What's the deal with a bunch of Ask sizes of "1"?

[d0rian]

It's considered illegal because it's chucked under the same realm as spoofing. Even spoofers have bonafide orders exposed to risk but "intent is not for the orders to execute." It's basically pro-algo/HFT/MM bullshit regulation.

So where does that leave my real-number example above, where a MM's algo just leapfrongs my 10 x 0.65 Bid with its own 10 x 0.70 bid (in a somewhat transparent attempt to make me pay a higher price...possibly even getting hit by the same MM's selling algo if I go to 10 x 0.75)?

I don't think it's inaccurate to say that the MM's doing exactly what, if I'm understanding correctly, you're saying it would be illegal for a retail trader to do?...That is: submit a quote whose purpose isn't necessarily to get executed, but rather to entice retail buyers to bid higher? I suspect I already know the answer...? That it's just one of those things where yep, unfairness and all, the individual investor is getting screwed by a regulation that penalizes him for doing something MM's are allowed to do?

 

[d0rian]

It's basically pro-algo/HFT/MM bullshit regulation. I'll see if I can find the story on the guy that got nailed for midpoint manipulation somewhere.
Update: so it turns out my memory was wrong on what the actual manipulation was, as it wasn't just jacking the midpoint around... but here you go: http://www.bloombergview.com/articles/2015-12-03/tricky-twins-spoofed-trading-computers

I have a feeling these 1 lots you are seeing *are* trying to game MMs though.

Also, thanks for that link. Eye-opening and...wow, kind of made me really angry. The HFT's and MM's are essentially whining that their super expensive algos are getting outwitted. It's an incredibly exploitable loophole to have your algo Bid for 150+ contracts at $x simply because it sees a 1-lot order come in, also at $x. It would be one thing if that type of spoofing were disallowed by all market participants, retail or automated/professional. But -- as in my 0.65 / 0.70 / 0.75 example above -- it's something that MM/HFT/algo's seem to have specific permission to do(!) They leapfrog retail orders for the sole purpose of juicing up the price. And -- I forget where I saw this -- but there's some famous article / site that shows that only a tiny fraction of orders submitted to exchanges are ever actually executed...that millions of orders are submitted by algos and then nearly-immediately yanked. Clearly they're being submitted to manipulate the market in some respect...why are they permitted to do so, when by the logic applied by the SEC to the twin case, that activity should be as forbidden as what they got nailed for?? I tend to agree with the law firm's characterization from the article that so long as the investor is submitting a bona fide offer where he's exposing himself to risk (aka he could get hit on his bid/offer), it shouldn't be considered 'manipulative'. As conclusion to article puts it:

"But despite those chats, they don't exactly come across as evil masterminds putting one over on a bunch of innocents. They come across as guys trying to outsmart computers in a race to earn trading rebates. The problems here, if they are problems, are about the structures and incentives involved in liquidity rebates and hidden order types. Kenny and the Afshars were just proactive about exploiting them."​

 

[i960]

Basically, the thing that should make you angry here, and which makes me angry, is that regulation is siding on the side of exchanges, HFTs, and MM entities in nearly all cases involving a situation where they get outplayed. While spoofers are kinda annoying during trading, they're the necessary yang to the yin - and keep algos/HFT in check. If only one side is allowed to be the sole surviving predator in the game (which at this point are the established HFT/MM entities) then there's no balance and they basically win through attrition.

 

[d0rian]

And...to ask a clumsy but necessary question: just what's the principle that enables HFT/MM's to do what is literally illegal for retail investors to do? I'm sure many dif laws apply, there are different exchanges, State laws, etc, but...well what's most galling is that this doesn't appear to be one of those instances where a law happens to be neutral-on-its-face, but simply disadvantages retail investors more than HFT/MM's. This appears to be a situation where it's explicitly illegal for a retail investor do X, but OK for an HFT/MM; e.g. my leapfrogging quote example above would seem to be a clear as day example of orders that aren't meant to be executed but serve primarily to goose the price up. I feel like this thread has taken on the quality of "Old Man Yells At Cloud", and that all of this is old news to you more seasoned options traders...
In any case, I can't think of any other reason than some sort of fraudulent manipulation for a whole batch of 1-lot orders to show up (usually at the same time of day.) I don't know exactly what the scheme is, though I'm amazed that it's being done so transparently, I suppose.

 

[sprstpd]

Life isn't fair.

The thing is, if you give a black and white definition of spoofing involving high cancel rates and no intent to fill every order, then HFTs should be just as guilty as spoofers themselves (and probably more so). I don't see how any court of law could actually prove someone was spoofing because every order on the book is able to be executed against. Just by entering an order into the book gives intent that you want to be filled, regardless of whether you want it or not. Judging someone's "intent" is very near impossible. But having a black and white law is easy, but it would not suit the HFT scum nor the exchanges who bend over backwards for them. The only reason spoofing is being prosecuted is so that HFTs have an easier playing field.

And one final point - I can't believe HFTs are so dumb that they lose regularly to spoofers. Maybe instead of bitching about the spoofers, the HFT programmers should make more robust algorithms that compete on a level playing field instead of one that is rigged. Isn't capitalism about the survival of the fittest? I guess capitalism is a shadow of its former self nowadays, where bankrupt companies are allowed to survive indefinitely largely due to Fed policy.

 

[marketsurfer]

HFT trades very small size per trade.

 

[Sig]

Yup, that thought crossed my mind as well: i.e. the idea that it's not so much using the 1-lot as a 'trip wire' for some larger order or strategy, but rather an attempt to manipulate the MM's Bids/Asks, one purpose of which could be, as you note, to create a new midpoint.

For example, I've noticed in the past that MM's sometime auto-react to your orders by changing their quotes: e.g. if Bid/Ask is 10 x 0.60 / 10 x 1.00, if you place a 10 x 0.65 Buy order, sometimes the MM will 'jump' your order and move its best Bid to 10 x 0.70. Or sometimes it'll just 'match' yours, so that the best bid shows as 20 x 0.65.

But what did someone get "busted" for doing? And what was illegal about it? In my example above, if I'm putting in a legitimate bona fide bid of 10 x 0.65, and the MM beats mine with 10 x 0.70, why shouldn't I be able to unload 10 contracts at 0.70?? Sure, perhaps it was my intention all along to Sell 10 contracts I was holding but wanted to 'goose' up the MM's Bid by $0.10...but my 10 x 0.65 was nonetheless a bonafide, executable Buy order that someone could have hit, right? Or to put it another way, the MM jumped my 0.65 Bid for the sole purpose of trying to get me to leapfrog his to 0.75, right? IOW doing exactly the same thing that I'm doing...turnabout fair play, no?

Do you happen to remember what option you saw that behavior on? I've never been able to get the MM to move in response to my orders, so I always assumed they're working off their own pricing model and ignoring the other orders. However I'm always on pretty liquid index options, so very possible that it's different there. Out of curiosity I'd love to find a place where this does happen.

 

[d0rian]

Do you happen to remember what option you saw that behavior on? I've never been able to get the MM to move in response to my orders, so I always assumed they're working off their own pricing model and ignoring the other orders. However I'm always on pretty liquid index options, so very possible that it's different there. Out of curiosity I'd love to find a place where this does happen.

But isn't that, if I'm understanding correctly, specifically what's prohibited...? i.e. "Getting MM's to move in response to an order"? If the threshold for illegality isn't simply a black-and-white examination of what percentage of an HFT's order flow is yanked before execution (isn't the # something like 95%+?)...then, just what IS the "test" as to whether something's spoofing or not? Is it only prosecutable when, as in the case of the twin brothers, the SEC has a friggin chat record of them talking about how they're trying to bait the algos? The twin article says the definition is "bidding or offering with the intent to cancel the bid or offer before execution." But just what's the test for that, and how can an HFT who yanks 95% before execution (again: my estimate, forget what the actual # is but it's absurdly high) possibly fall on the right side of that?

(Thanks for replies all; very educational even if this is 101-level stuff for you.)

I'll monitor today and see if I can get more insight into all the 1-lots.

 

[Sig]

But isn't that, if I'm understanding correctly, specifically what's prohibited...? i.e. "Getting MM's to move in response to an order"? If the threshold for illegality isn't simply a black-and-white examination of what percentage of an HFT's order flow is yanked before execution (isn't the # something like 95%+?)...then, just what IS the "test" as to whether something's spoofing or not? Is it only prosecutable when, as in the case of the twin brothers, the SEC has a friggin chat record of them talking about how they're trying to bait the algos? The twin article says the definition is "bidding or offering with the intent to cancel the bid or offer before execution." But just what's the test for that, and how can an HFT who yanks 95% before execution (again: my estimate, forget what the actual # is but it's absurdly high) possibly fall on the right side of that?

(Thanks for replies all; very educational even if this is 101-level stuff for you.)

I'll monitor today and see if I can get more insight into all the 1-lots.

Probably, I guess I'm opening myself up to prosecution for spoofing! My interest is more academic on this, i.e. I only place orders that I want filled at the time I place them, I just always thought the market should/might respond to my price signal and it generally just steadfastly ignores me. I guess I just want to feel the power of watching that bid/ask move in response to my order! Thanks for posting the OP though, this is very interesting stuff.

 

[i960]

And one final point - I can't believe HFTs are so dumb that they lose regularly to spoofers. Maybe instead of bitching about the spoofers, the HFT programmers should make more robust algorithms that compete on a level playing field instead of one that is rigged. Isn't capitalism about the survival of the fittest? I guess capitalism is a shadow of its former self nowadays, where bankrupt companies are allowed to survive indefinitely largely due to Fed policy.

This is the point: by only targeting spoofers/anti-algo gamers they're enabling HFT/MMs to get by without having to adapt while the rest of the playing field is forced to adapt.

and how can an HFT who yanks 95% before execution (again: my estimate, forget what the actual # is but it's absurdly high) possibly fall on the right side of that?

"MMs are excluded from these regulations" or some nonsense along those lines.