4 ms·

> suggest a lack of understanding about what normal exchange activity can look like. So you are suggesting that the experts who were cited in the article, whos

> suggest a lack of understanding about what normal exchange activity can look like. So you are suggesting that the experts who were cited in the article, whose field of expertise is this very thing, do not understand how markets work? Meanwhile you don't seem to understand why it is remarkable that the market can absorb a huge order that should move the market because of its sheer size while much smaller order have a much larger impact. When a buy order of quantity 37.5 moves the price by 0.0002, but the following buy order of quantity 13,076.389, that is almost 350 times the size of the previous order, moves the market by half of that, something is very strange about the liquidity situation.

arisAlexis 8y ago
yes it's pretty common for "experts in crypto" to write complete nonesense just becUse there are no experts in this field yet
lend000 8y ago
These "experts" are just journalists who wouldn't be writing articles if they had enjoyed any success analyzing orderbooks. This article is perhaps the most clueless garbage I've read on the topic of crypto, and there is a LOT of cringeworthy stuff out there. What's that saying about how people scoff at the inaccuracies when reading about something in the news they know intimately, and then read the rest of the paper as though it's going to be any better? > Meanwhile you don't seem to understand why it is remarkable that the market can absorb a huge order that should move the market because of its sheer size while much smaller order have a much larger impact. Is the concept of an orderbook so foreign to non-finance/math people? Firstly, 13k is not a "huge" order. Secondly, someone clearly just had a somewhat large maker order placed at $1 for the example cited which absorbed a few orders, and the book was thinner for a subsequent price movement (liquidity vacuum). Explained by the market being low volume, low volatility, likely dominated by a handful of traders/algos, and plain old random variation (individual players' orders have a more significant effect in low volume markets like Tether/USD). It's also important to note that psychological price levels (like 1.0) tend to accumulate larger walls as a result of human nature. For Tether in particular, it makes sense that the price never breaches 1.0. It's an easy short above 1.0, as it is impossible for Tether to ever become worth more than the price it is pegged at, whereas it could become worth less if it were insolvent. So there is probably a huge wall sitting there on every exchange.
- neom 8y ago
  The part I'm confused about is if you look at order size to volume, isn't it weird that price flux doesn't seem to match the spay in order size? Also, is tether formally a fixed-exchange-rate system, if so how is the pegging being mechanized?
  - lend000 8y ago
    You can't read into it much for such a low volume exchange. It's akin to a small sample size resulting in behavior that can stray farther from the mean. This article also makes a glaring omission of using the Tether/USD index price (since Kraken is not the only exchange trading the pair), so prices can move in the market without any volume occurring on Kraken, if the traders shuffle the orderbook before arbitragers make trades.
- wpietri 8y ago
  Did you read the article? These experts are experts. Including: "New York University Professor Rosa Abrantes-Metz and former Federal Reserve bank examiner Mark Williams"
  - lend000 8y ago
    Those references are not the authors, and keep in mind that professors usually would not be professors if they had a sufficient understanding of markets to beat them. No one cited in the article has seen a live exchange with as low of volume as Tether/USD, simply because such low volume exchanges are typically considered irrelevant. Lower volume == fewer players == higher impact of a smaller number of traders == attribution of the inefficiencies of a small group of traders to market-wide patterns == clueless articles like this one.
    - wpietri 8y ago
      They are experts. The article is written by skilled financial journalists based on the work of experts. I used to write trading systems for market makers, and still have friends in the business. I also have friends who are professors, including a finance professor. Trading is not really about market knowledge; it's a skill and an orientation. Being a professor is much more about knowledge, but it's also about skills and orientation. They're different kinds of people, doing very different work. Your "the only people who can possibly understand this are the ones who agree with me" approach is essentially religious thinking. You're making an argument from authority, except the only cited expert is you, an anonymous person with no credentials and no provable experience.
lowdest 8y ago
I'm suggesting one or more of the authors and whoever proofread this article before publication do not understand limit order books, and they therefore don't understand this particular market at the micro level. Everything they have highlighted in the red and green graphs in the latter half of the article can be explained by a completely ordinary order flow. Much weirder stuff happens in the order books every day: iceberg orders that appear small but are bottomless, giant walls that retreat as soon as they're touched, spoof orders that are placed and canceled repeatedly. None of that funny stuff is in evidence here. Let's focus on one particular error from the article and I'll explain the misunderstanding behind it. Take a look at the diagram titled "Tether-USD Trades on May 9 From 20:59:13 to 21:01:00 GMT." If you don't understand the two prices of a limit order book, you might think that diagram shows 7 different price movements. But in fact, there are only two price movements: when the Ask price moves from .9991 to .9992, and when the Ask price moves from .9992 back to .9991. Something to remember about limit order books: market buy orders will execute at the Ask price, which is at least one increment higher than the Bid price. Market sell orders will execute at the Bid price, which is at least one increment lower than the Ask price. Why can't the Ask and Bid prices be the same? If you had a limit buy order on the books at $100 and a limit sell order on the books also at $100, they match to each other, cancelling each other out until only one type of order (buy or sell) is left at the contested price. When looking at a price history graph like this, what is actually plotted is the Last Executed Price. If the Bid and Ask prices are completely static, the price history will show oscillation between the Bid and Ask prices as market buy orders and market sell orders are executed in turn. The gap between the Bid and Ask prices at a point in time in the order book is called the Bid-Ask Spread. An important point is that when the Bid and Ask are not moving, the Last Price graph will still show oscillation across the Bid-Ask Spread. So now look again to the diagram caption: "Multiple sell orders of exactly 13,076.389 are executed, and the price didn’t budge..." What this means is that there was more than 183,076.46 in buy limit orders at the Bid price of $0.9990, which is fairly typical to witness. Certainly not evidence of manipulation or even out of the ordinary. The second caption "...but a buy order of just 75 moved the price up .0001" is more problematic. Neither the Bid price nor Ask price has actually moved in response to the size 75 order, this is just the normal oscillation across the Bid-Ask Spread. Furthermore the minimum price increment on Kraken is 0.0001, so the minimum any buy order following a sell order can move the Last Price is 0.0001 (barring a simultaneous drop in the Ask price.) Let me repeat: Unless the Ask price drops, any and all buy orders following a sell must move this price graph upward by at least .0001 by definition. This appears to be lost on the authors, who highlight the .0001 move upward as something notable, an outsized price move for such a small order. They might be shocked to discover that a size 0.01 order would also move the price graph upward as much as the size 75 order.

arisAlexis 8y ago

yes it's pretty common for "experts in crypto" to write complete nonesense just becUse there are no experts in this field yet

lend000 8y ago

These "experts" are just journalists who wouldn't be writing articles if they had enjoyed any success analyzing orderbooks. This article is perhaps the most clueless garbage I've read on the topic of crypto, and there is a LOT of cringeworthy stuff out there. What's that saying about how people scoff at the inaccuracies when reading about something in the news they know intimately, and then read the rest of the paper as though it's going to be any better? > Meanwhile you don't seem to understand why it is remarkable that the market can absorb a huge order that should move the market because of its sheer size while much smaller order have a much larger impact. Is the concept of an orderbook so foreign to non-finance/math people? Firstly, 13k is not a "huge" order. Secondly, someone clearly just had a somewhat large maker order placed at $1 for the example cited which absorbed a few orders, and the book was thinner for a subsequent price movement (liquidity vacuum). Explained by the market being low volume, low volatility, likely dominated by a handful of traders/algos, and plain old random variation (individual players' orders have a more significant effect in low volume markets like Tether/USD). It's also important to note that psychological price levels (like 1.0) tend to accumulate larger walls as a result of human nature. For Tether in particular, it makes sense that the price never breaches 1.0. It's an easy short above 1.0, as it is impossible for Tether to ever become worth more than the price it is pegged at, whereas it could become worth less if it were insolvent. So there is probably a huge wall sitting there on every exchange.

neom 8y ago
The part I'm confused about is if you look at order size to volume, isn't it weird that price flux doesn't seem to match the spay in order size? Also, is tether formally a fixed-exchange-rate system, if so how is the pegging being mechanized?
- lend000 8y ago
  You can't read into it much for such a low volume exchange. It's akin to a small sample size resulting in behavior that can stray farther from the mean. This article also makes a glaring omission of using the Tether/USD index price (since Kraken is not the only exchange trading the pair), so prices can move in the market without any volume occurring on Kraken, if the traders shuffle the orderbook before arbitragers make trades.
wpietri 8y ago
Did you read the article? These experts are experts. Including: "New York University Professor Rosa Abrantes-Metz and former Federal Reserve bank examiner Mark Williams"
- lend000 8y ago
  Those references are not the authors, and keep in mind that professors usually would not be professors if they had a sufficient understanding of markets to beat them. No one cited in the article has seen a live exchange with as low of volume as Tether/USD, simply because such low volume exchanges are typically considered irrelevant. Lower volume == fewer players == higher impact of a smaller number of traders == attribution of the inefficiencies of a small group of traders to market-wide patterns == clueless articles like this one.
  - wpietri 8y ago
    They are experts. The article is written by skilled financial journalists based on the work of experts. I used to write trading systems for market makers, and still have friends in the business. I also have friends who are professors, including a finance professor. Trading is not really about market knowledge; it's a skill and an orientation. Being a professor is much more about knowledge, but it's also about skills and orientation. They're different kinds of people, doing very different work. Your "the only people who can possibly understand this are the ones who agree with me" approach is essentially religious thinking. You're making an argument from authority, except the only cited expert is you, an anonymous person with no credentials and no provable experience.

lowdest 8y ago

I'm suggesting one or more of the authors and whoever proofread this article before publication do not understand limit order books, and they therefore don't understand this particular market at the micro level. Everything they have highlighted in the red and green graphs in the latter half of the article can be explained by a completely ordinary order flow. Much weirder stuff happens in the order books every day: iceberg orders that appear small but are bottomless, giant walls that retreat as soon as they're touched, spoof orders that are placed and canceled repeatedly. None of that funny stuff is in evidence here. Let's focus on one particular error from the article and I'll explain the misunderstanding behind it. Take a look at the diagram titled "Tether-USD Trades on May 9 From 20:59:13 to 21:01:00 GMT." If you don't understand the two prices of a limit order book, you might think that diagram shows 7 different price movements. But in fact, there are only two price movements: when the Ask price moves from .9991 to .9992, and when the Ask price moves from .9992 back to .9991. Something to remember about limit order books: market buy orders will execute at the Ask price, which is at least one increment higher than the Bid price. Market sell orders will execute at the Bid price, which is at least one increment lower than the Ask price. Why can't the Ask and Bid prices be the same? If you had a limit buy order on the books at $100 and a limit sell order on the books also at $100, they match to each other, cancelling each other out until only one type of order (buy or sell) is left at the contested price. When looking at a price history graph like this, what is actually plotted is the Last Executed Price. If the Bid and Ask prices are completely static, the price history will show oscillation between the Bid and Ask prices as market buy orders and market sell orders are executed in turn. The gap between the Bid and Ask prices at a point in time in the order book is called the Bid-Ask Spread. An important point is that when the Bid and Ask are not moving, the Last Price graph will still show oscillation across the Bid-Ask Spread. So now look again to the diagram caption: "Multiple sell orders of exactly 13,076.389 are executed, and the price didn’t budge..." What this means is that there was more than 183,076.46 in buy limit orders at the Bid price of $0.9990, which is fairly typical to witness. Certainly not evidence of manipulation or even out of the ordinary. The second caption "...but a buy order of just 75 moved the price up .0001" is more problematic. Neither the Bid price nor Ask price has actually moved in response to the size 75 order, this is just the normal oscillation across the Bid-Ask Spread. Furthermore the minimum price increment on Kraken is 0.0001, so the minimum any buy order following a sell order can move the Last Price is 0.0001 (barring a simultaneous drop in the Ask price.) Let me repeat: Unless the Ask price drops, any and all buy orders following a sell must move this price graph upward by at least .0001 by definition. This appears to be lost on the authors, who highlight the .0001 move upward as something notable, an outsized price move for such a small order. They might be shocked to discover that a size 0.01 order would also move the price graph upward as much as the size 75 order.