3 ms·

Why do CFTC try to police this? For the money or optics? Orders far from the BBO are clearly irrelevant and those close to the BBO are not "free" for the spoo

by derriz 5y ago

Why do CFTC try to police this? For the money or optics? Orders far from the BBO are clearly irrelevant and those close to the BBO are not "free" for the spoofer given they have to bear the risk that the orders could be filled if the market moves. Effectively it means anyone cancelling an order has to worry that their action could be interpreted as "spoofing" - which will make market making more risky and expensive. The only ones this type of enforcement "protects" are are those who naively think they've discovered a strong trading signal based on order volumes away from the BBO. Putting a $1 bid on a Rolex on ebay which is currently attracting $10k bids isn't going to fool anyone into thinking that demand is increasing and is relatively harmless. For me real manipulation involves actual trades, not cancelled/unexecuted orders, and should be policed stringently. E.g. banging the market at close in order to nudge the closing price in order to boost the value of a position in a different book.

adriancr 5y ago
> anyone cancelling an order has to worry that their action could be interpreted as "spoofing" And getting investigated by CFTC, and CFTC finding proof that you never intended to execute it AND you profiting from steering market in the desired direction which would not have happened without the fake trade...
- derriz 5y ago
  There were no "fake trades" involved in this case - only orders which were cancelled.
  - adriancr 5y ago
    they were fined for 'spoofed' trades which in my view are fake trades - they never intended for them to execute, only to steer the market in a direction they want to fill in orders they had at a profit. if those trades ran the risk of executing they would have probably been cancelled fast and moved further.
    - derriz 5y ago
      The language of trading is fairly precise - trades and orders are quite distinct things. Orders may or (commonly) may not result in trades. An order that doesn't execute is not associated with any trade. JPMorgan were fined for spoofing orders not trades.
csee 5y ago
How do you know it's far from BBO
- carnitine 5y ago
  They don’t. They clearly cover both cases.
  - csee 5y ago
    > those close to the BBO are not "free" for the spoofer given I missed this earlier, but it's not a good argument. It's all about risk-adjusted expected value. There's a distribution of payoffs to spoofing, and it skews far to the right. The conceptual categories of "free" vs "not-free", or "risky" vs "riskless" is the wrong mental framework for evaluating this problem. Yes, it's not risk-free, but why is that relevant?
    - totalZero 5y ago
      It's relevant because a live quotation on an exchange can be traded and is thus by definition "real" regardless of the intention with which it was submitted.
      - csee 5y ago
        > a live quotation on an exchange can be traded and is thus by definition "real" This is just a restatement of the thing that I was questioning. Yes, it's real according to the definition that you've put out. The spoof order can be traded against. So? Why does this binary real/not real categorization matter for whether spoofing should or shouldn't be allowed? An intelligent spoofer isn't going to place large orders directly on the BBO when the fill probability is high. They're going to be doing it when the correlated markets aren't moving against their order, for instance. Or they're going to put it one tick below the BBO and algorithmically pull when a trade hits the level above. So even if they do get filled, which is unlikely if they're smart, it's no big deal since they've made lots of money up until that point and they can get out of the unwanted position without much slippage.
hatesinterviews 5y ago
The SEC regulates the exact same thing for US stocks. And nearly every other country / regulated exchange has similar rules. When you post an order to any exchange, you are doing so under the agreement that it is legitimate and that you actually want to be filled at that price. Obviously a participant’s desire desire to be filled at a specific price can change over time, so you’re allowed to cancel orders as well. That makes it difficult to detect and prove that a participant is acting maliciously. But if the BBO isn’t really moving and you keep posting/cancelling orders all day in certain patterns then eventually you’re going to get investigated and told to explain your behavior. And that’s where you might struggle to justify things. > Effectively it means anyone cancelling an order has to worry that their action could be interpreted as "spoofing" On human time scales, I think you’re vastly overstating the relevance of accidentally spoofing. For automated trading, yes you have to be careful that your signal doesn’t flicker right on the edge of your threshold. Sometimes that means you need to purposefully limit your order entry or sometimes it’s just a matter of improving your signal. > The only ones this type of enforcement "protects" are are those who naively think they've discovered a strong trading signal based on order volumes away from the BBO. I would say most of high frequency trading, which is the primary means of market making these days, relies on the state of the order book as a primary signal. Sure they incorporate other outside information, but when you’re trying to be the fastest, the only data you have to work with that is fast enough is the what the exchange says the order book is. No body’s paying attention to $1 quotes on a $10,000 stock. But a couple dozen price levels from the BBO on GOOG might easily be less 1% away and maybe spoofing there at large sizes might be enough to negatively affect HFT and cause spreads to widen, which hurts everyone. Basically these regulation exists because only legitimate activity is allowed. Trying to fool other market participants (spoofing) or trying to slow down specific matching engines (quote stuffing) are obviously not legitimate.
- derriz 5y ago
  Actually it was the SEC rules I had some exposure to but this was a few years ago and was communicated to us via the compliance dept. What I recall is that because of the difficulty in proving intent, the rules allow very broad interpretation on the part of the SEC. The bar for proving spoofing seemed very low as, for example, the SEC does NOT have to demonstrate that you made a profit from it. > But if the BBO isn’t really moving That might looks suspicious sure, but there may be other factors (not reflected in the book) influencing a change of desire on my part. For example, what if I only want to buy one of GOOG or APPL and optimistically stick in a low-ball bid on each. If one gets filled, I cancel the other. I never intended for BOTH orders to be filled; is this spoofing? Or half-spoofing? This isn't a real strategy but there are lots of strategies which can look like this. I agree the state of the order book drives a lot of the behavior of a strategy (but other factors like current position are also critical) but I've never come across a strategy that responds to volume changes far from the BBO. Not to say they don't exist - it's a secretive industry after all - but the changes at or very near the BBO clearly reflect real intent and are weighted accordingly. Quantity change far away from the action is mostly noise.
fnordfnordfnord 5y ago
>Effectively it means anyone cancelling an order has to worry No, it really doesn't. This is a single enforcement action that probably took months or years to put together. The behavior being policed is traders entering thousands of orders in a short period of time and then cancelling the orders after their real orders were filled, and then repeating this behavior over some period of time. Someone who fat fingers an order occasionally, or changes their strategy in response to intraday market movements has nothing to fear. >For me real manipulation Yeah those people also suck and I hope they get what's coming to them, even though I know they won't.

by derriz 5y ago

adriancr 5y ago

> anyone cancelling an order has to worry that their action could be interpreted as "spoofing" And getting investigated by CFTC, and CFTC finding proof that you never intended to execute it AND you profiting from steering market in the desired direction which would not have happened without the fake trade...

derriz 5y ago
There were no "fake trades" involved in this case - only orders which were cancelled.
- adriancr 5y ago
  they were fined for 'spoofed' trades which in my view are fake trades - they never intended for them to execute, only to steer the market in a direction they want to fill in orders they had at a profit. if those trades ran the risk of executing they would have probably been cancelled fast and moved further.
  - derriz 5y ago
    The language of trading is fairly precise - trades and orders are quite distinct things. Orders may or (commonly) may not result in trades. An order that doesn't execute is not associated with any trade. JPMorgan were fined for spoofing orders not trades.

csee 5y ago

How do you know it's far from BBO

carnitine 5y ago
They don’t. They clearly cover both cases.
- csee 5y ago
  > those close to the BBO are not "free" for the spoofer given I missed this earlier, but it's not a good argument. It's all about risk-adjusted expected value. There's a distribution of payoffs to spoofing, and it skews far to the right. The conceptual categories of "free" vs "not-free", or "risky" vs "riskless" is the wrong mental framework for evaluating this problem. Yes, it's not risk-free, but why is that relevant?
  - totalZero 5y ago
    It's relevant because a live quotation on an exchange can be traded and is thus by definition "real" regardless of the intention with which it was submitted.
    - csee 5y ago
      > a live quotation on an exchange can be traded and is thus by definition "real" This is just a restatement of the thing that I was questioning. Yes, it's real according to the definition that you've put out. The spoof order can be traded against. So? Why does this binary real/not real categorization matter for whether spoofing should or shouldn't be allowed? An intelligent spoofer isn't going to place large orders directly on the BBO when the fill probability is high. They're going to be doing it when the correlated markets aren't moving against their order, for instance. Or they're going to put it one tick below the BBO and algorithmically pull when a trade hits the level above. So even if they do get filled, which is unlikely if they're smart, it's no big deal since they've made lots of money up until that point and they can get out of the unwanted position without much slippage.

hatesinterviews 5y ago

The SEC regulates the exact same thing for US stocks. And nearly every other country / regulated exchange has similar rules. When you post an order to any exchange, you are doing so under the agreement that it is legitimate and that you actually want to be filled at that price. Obviously a participant’s desire desire to be filled at a specific price can change over time, so you’re allowed to cancel orders as well. That makes it difficult to detect and prove that a participant is acting maliciously. But if the BBO isn’t really moving and you keep posting/cancelling orders all day in certain patterns then eventually you’re going to get investigated and told to explain your behavior. And that’s where you might struggle to justify things. > Effectively it means anyone cancelling an order has to worry that their action could be interpreted as "spoofing" On human time scales, I think you’re vastly overstating the relevance of accidentally spoofing. For automated trading, yes you have to be careful that your signal doesn’t flicker right on the edge of your threshold. Sometimes that means you need to purposefully limit your order entry or sometimes it’s just a matter of improving your signal. > The only ones this type of enforcement "protects" are are those who naively think they've discovered a strong trading signal based on order volumes away from the BBO. I would say most of high frequency trading, which is the primary means of market making these days, relies on the state of the order book as a primary signal. Sure they incorporate other outside information, but when you’re trying to be the fastest, the only data you have to work with that is fast enough is the what the exchange says the order book is. No body’s paying attention to $1 quotes on a $10,000 stock. But a couple dozen price levels from the BBO on GOOG might easily be less 1% away and maybe spoofing there at large sizes might be enough to negatively affect HFT and cause spreads to widen, which hurts everyone. Basically these regulation exists because only legitimate activity is allowed. Trying to fool other market participants (spoofing) or trying to slow down specific matching engines (quote stuffing) are obviously not legitimate.

derriz 5y ago
Actually it was the SEC rules I had some exposure to but this was a few years ago and was communicated to us via the compliance dept. What I recall is that because of the difficulty in proving intent, the rules allow very broad interpretation on the part of the SEC. The bar for proving spoofing seemed very low as, for example, the SEC does NOT have to demonstrate that you made a profit from it. > But if the BBO isn’t really moving That might looks suspicious sure, but there may be other factors (not reflected in the book) influencing a change of desire on my part. For example, what if I only want to buy one of GOOG or APPL and optimistically stick in a low-ball bid on each. If one gets filled, I cancel the other. I never intended for BOTH orders to be filled; is this spoofing? Or half-spoofing? This isn't a real strategy but there are lots of strategies which can look like this. I agree the state of the order book drives a lot of the behavior of a strategy (but other factors like current position are also critical) but I've never come across a strategy that responds to volume changes far from the BBO. Not to say they don't exist - it's a secretive industry after all - but the changes at or very near the BBO clearly reflect real intent and are weighted accordingly. Quantity change far away from the action is mostly noise.

fnordfnordfnord 5y ago

>Effectively it means anyone cancelling an order has to worry No, it really doesn't. This is a single enforcement action that probably took months or years to put together. The behavior being policed is traders entering thousands of orders in a short period of time and then cancelling the orders after their real orders were filled, and then repeating this behavior over some period of time. Someone who fat fingers an order occasionally, or changes their strategy in response to intraday market movements has nothing to fear. >For me real manipulation Yeah those people also suck and I hope they get what's coming to them, even though I know they won't.