The primary market dynamic that favors trading bots is low . In low condiions these bots effectively can dictate price by "entraining" buyers or sellers and other bots. For example, the idea of a "sell wall" or "buy wall", that is, a large cluster of orders that one can see in the order book or on the order book graph, will influence other buyers and sellers as you will either be "entrained" or you will "get out of the way." Both of these scenarios present problems as the optimal outcome may not be clear if you're under pressure to make a decision that hasn't been planned out according to an overall understanding of market conditions. Not understanding fundamental market dynamics is a key reason fails. Charts won't actually tell you exactly what is happening in a market with data. It is usually suggesting a direction and often only telling you something important after it's too late (see for example how bad is at prediction).
But I'll talk more about the mechanism of how algo bots work later. If you want to read my streamed commentary you can see a recent piece that I published about a greater than 50% probability of another decline where I slowly built my (admittedly still weak) understanding of how these bots influence price action. And if you're skeptical about what I'm describing then I'm open to constructive feedback or corrections. But I'm going to cut to the chase while skipping over some of the technical aspects.
In the current market we're looking at what appears to be a strong recovery from a 31% decline from $8400 to $5800 (on GDAX) and a retrace of 58.3% of that price movement.
Given the first wave of price movement off of the ATH took 26 days to hit a reversal and the second wave took 15 days, the current reversal will likely be very soon, as we are about 5 days past the last reversal (Feb. 2) today, near the end of the day Feb. 6/start of Feb. 7.
In order to signal a reversal this move needs to hit $8k (low of the previous wave on GDAX prior to reversal) and ideally it goes above $8.4k which was the high point of the current draw down to fully signal a potential (but improbable) move higher.
The fundamental reason for momentum (i.e. a pump) is that buyers as an aggregate can see when there will be sufficient momentum for a price movement and they are "entrained" so as not to be left behind. This is basic pack/herd behavior as I would surmise is understood by evolutionary game theory ( not going to get into that ). Those who choose to stick with the herd have higher probability of survival/success. Venturing away from the herd makes you more likely to become a victim of your "individuality".
Clearly, the market is positioning for selling more than buying. This kind of signaling plays a role in warning buyers to enter the market with caution (at least for longer duration trades). For example, if someone is quite eager to sell you a car at a great price, you might think, what's wrong with this car? And in this instance, we should think, what's wrong with this market? Clearly, there is fear. And then there is also now a conditioned expectation that another huge sell off could happen at anytime. This makes for enough weak hands for more .
The ONLY way this could be reversed is if some HUGE buyers enter the market. So let's say a bunch of millionaires all collude to go and buy Bitcoin so the price goes up.
So the fundamental idea here goes:
1) we're not seeing enough momentum / order positioning to break the $8k SELL WALL (soon to be $7.5k SELL WALL)
2) algo trading bots will take advantage of declining volume and momentum to push prices back down to a level where a lot of other buyers are positioning themselves given the right conditions (i.e. near $8k) (BEWARE OF WEEKENDS and OVERNIGHT). Once the "sell train" starts rolling you either get out of the way or get killed. Even Satoshi Nakamoto couldn't stop a huge sell train.
3) Positioning yourself with THE HERD will protect you from entering or exiting the market at the wrong time. Look at order books/graphs to figure out where the BEST place is for a longer duration trade amongst the largest cluster of BUYERS/SELLERS. You can also use this for short duration, smaller order size trades profitably. I feel like this actually works better in a downward trending market as it can be hard to beat buy and hold in an upward trending market. Obviously you won't profit if you hold during a downtrend but these short duration trades can offset some losses in your longer duration portfolio if you do it correctly. (BIG IF)
4) We're in the very near term end of a bearish price move that could be like a descending wedge we'll likely continue to see massive volatility at increasingly short intervals. Traders are conditioned now to be extremely sensitive to a rapid decline in price. Given equity market volatility this could all be contagious to risk aversion in markets overall. So be warned.
5) That being said, the overall risk:reward for Bitcoin/Litecoin and a few others appears to be very strong at current prices. So in theory you could do nothing for 6 months at this price level and have a high probability of making money. So if you're in a position started at $7.5k or less I would say that's pretty good. We might see a final draw down to 50% of that, but basically the sky is still the limit for Bitcoin prices over the 2-4 year time horizon.
I would set aside 25% of your trading portfolio for a move to less than $5k and try to enter with at least 50% at $5.5k (or the equivalent for some other coin that you think will bounce, like BCH, ETH, or LTC). Then you still have another 25% set aside for the sake of risk management/contingencies. This will depend on what the order book is telling me so that can change at any time.
ETH had an incredible bounce and I fully expect any draw down to below $500 could reverse for 100% gains over a short duration. But I haven't looked at the order books or anything so that's just a guess. And Ethereum also has relatively more regulatory risk related to it compared to Bitcoin.
Also, observing Bitcoin Cash (BCH) it is VERY obvious that an algo bot is pumping the price with an unusual 350 BCH order that bounces around in the BID order book somewhat close to the market price but not too close creating the illusion of demand. This is by far the largest order on either the buy or sell side of the order book and clearly has the ability to influence other bots/traders into aligning around its behavior.
All of the order volume above $6.5k is roughly equal to the order volume positioned between $5.5k and $6k. And the order volume between $6k and $6.5k is maybe 20% less than that between $5.5k and $6k.
So about 1/3 of all order volume above $5k is set between $5k and $5.5k, another 24% is above $6.5k and another 43% set between $5.5k and $6.5k in a fairly linear distribution.
The bot is very obviously inflating the buy side order volume by just enough to get the price towards $7800 where there appears to be enough selling volume to push the price back down.
However, this bot is still clearly at work, throwing in and cancelling orders at just the very edge of the order book graph when you're zoomed in to keep the volume from dropping off completely.
It will be interesting to see how this plays out.
I just don't understand how the price is staying up like this despite almost all of the buying volume drying up. If the algo bot intends to hit a price target then it wouldn't want to signal that price target unless it wants to attract more sellers to it. That would make it harder to reach the price target. So then maybe it's just a general pressuring of buy volume without actually any buying, which could eventually exhaust a large sell wall and allow price to creep higher to an even larger sell wall/price target. I suppose if you're able to build enough momentum on the buy side then the volume will increase and more buyers will buy into the FOMO momentum. On declining volume maybe this doesn't work.
I just saw the price skyrocket 1.6% in 5 minutes on almost no sell volume simply due to these small orders. Clearly, a rational human trader isn't setting tiny little sell orders at weird prices throughout the order book. I wonder how much activity happening in exchanges is "real" and how much is created by bots.
So this would explain why areas of the order book in between "notable" numbers like $8k, $8.5k etc. could see very little support or resistance due to the ease in which algo bots can "seed" the order book to facilitate price movement.
Shanghai SE Composite is down 1.82% and the Hang Seng down 0.89% so if this global volatility continues it could trigger another quick reversal back to previous lows. It will be interesting to see if global equity markets volatility has any effect on crypto. Judging by the reversal today I would guess not.
That would be extremely interesting if it were to happen. During the last period in November 2015 when prices dipped this low it took over a month for prices to recover and then dip again before finally reversing fully. But that was at least 80% off the ATH and the 80% off the ATH level would be around $4k. So not exactly a great sign if we're at $8k right now and still about 10X above the low from a year ago.
And as I write this the volume above 6k magically increased by 30% so now it's even more bizarrely shaped.
Not sure if we're getting some major FOMO positioning or what.
So on GDAX it would be about 20 M, 10 M, 30 M for this chart. In a linear demand world we'd be seeing something more like 30, 20, 10 (hypothetically and not precisely) which would be consistent with yesterday where we saw 20, 15, 20 (i.e. 10, 5, 2.5 if you distribute it the same).
I would expect prices to bounce around 7250 based on the demand. There appears to be unusually sized orders placed between 7160 and 7250 making the order volume balloon in that area. It's the same between $8300 and $8500.
I wonder if this inflated order volume is an attempt to manipulate the market. Overall Bitcoin "Direct Volume for the last 24 H) has dropped 30-40% so whatever enthusiasm / momentum occurred yesterday and the day before has not carried over at all. If we see another lull with volume dropping by another 50% then that will be a sign that the markets could be easily manipulated.
If you look at GDAX order book with orders aggregated to $10 intervals (the + button) you will see at $980-$990 order volume spiking and then going lower (with no other numbers moving). Clearly someone is using a bot to place and cancel large orders to create the illusion of more buying volume than exists. In regulated markets this is illegal.
This article is pretty thorough (and where I got the link above):
I feel like we're very close to "capitulation," but perhaps we'll continue to see medium term strength in the markets due to how much some people have lost so far (i.e. denial). I would be extremely strict about portfolio allocation, with no more than 20% in long duration crypto assets and no more than 5%
I just don't know where this volume comes from because there isn't much order positioning and buy volume looks terrible still...
Also, Tethers were supposedly printed so if we start to see a big + spread on some of the Tether exchanges then that could be a possible sign of wash trading with fake money...
For example if I own 80% of all XRP trading in S. Korea (or collude with a large enough percentage of XRP owners that we control over 50% of the volume) that I bought at less than 0.10 I'm going to have extreme price control in the market. This is an extreme hypothetical scenario. Essentially I could have a monopoly simply by buying more XRP by pressuring the market price lower (as long as you buy more than you sell) and then command higher prices by restricting supply to increase bids while pumping XRP on social media, etc. This model is likely completely automated by many of the trading groups/individuals who use bots for both trading and social media.
This price manipulation will lead some to believe in the correlation between XRP news, positive price action, and social media sentiment and then the FOMO positive feedback loop begins as people gamble on crypto believing they can become insta-rich even though they are "noobs" and don't understand any of this stuff.
If you are a savvy pump and dump schemer this is basically free money. You take your trading fund and buy up as much of some crypto as possible when it is very cheap and you believe there will be positive sentiment at a later time. Then you deploy bots for trading and social media presence (easily sourced through internet freelance sites) and maybe even build your own content site to pump the coin.
Ideally you want a limited presence on a handful exchanges, like with an ICO token, so you can corral enough volume on those exchanges to start manipulating prices with your huge stack relative to the total volume being traded.
This is essentially every ICO and pre-mine/insta-mine coin out there and has been documented here:
On another note, on-chain transactions for Bitcoin have only been averaging 203,000 per day over the last 6 days. Not a good sign if we're thinking about Bitcoin as a network that gains its value through growth. Even if it is a store of value you would think that the network is growing rapidly enough to see an overall increase in on-chain transactions over time, but instead these numbers have more or less peaked and declined to the same levels as late 2015.
What we're seeing is an increase in transaction volume volatility, which might be a sign that a lot of people are going in and out of the market, so instead of keeping a stable value, we're seeing these massive price changes that reflect extreme uncertainty in the market about the current valuation. This would seem to be a sign that the market could stabilize after a fiery and unpleasant decline that results in an extended period of relative calm.
Looking back at Nov. 2013 to late 2015 you can clearly see transaction volume spike (from the price manipulation of Mt. Gox / bots) and then decline siginificantly. After this crash transaction volume stayed fairly steady for about 6 months before increasing steadily through late 2015 when transaction volume volatility began to return to the market.
During this period many companies which had organized to enter around the time of the big "boom" in crypto flamed out and by 2015 the market was probably finally starting to recover enough investment interest that transactions picked up significantly.
Given the current state of the market I would be very cautious about taking any long duration positions with the expectation of hitting a 6+ month price target. It's highly likely that we'll be seeing a bear market where prices decline back to previous levels and until exchanges signal that they've leveled the playing field by weeding out bad actors the market could get worse before it gets better.
Since I've decided to sleep more during peak algo bot trading hours when volume is low I haven't really noticed much suspicious activity lately. There still appears to be what I think is layering but not really any spoofing and probably less wash trading too as I'm not seeing these rapid changes in price through thin sections of the order book.
Given the charting suggests we need to visit $5k before a reversal I could see a reverse head and shoulders that forms over the course of early March. This would be a good sign that the market has stabilized.
However, given Bitfinex's sketchy ass position in driving price discovery and the very distinct possibility that they have essentially been repeating the mistakes of Mt. Gox, that is, 1) get hacked and lose a ton of money 2) fake a bunch of price action so as to keep your business running 3) fake print a bunch of crypto and allow sketchy market cap cryptos (aka Bitcoin Gold) to be traded to inflate your valuation / volume 4) market loses faith based on big negative news/sentiment, volume and price drops off considerably and you go bankrupt.
I'm not seeing rampant bot action and a pretty calm market so now I'm feeling like maybe I've blown this all out of proportion... but in general, when you look at the systemic risks of the global crypto market, this is still a ticking time bomb. Perhaps Coinbase is getting their act together. Until Bitfinex has opened themselves up to being transparent about their finances and exactly what is going on with Tethers there's a lot of reason to be concerned about current valuations.
The weekends are a dangerous time to hodl so I'd watch what happens during the weekend, especially when volume gets low, and see if these trading bots start trying to manipulate prices up or down.
One thing I haven't discussed is how very large coin holders have invested or directly own significant media businesses in the crypto space. This allows them to coordinate both trading price manipulation as well as sentiment manipulation (which you could say is much harder to do but still possible given enough noobs visiting, setting up wallets, trusting the "bitcoin.com" name, etc.).
In general a lot of the key activity happens during low volume trading hours, which is when I like to trade/observe the market. Given that equities markets didn't blow up today that's generally a good sign.
I believe crypto doesn't function as a "haven" like gold, which makes the idea that it will replace gold fundamentally unsound. I think it works like gold as a hedge against central bank activity, but in the opposite way, which is as a speculative store of value instead of a safe store of value.
So when central banks drag their interest rates to nothing people start looking for higher returns and increase their risk appetite. This is where crypto becomes attractive. When interest rates increase the appetite for risk also decreases.
Given that interest rates are expected to increase sharply and risk appetite by investors will decrease sharply as well, I think the overall crypto space will trend bearish over the same duration in which we see these central bank unwinding of QE, interest rate announcements, etc. This will be the first time since Bitcoin was created where there will be a large enough volume of activity and a sharp enough change in central bank sentiment to get a sense of whether or not there is correlation. Given the sharp disparity between Japan and the US over central bank policy we could see a really interesting divergence play out, but that seemed to have been the case at least before the big correction, when there was a premium in many Asian markets, but now that seems to have reversed.
Look forward to your ongoing commentary on what you are seeing unfold. Cheers
Also check out his other articles. This guy has really done some great research, but knowing what's going on in the markets anyone participating should be deeply concerned about systemic risk.