Market Mechanics – Part 6, Market pitfalls

Part of a 6-part series on market mechanics:

  1. Introduction
  2. Functions of an exchange
    1. Custody
    2. Settlement
    3. Tax reporting
    4. Matching
    5. Market data
  3. Types of exchanges
    1. Order book
    2. Over-the-counter (OTC)
    3. Derivatives
  4. How order books work
    1. Mechanics
    2. Types of orders
      1. Market
      2. Limit
      3. Stop
    3. Matching
  5. What’s in a price?
  6. Market pitfalls
    1. Slippage
    2. Market cap
    3. Hidden supply
    4. Flash crash

This is a brief list of unexpected behavior that can sometimes happen because of the structure of markets.


Slippage was mentioned earlier as a pitfall specifically affecting market orders.

On a broader level, though, slippage affects anyone who wants to buy or sell a large quantity of a particular asset. The reason is the structure of the market: if the price of Bitcoin is $8000 and there are currently 10K Bitcoin for sale in the order book, that doesn’t mean you can buy all 10K Bitcoin for $8000 each, nor does it mean that there’s $80M worth of Bitcoin for sale. Rather, you can buy your first 5 BTC for $8000, then another 0.2 BTC for $8000.10, then another Bitcoin for $8000.12, and so on. A number of sellers might be optimistically holding out for $1M Bitcoin before they sell. The total amount of Bitcoin for sale therefore might be more like $200M or more.

Limit orders do not help you get around this. A limit order guarantees you a price, but it does not guarantee execution. If you were to break up your your $80M order into 10K limit orders of 1 BTC each, only the first 5 would execute, and then the price would be $8000.10. The rest would sit in the order book until the time when Bitcoin is back under $8000, which is potentially never.

The way that investors in the public stock markets deal with this is to enter large positions gradually over time. There’s a continuous flow of new orders into the market, and most investors expect that the price will be relatively stable. So an institutional investor that wants to buy that $80M of Bitcoin might put in a limit order for 5 BTC @ $8K, and then wait a few minutes for the market to recover. Then they put in another order, and wait, and so on. If they’re smart about it, they also won’t put all orders in at the same price – they’ll scatter them by a few dollars in either direction and make a bunch of small purchases, so it looks like random retail investors trading instead of an institution taking a big position.

Most crypto investors do not do this, although it is common in the stock market.

Incidentally, slippage is commonly overlooked in a number of discussions about high-net-worth individuals. For example, Jeff Bezos is the richest man in the world, with a net worth of about $103B. And people frequently ask “If you’re so rich, why can’t you just donate $90B to education and increase the average salary of the nation’s 3.2M teachers by 50%?” The thing is – Jeff Bezos does not own $90B in cash. He owns 59M shares of Amazon, worth $103B at current market prices. If he were to sell all of his Amazon shares at once, it would tank the price to zero: only 3M Amazon shares are transacted in all of the public markets on a typical day. The actual amount of cash that Jeff Bezos can access instantaneously is much, much lower than his net worth.

The same situation applies to cryptocurrency whales who got rich because they own large quantities of a coin that appreciated significantly. They have large net worths on paper, but if they were to sell all of that coin at once, there’s not enough buyers in the market to get anywhere near their paper net worth.

Market Cap

A related pitfall applies when measuring the total worth of a cryptocurrency. The most commonly cited number for that is the market cap: it’s the current price * the total supply of the coin.

Prices are volatile – they’re set on the margins, by a small number of people who happen to be trading that day. And the market cap multiplies that out by all coins ever minted, which often includes large amounts held by HODLers indefinitely or by the development team.

That lets a small number of people have a big impact on the market cap. Bitcoin, for example, has a market cap of $139B at a price of $7800. When the Bitstamp flash crash happened, a single seller with $34M of Bitcoin managed to tank that to $110B, a loss of $29B in market cap within 10 minutes. The total dollar value transacted was less than 1/1000th the drop in market cap.

This isn’t to say that market cap is useless – it can be a convenient shorthand to encapsulate roughly what people believe a cryptocurrency is worth, and is also handy with fundamental analysis strategies that try to determine if the current price is worth it based on other non-price metrics. But it’s worth being aware of other metrics like volume and order book depth so you can check to make sure that large changes in market cap are actually reflective of changes in the coin’s worth.

Hidden Supply

Just as the current price reflects only the most recent trade and ignores everything in the order book, the order book only reflects traders who have active orders out. The majority of people who own cryptocurrency are not actively trading it: they might be developers sitting on coin reserves, HODLers who are waiting for it to become the dominant payment method, wealthy individuals holding crypto for diversification, or just retail investors who bought some in the last bubble and forgot they owned it.

Normally, individuals who don’t trade have no effect on the price. However, there are a few cases when people might suddenly realize that they own Bitcoin and want to sell it:

  1. When the price goes up suddenly and someone burned by the last bubble decides that they’ve had enough and ought to cut their losses while they can.
  2. When the price goes down suddenly and existing HODLers get scared about crypto’s long-term future and want to exit immediately.
  3. When a whale or market manipulator drops a lot of crypto on the market at once, crashes the price, and people waiting on the sidelines decide that it may be a good buying opportunity.
  4. When there’s a major news story – either positive (eg. Facebook launching their cryptocurrency) or negative (eg. an exchange declaring bankruptcy) – and people on the sidelines decide that it may be time to buy or sell.

We’re in situation #1 right now: a large number of retail investors bought for prices between $7K and $13K during the December 2017 bubble. Bitcoin had a rapid run up from $8K to $14K in the second half of June 2019 – and then promptly fell down under $10K. Then it had another run up to $13K in July – and then fell back to $9K. Then it had another run up to $12K in August, and fell back to $9500, and a run to $10.5K in September and fall to $8K.

Every time there’s a rapid increase in price, the news media covers it, which reminds buyers from 2017 that the asset that they lost 80% of their money on is close to break-even again. Some may ignore that, but others decide that they don’t want to go through that again and hit the exchange. At some points the order books have looked very bullish, but as soon as the price hits the news media, sellers come online and the rally ends. There won’t be a sustained rally until everyone who was traumatized by early 2018 gets rid of their Bitcoin.

Flash Crash

With slippage, the price you get for a large sell may be nowhere near what the price was when you initiated the trade. What happens when there’s an extreme version of this, and there’s actually a gap in the order book. What if, for illustration, the order book for Bitcoin looked like this:


And then somebody puts in a market order to sell 100 BTC?

The answer is that the price would crash to $10, and one lucky person would get a whole lot of really cheap Bitcoin. The order would first eat up all bids > $7K, for a total of 40 BTC. But there’s still 60 BTC left in the market order. 5 of them eat up the price levels between $4500-5000, and then there’s nothing left in the order book other than the $10 bid. So that’s what gets filled.

Flash crashes are rare in heavily-traded markets like BTC/USD or ETH/USD, because there’s enough liquidity in the order book to absorb most large orders. The most you’ll see are things like the Bitstamp flash crash mentioned above. They have happened in the past, however, particularly in pre-bubble times (like the June 2017 Ethereum crash that took it from $319 to $0.10) or on less-common trading pairs on smaller exchanges (like the June 2019 BTC/CAD crash on Kraken that took it to $100).

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s