would give me figures of 50% and -70%.
  8. The final step is to multiply these figures together to find a volatility ratio for the given duration. In this case, it would be -0.35 over the past 30 days.
  9. The closer to 0, the less volatile the market during that period of time, and vice-versa.

This process is particularly useful for cross-comparing the volatility of altcoins over the same time-period. It is rudimentary in its methodology, but gives us some form of concrete figure to apply to our risk management. Those that are risk-averse may opt to only enter positions in coins that have volatility between 0 and -0.1, for example.

For liquidity:

  1. This is even simpler than the calculations made for volatility. The first step is to calculate the buy support across listed exchanges for the coin you’re interested in, within 10% of the current price. Calculate this in BTC-denomination.
  2. Now divide this figure by the market cap of the coin (again, use the BTC figure).
  3. Multiply the result by 100 to get the buy support as a percentage of the market cap.
  4. Anything lower than 0.1% is highly illiquid. Anything higher than 1% is highly liquid. Most altcoins tend to be between these two figures.
  5. Do this once a day for a week and calculate the average to get a more reliable figure.

Now, the most important thing to do with this information is devise a benchmark that works for your personal relationship with risk. And stick to it. For some, this will be a commitment to only entering positions in coins with greater than 0.5% liquidity and between 0 and -0.05 volatility. Just make sure you know what works for you.


Risk, as in likelihood of success or failure:

This second definition of risk is, as mentioned earlier, more abstract than the other two. We often use low-risk synonymously with high-probability in everyday conversation, or high-risk synonymously with low-probability. The utility for speculators comes from finding historically similar scenarios to those we are expecting to profit from and evaluating their successes and failures. To make this clearer, let’s use a simplistic example:

First we must define the terms of the position we are considering. Let’s say I am considering an entry on X at 3000 satoshis. I am anticipating prices above 6000 satoshis, and would consider my trade idea incorrect below 2000 satoshis (which would be my soft stop-loss). I am willing to hold the position for 3 months.

Given these points-of-reference, we would simply backtest the trade using the coin’s price-history. Every time price reaches 3000 satoshis, we would enter an imaginary trade; does price reach 6000 satoshis? Does it reach it within 3 months? How many times would the position be stopped out? Ask all the relevant questions and compile an historical evaluation of your trade idea. If we’re looking at a trade that has been successful 80% of the time in the coin’s price-history, it gives us some degree of confidence that our own position will be successful, also. Of course, to be able to evaluate your idea to this degree, you first need to know all the critical information regarding exposure, entries, exits and risk versus reward. As such, the most important aspect of risk management outside of using capital you can afford to lose is found in the third definition.


Risk, as in exposure and returns:

Risk management for traders is mostly concerned with this third definition of risk that concerns all things quantitative, and for good reason. For me, calculating exposure is a prerequisite to entering a new position. It is the primary element upon which the rest of the trade is structured. Speculating without a clearly defined plan for capital exposure is a sure-fire way to wipe out your portfolio, and we don’t want that, if we can help it…

I will, at a later date, be writing an in-depth post on position sizing, which itself is a integral part of managing exposure, but for now let’s consider the basics. You could, of course, create an intricate plan of position sizing based on the volatility and liquidity calculations I mentioned earlier – almost as though you’re basing your risk on, well… risk itself. Riskception. But for the purposes of this post, let’s stick to the most common method of determining position size, which is focused on market cap or network value, however you like to refer to it:

  1. Firstly, you need to calculate the value of your portfolio. This is the base figure that you will use to calculate exposure for a new position. Let’s say it is 10 BTC, or ~$40,000 at current prices.
  2. Now, figure out whether the coin you are considering a position in is a microcap, lowcap, midcap, highcap or megacap. These are arbitrary terms, of course, but I can only offer my approach here. I categorise these using the following figures: microcap = 0-25 BTC; lowcap = 25-250 BTC; midcap = 250-2500 BTC; highcap = 2500-25,000 BTC; and megacap = 25,000 BTC or higher. These, again, are subjective numbers based on my own experiences in the space. If you wanted to use $ figures (though I advise against it, as these are heavily dependent upon the price of Bitcoin), then I’d opt for 0-$250k for a microcap; $250k-$2.5mn for a lowcap; $2.5mn-$25mn for a midcap; $25mn-$250mn for a highcap; and $250mn or higher for a megacap.
  3. Now, for each of these market cap-based groups, I have a different band of exposure based on the original value of my portfolio prior to entering the position: 0-1% for microcaps; 1-3% for lowcaps; 3-5% for midcaps; 5-10% for highcaps; and 10% or more for megacaps. I do not commit to the minimum percentage exposure within these bands, but I explicitly do not exceed the maximum for the given market cap. For example, I might choose to only allocate 5% of my capital to a megacap, but I would never allocate 5% of my capital to a microcap.
  4. Of course, there are some caveats here. Firstly, this approach is known as fixed-risk, wherein one allocates a fixed percentage of capital to a position often in lieue of setting a stop loss (but not always, as we’ll come to shortly). The position is then held until: it reaches its target price(s); it fails to reach its target within the predetermined duration of the trade, at which point it is exited; or, the coin dies. This is a common approach with microcaps and lowcaps, but makes less sense when one is concerned with the larger coins.
  5. When it is these larger coins that are being considered, the bands of exposure are still used, but a stop-loss (hard or soft) is added as a second risk-mitigator. My approach to stop-losses is that they should be based on technical factors rather than predetermined percentages, such as the break of long-term support or something similar, but it is often useful to have a maximum percentage stop-loss in place. For example, let’s say we were looking to enter a position in ABC. ABC is a highcap and so our capital exposure is a maximum of 10% of the value of our portfolio. Further, since it is a highcap, we choose to place a stop-loss. The maximum we are happy to lose is 25% of the initial capital, and thus a stop-loss is placed 25% below the average entry price. This equates to 2.5% of the value of our portfolio, which is our maximum capital loss.

Now, there are numerous other avenues one could go down when devising an approach to risk management, but I believe this approach will suffice for most. The issue is that stop-losses can and must (in my opinion) be linked to the risk versus reward of the trade. So let’s discuss the final aspect of risk for this post: returns.

The goal in investing is asymmetry – Howard Marks

Asymmetrical opportunities are the real secret to profitable speculation and proper risk management. Finding opportunities that present returns many multiples greater than the potential risk is what is so special about this space – they are ubiquitous.

Reward is almost always predicated on the price paid for the position, which is why buying low is so important. It allows for the low-risk (here meaning minimal amount of capital loss), high-reward opportunities. The most important thing to take away about risk versus reward is to exclusively go after opportunities that offer at least twice the reward against the risk. So, if, after calculating your capital exposure and your stop-loss, you have a maximum capital loss of 2.5% of the value of your portfolio (as in the earlier example), then your opportunity must present a reward equating to 5%. This is why the fixed-risk approach is suitable for the smaller altcoins; the potential rewards are so large that the trades are often asymmetrical in our favour despite the potential loss of the entire position.

Now, to conclude this post, how do we tie it all together? Well, what you can do is create a risk framework that all future trades must adhere to, and this would be based on your own level of risk tolerance. For example, you could decide to only enter positions in coins that have 0.5% liquidity, between 0-0.1 volatility over the past 90 days, at least one instance of success of a similar scenario in the coin’s price-history and at least 3:1 reward-to-risk. Play around with the numbers to see what works for you – the important thing is to have a consistent framework to which you always adhere.

I hope this post has proved useful. Feel free to leave any comments and questions below and I’ll get back to you!


If you’ve enjoyed this post and want to receive new posts straight to your inbox, I’ve set up a RSS-to-Email feed that will be sent out weekly; every Monday, 12pm. Just submit your email and I’ll make sure you’re included in the list. Cheers.

Orderbook Reading 101

Orderbook reading is a key component of my trading toolbox. It is a technique I developed myself, back in 2014, and one that there is little-to-no quality information on online. (Seriously, Google “orderbook reading” and you’ll be shocked by the lack of resources.) In my book, I dedicated an entire 5,000-word section to orderbook reading, and, given the lack of material readily available elsewhere, I figured that it might be useful for anyone interested in the technique to have a primer written; if you find orderbook reading compelling, you can take a look at the more advanced material in the book.

Of course, I don’t doubt that there were others who had dissected the orderbook in a similar way to myself prior to 2014, and I don’t take any credit for the concept of orderbook reading; but it is the one technique that I learnt entirely via my own experience, with no help from resources such as those one would look to when learning other aspects of technical analysis. Now, I say it is an aspect of technical analysis, though strictly that isn’t true. TA is exclusively concerned with the chart, but orderbook reading resides in that grey area between fundamentals and technicals. The technique itself feels more like technical analysis, intuitively, but given that it is not derived from the chart, the analytical grey area is where it will remain.

So, what is orderbook reading? In short, it is the study and subsequent analysis of the ledger of orders for any given market. Orderbooks contain a list of all the buy and sell orders currently placed within a market on a specific exchange, and it is this transparency on which I learnt to capitalise. By studying the orderbook, one can often find clues as to the plans of those manipulating price. This can add confluence to our technical targets for entries and exits, as well as boost confidence in the validity of our positions.

But what’s the process? Below, I’ve outlined a breakdown of what I tend to look for when studying an orderbook:

  • There are three key components to orderbook reading: Order Depth, Order Patterns and Time. In this post, we will be looking at the first two.
Order Depth:

This is the simplest form of orderbook reading, and is predicated on studying the value of the orders in the bid and ask (buy and sell) sides of the orderbook.

  1. Pick any given market on any given exchange. Ideally, you’ll want to use exchanges that provide the most transparency with their orderbooks. Bittrex and Cryptopia are two exchanges that I like to use.
  2. Calculate the total value of orders in both the altcoin itself and Bitcoin for the bid-side and ask-side of the orderbook. For example, Vertcoin on Bittrex currently has ~20 BTC-worth of orders in the bid-side, totalling ~28,800,000 VTC. It has ~9.5 BTC-worth of orders in the ask-side, totalling ~100,000 VTC.
  3. At this point, you want to calculate the average order size in both sides of the orderbook. Using the above values, the average buy order is 69 satoshis. This doesn’t make a lot of sense, at surface-level, and we’ll come back to it in a second. The average sell order, however, is 9500 satoshis, or ~30% above current prices.
  4. A quick thing to note before we continue is that, from this point, one can also make another surface-level analysis based on the total Bitcoin values of the bid and ask sides: given 20 BTC-worth of buys and 9.5 BTC-worth of sells, it is evident that there is greater demand than supply. This, as mentioned, is surface-level, and does not account for orderbook manipulation, which I won’t go into here.
  5. Returning to our average order values, why is it that the average buy order for Vertcoin is so low relative to current prices? If we inspect the latter pages of the orderbook, we find a buy order worth 1.13 BTC at 4 satoshis, totalling ~28,200,000 VTC. This explains everything. At your discretion, you can then discount this order from the calculations. Doing so would give us ~600,000 VTC remaining in the buy side at a total of ~18.87 BTC, which equates to an average buy order of 3145 satoshis. Far more insightful a figure.
  6. Now, the ask-side calculation also comes with a caveat. Despite there being less than 10 BTC-worth of orders listed, we can see, if we look to the last page shown, that we are only being shown 20 pages of data – in this case, orders up to ~11,000 satoshis. Undoubtedly, there will be orders above that point, but this is a disadvantage of Bittrex; it only shows 20 pages of orderbook data for either side. Cryptopia, and many other exchanges, offer full transparency of the orderbook.
  7. Another point to note is that these values are dynamic, as orderbooks are dynamic. Consistent monitoring and study is required in order to garner a more accurate understanding of order depth.
Order Patterns:

Order patterns are perhaps the most complicated aspect of orderbook reading, as they illuminate much of the manipulation that occurs in altcoin markets. Whilst I won’t go into the more advanced stuff here, it is significant to highlight the four types of order pattern: clean orders, bot orders, walls and, for lack of a better term, non-clean orders or messy orders.

Clean Orders: a clean order is simply an order that is unusually perfect, mathematically. These are often orders comprising of multiples of 5s or 10s that occur at regular intervals in the orderbook. For example, buy orders of 50,000 Vertcoin at 5000 satoshis, totalling 2.5 BTC, with corresponding orders at 1000-satoshi intervals. These are indicative of significant price-levels for the market-maker; perhaps levels at which accumulation is taking place. On the ask-side, you may find similar patterns of clean orders that are set up as future targets for price-action. Again, this is simplistic, and not taking into account orderbook manipulation, but it is obvious to see how this can be useful when looking for entries and exits.

Bot Orders: a bot order is one that defies human ability in its execution. Often, algorithms are in place to bid up a buy order or push down a sell order. This is easily noticeable and you have likely experienced it yourself. Recall a time when you’d place a buy order at the top of the orderbook, only for it to be displaced within milliseconds by another buy order; this is a bot order. The purpose of these orders can be two-fold: either to drive you to make irrational decisions (as we’ll discuss in the next section) or to beat you out for the purposes of active and immediate accumulation or distribution.

Walls: Most are familiar with walls. A wall is an unusually large order in the orderbook. However, most tend to react in the most irrational manner when it comes to these orders. Large buy orders being pushed and pulled near current prices are often viewed by the masses as the perfect time to enter, lest one miss out on what must be an imminent bullish move. Large sell orders are viewed as the perfect time to exit, lest we lose all our money when the market crashes and burns. This is orderbook manipulation at its finest. Buy walls are an attempt at causing irrational market participants to buy up the orderbook, and vice-versa  for sell walls. This allows for better pricing to be had for the puppet-master for both accumulation and distribution purposes.

Non-Clean Orders: much like clean orders, non-clean orders are about determining symmetry and patterns in the orderbook. However, they are more difficult to notice, as, unlike their clean counterparts, they are comprised of seemingly random numerical values. For example, an order of 24879.284733 Vertcoin at 6435 satoshis might appear in the orderbook. This order would seem irrelevant to most, but perhaps you notice another order of that exact same amount at 5733 satoshis… and another at 5210 satoshis. Coincidence? I think not. Non-clean orders, such as these, are simply a more discrete way for market manipulators to mark out significant levels for future reference.

Note these various orders down as and when you come across them – and you will come across them – and you’ll begin to piece together a trail of footsteps that must be left by those manipulating price.

I hope this post has proved somewhat insightful and piqued your interest in orderbook reading. For anyone compelled to learn more, there is an advanced section in my book that goes into far more detail.


If you’ve enjoyed this post and want to receive new posts straight to your inbox, I’ve set up a RSS-to-Email feed that will be sent out weekly; every Monday, 12pm. Just submit your email and I’ll make sure you’re included in the list. Cheers.