The Rolling Naked Tesla Short

[luvb2b]

Continuing my series of naive questions: couldn't the original seller just buy them back to close their position? That would also reduce the open interest, wouldn't it?

ggr, theoretically what you say is possible. however the evidence indicates that's not it. here's why:

(1) on some days the prints are so close together (minutes or seconds) that there's no consistent rational explanation why someone would get into 1000 contracts (which represent 100,000 shares) and then seconds later decide to flip out of it. and then repeat every tuesday.

(2) there have been days where some contracts trade 5 blocks of 100 contracts each with open interest of (say) 90 contracts. so you know the 100 contract blocks must be new activity, as no trader has 100 contracts to trade. with 5 blocks, only 4 could offset each other, implying open interest should go up by at least 100 contracts. instead the open interest doesn't change or declines, which means all traded contracts are getting exercised.

(3) the size of the trades is generally very large, consistently representing 25,000 to 150,000 shares per printed trade. if you've ever tried to move a position this large, you know you can't do it on a dime. even if a market maker let you get into it, he won't let you out moments later without exacting his pound of flesh.

(4) the regularity with which these trades happen and the way they happen regularly on tuesdays to make sure shares settle on friday before the weekend are not likely to be a day trader.

@luvb2b
That's a $202million commitment in one day. Seriously, if this continues, my theory of someone using options to secretly corner the market like VW replay is more and more likely. Except this time, they learned the lesson that having options without the capital backing to exercise is not enough.

the 3 stooges theory means there is no corner going on. larry & moe trade a spread with each other. so effectively larry is buying from moe who is selling to larry who is forced to sell to moe who is buying from larry. got that? :smile:

when they trade a call spread with each other and both exercise one side, they can effectively offset hundreds of thousands of shares with each other, leaving themselves only the stock and open interest they seek to harvest from shemp. go back and work through the original 3 stooges post you'll see what i mean.

the one thing you can be sure of - this is most likely two mid-size hedge funds or equity derivatives desks (like maybe at banks) that are doing this trade. the amount of capital required to exercise so many calls on both sides means both parties have to have substantial trading capital. they also have to have access to get into the lending pools for tesla shares pretty easily. if it's illegal i'm very curious to see who the regulators identify as the culprits.

it can't be a corner, because we've seen millions of shares equivalent cross almost every week for a few weeks now. if there were a corner, these guys would have to file with the sec as soon as they got over 5.5 million shares. also you know these positions are not unhedged positions, because you couldn't even put on trades like these with the volume as low as its been. not sure how to explain it but think about trying to buy 4.5 million shares on a day when only 3 million shares trade. you can't do it.

one could spend an eternity trying to figure out an algorithm. you never will. the trades are being posted from off the floor, most likely posted manually. they really don't care when exactly the trades get posted, only that they hit the strikes that they want with the volume that they want.

the 3 stooges theory allows you to predict: (a) which strikes they will do, (b) how much relative volume they will do in each strike, (c) when they will do it, and (d) why they do it. i posted where they would trade and relatively how much before yesterday's open to demonstrate that. the theory is completely consistent with the observed facts which are: (a) high rate earned for loaning out shares, (b) tens of thousands of contracts trading with no change in open interest and little discernible price impact, (c) the absence of increases in the open interest the next day, (d) suspicious trades only in strikes with zero time premium and meaningful open interest, and (e) daily share volumes which are below the volume that would be implied by the suspicious options volume.

originally i started this thread thinking it was a naked short, but that theory was falling apart as the trading volumes increased because (a) high naked shorts would put tesla on the reg sho list which didn't happen, (b) there was no way to cover the shares to match the rolling of the short when volume was far below the options volume, and (c) there was no good explanation of how they covered the naked short to avoid appearing on the reg sho list.

unless an alternate theory is presented that is consistent with all the observed facts, i will stand by the stooges.

to me the meaningful remaining questions are (1) is it legal, and (2) are there ramifications of this activity that create predictable price patterns worth trading. the subject remains interesting to me because it's a fascinating use of the technicalities of options to steal valuable interest bearing positions out of someone else's account.

 

[PureAmps]

(1) on some days the prints are so close together (minutes or seconds) that there's no consistent rational explanation why someone would get into 1000 contracts (which represent 100,000 shares) and then seconds later decide to flip out of it. and then repeat every tuesday.

So why are these trades consistently performed in "pairs" of equal sizes? What purpose does it serve, do you think? The only other explanation I have is that it is intended to "appear" like an open/close or spread trade to mask the true intention of the trade.

it can't be a corner, because we've seen millions of shares equivalent cross almost every week for a few weeks now. if there were a corner, these guys would have to file with the sec as soon as they got over 5.5 million shares.

Not a corner, but could be a cover of a short position. A very creative way to avoid buying shares on the open market and bidding the stock price up, reducing the effect of a short squeeze.

one could spend an eternity trying to figure out an algorithm. you never will. the trades are being posted from off the floor, most likely posted manually. they really don't care when exactly the trades get posted, only that they hit the strikes that they want with the volume that they want.

An eternity? I think you have stated the algorithm pretty well in your previous posts. They appear to be trading in-the-money calls with open interest. I'm just trying to figure out their criteria in a more precise manner. The trades could be posted manually or through a program, but given the sophistication of this trade I'd lean towards a program, or at least somebody has a spreadsheet model they are looking at carefully before punching in the trade. This is not a "seat of your pants" kind of trade...

 

[kenliles]

Not a corner, but could be a cover of a short position. A very creative way to avoid buying shares on the open market and bidding the stock price up, reducing the effect of a short squeeze.

This seems like the natural motivation to me. And if this is the result (I don't quite see where the actual shares come from), then it damn well should be illegal as it changes the risk of going short in the first place, transfers the benefit of short covering from longs to neutral. And does in effect change the resulting trade price of e underlying stock by altering the demand- by trading 'off the board' with a non arms length collusive relationship. Since the MM must know this is happening, it seems like they would work to stop it, unless they are benefiting in some way. (Its this kind of crap that brought the real-estate market down - sorry for the vent)

 

[30seconds]

any other stocks with high costs to shorts where this pattern can be confirmed? If I've learned anything from my friends in that part of the business world is that if something works in one place then the next thought is to expand. could other stocks with large short vs % of float and days to cover also be subject to a harvest trade like this?

 

[PureAmps]

This seems like the natural motivation to me. And if this is the result (I don't quite see where the actual shares come from)

The stock they are acquiring is from the market maker's hedges and existing "retail" covered calls and spreads. If you look at how the trades have evolved over time, they began with huge volume the week of 3/25 and that volume was matched with corresponding increase in open interest. These are most likely long calls acquired, possibly as insurance, or in preparation for exiting the short position. These calls were most likely "sold" by the market maker. These shares if exercised would come from the market maker's inventory, which would have to be replenished via market operations at some point, possibly contributing to the "short squeeze".

The more recent trades seem to be designed to "shake the tree" and harvest the open interest of longs that are short calls in the form of covered calls or spreads. We have at least one data point on this thread confirming that covered calls were assigned before expiration. These shares would normally be unavailable to the short as they would be exercised by the market maker on expiration and go into the market maker's inventory. This is basically freeing up additional liquidity that wouldn't normally be available in the open market, and freeing it up sooner rather than later.

Again, this is all just speculation...

 

[luvb2b]

So why are these trades consistently performed in "pairs" of equal sizes? What purpose does it serve, do you think? The only other explanation I have is that it is intended to "appear" like an open/close or spread trade to mask the true intention of the trade.

Not a corner, but could be a cover of a short position....

....I'm just trying to figure out their criteria in a more precise manner. The trades could be posted manually or through a program, but given the sophistication of this trade I'd lean towards a program, or at least somebody has a spreadsheet model they are looking at carefully before punching in the trade. This is not a "seat of your pants" kind of trade...

the exact sizes of the prints don't matter. they just put together some spread combination or some pairs of individual trades and cross them with each other. what matters to them is the total volume they can do in each contract relative to the open interest. that allows them to harvest proportionally an equal amount from various strikes.

the exercise of both sides of the spread results in a hedged position in larry & moe's accounts: long stock + short deep in the money call. if they are short, it does temporarily give them cover on part of the position: short stock cancels with long stock, leaving them short deep in the money calls. this position has some advantage in that it has the same delta as being short stock for small price moves, but allows them to save the cost of borrowing. as you said, if these positions are stolen from longs who are doing buy-writes in accounts with no lending of shares, it does temporarily free up more shares and make them available in the lending pool. freeing more shares for shorting and lowering the lending rate - both are advantages for short sellers. how many shares they get this way vs. what they steal from the market maker (who is already lending) makes a big difference. shares taken from a market maker who is already lending to earn the rebate won't change the available pool of short shares or lower the lending rate.

anyway the problem is at expiration the options get exercised and larry & moe go back to being short again. we'll see what happens here at the april expiration. they harvested a lot of open interest out of april, and i'm pretty sure if they're using as a temporary short cover you'll see a lot of tightness in the availability of shares and higher borrowing costs for shares next week.

usually a trading algorithm has a lot more details to it than what i posted. if you want to "forecast" their trades, here's a pretty good formula. try it out next week.
(a) wait til monday night.
(b) note all options strikes where calls are deep in the money and have meaningful open interest.
(c) watch trading closely on tuesday.
(d) trades will cross on phlx in the options identified in (b) on tuesday.
(e) volume in each strike will be proportional to the amount of open interest that's available. typically they need to do several multiples of the open interest in each strike to harvest effectively.

- - - Updated - - -

any other stocks with high costs to shorts where this pattern can be confirmed? If I've learned anything from my friends in that part of the business world is that if something works in one place then the next thought is to expand. could other stocks with large short vs % of float and days to cover also be subject to a harvest trade like this?

you're correct that you should see this elsewhere too. i've never seen a stock with borrowing costs so high for so long as tesla, so i have not seen it before.

basically the criteria you need are:
(a) high borrowing cost for shares
(b) an optionable stock
(c) speculators using deep in the money calls to be long shares, or a stock that moves higher quickly making many at-the-money options deep in the money.
(d) visible open interest in the deep in the money strikes, the evidence for (c) essentially
(e) the deep in the money calls trading with zero time premium

i suppose you could scan for such stocks and find where this stuff is happening. some hedge funds would be doing very well on these trades.

 

[Causalien]

So why are these trades consistently performed in "pairs" of equal sizes? What purpose does it serve, do you think? The only other explanation I have is that it is intended to "appear" like an open/close or spread trade to mask the true intention of the trade. Not a corner, but could be a cover of a short position. A very creative way to avoid buying shares on the open market and bidding the stock price up, reducing the effect of a short squeeze. An eternity? I think you have stated the algorithm pretty well in your previous posts. They appear to be trading in-the-money calls with open interest. I'm just trying to figure out their criteria in a more precise manner. The trades could be posted manually or through a program, but given the sophistication of this trade I'd lean towards a program, or at least somebody has a spreadsheet model they are looking at carefully before punching in the trade. This is not a "seat of your pants" kind of trade...

I agree that hidden short covering is actually a better explanation based on occam's razor. Which also has the effect of a delayed rise in price, offsetting the side effect of covering a huge short where price runs up as you cover. But then, this will mean that the MM facilitating this will hurt a lot if the stock in turn goes back down... if they didn't hedge. BTW, maybe I missed this mentioned before, but has anyone went to the tick by tick of the actual order to see whether or not the orders landed on the bid or ask side? Are we so sure that it was a buy order? If not, I'll try to take some time to look at it if somebody has a time stamp. EDIT** If the HFT theory is true, then someone most definitely pointed their text scraping bot on this forum and our conversations in the past week might have triggered a rally. Not sure how they'd do it, the only way I can think of is to measure the ratio of positive to negative words on all posts that are mentioned in a day and to do an association with a number mentioned within 5 words from the sentiment in question. I'd like to test it out by getting everyone to mention a few words of positivity for a set amount of days, but SEC will probably slap someone.

 

[avatar]

This seems like the natural motivation to me. And if this is the result (I don't quite see where the actual shares come from), then it damn well should be illegal as it changes the risk of going short in the first place, transfers the benefit of short covering from longs to neutral. And does in effect change the resulting trade price of e underlying stock by altering the demand- by trading 'off the board' with a non arms length collusive relationship. Since the MM must know this is happening, it seems like they would work to stop it, unless they are benefiting in some way. (Its this kind of crap that brought the real-estate market down - sorry for the vent)

I agree. this is frustrating. However it feels like they cannot stop the tsunami - only slow it. The price for TSLA has been gradually ascending (perhaps slower than it otherwise would). I wonder though if this guy is the only thing stopping a fully blown squeeze right now..

 

[Elshout]

As of 11:30 PDT TSLA is up $1.84 in a market down almost a percent. The shorts are getting their gonads squeezed and I love it!

 

[kenliles]

I wonder though if this guy is the only thing stopping a fully blown squeeze right now..

well relative to volume, this 'guy' IS the market right now;
the registered MM has no current relevance (except for what they are being forced to do by the new MM); nor do current traders, so I'd have to say yes to your excellent musing, and mission accomplished by the stooges (yuk-yuk);
this following is speculatory(but not by much - yes I know there's no such word):
I'd also say it introduces a whole new parameter of short squeeze characterization for all future investors to consider-
where apparently 'mutual-funds' become 'mutual-buds', create their own market, co-opt and replace the MM, becoming the new unregistered default MM...
until they clear their shorts of crapola, handing them unwashed back to the original MM and the rest of the market, namely us

@PureAmps- thanks for detailing the answer to my ? about where the stock accum is coming from- much appreciated

 

[luvb2b]

tomorrow is tuesday again, so our friends larry & moe should be back. i'll try to post where i expect to see them working tonight.

 

[blakegallagher]

Cant wait ... thanks for staying at it Luvb2b it has been very interesting watching this progress.

 

[TD1]

tomorrow is tuesday again, so our friends larry & moe should be back. i'll try to post where i expect to see them working tonight.

If they should return tomorrow with a new naked short, how you think that would influence the stock if it would finish today at 49.90$ ?

 

[luvb2b]

If they should return tomorrow with a new naked short, how you think that would influence the stock if it would finish today at 49.90$ ?

the naked short theory was replaced by the 3 stooges theory (a couple pages back). 3 stooges fits the behavior better than the initial theory i posted. perhaps the thread title should be changed to the 3 stooges tesla options traders.

ok, here's what i expect tomorrow, look for volume in these strikes, proportional to the open interest. i posted tonight's open interest, but tomorrow morning's numbers will be the ones people will trade on. expect especially heavy activity in the may 40's, jun 40's and jun 38's as there's good open interest at those strikes.

may 35 call oi 436
may 37 call oi 123
may 38 call oi 130
may 39 call oi 282
may 40 call oi 2266
may 41 call oi 503
may 42 call oi 862
jun 32-40 calls, oi's 268, 304, 153, 757, 655, 616, 1651, 688, 9077 (respectively)

 

[Nicu.Mihalache]

luvb2b, just to say once more that I appreciate your efforts; and I completely agree with your scientific approach: look at the data, conceive a theory, make predictions, adjust or KILL the theory and start over - that last part of killing it instead of adding circles on circles is most of the time the real key to great discoveries; it seems even Einstein made all possible (but intelligent) errors before coming up with his relativity theory

keep up the good work

 

[luvb2b]

luvb2b, just to say once more that I appreciate your efforts; and I completely agree with your scientific approach: look at the data, conceive a theory, make predictions, adjust or KILL the theory and start over - that last part of killing it instead of adding circles on circles is most of the time the real key to great discoveries; it seems even Einstein made all possible (but intelligent) errors before coming up with his relativity theory

keep up the good work

thanks very much for the kind words.

here's the updated view on where the stooges will be working today, based on the morning's open interest figures:

may 35 call oi 436
may 39 call oi 208
may 40 call oi 1486
may 41 call oi 535
may 42 call oi 860
jun 32-40 calls, oi's 268, 304, 151, 575, 489, 310, 979, 675, 8947 (respectively)

the heaviest volume will come in the jun 40s, may 40s, jun38s, and may 42s. those jun 40s are kind of on the cusp, so if the stock goes down you may not see the stooges working there, on the other hand if the stock goes up the jun 41s and maybe the may 43s will also come into play.

 

[smorgasbord]

Here's a similar theory posted on Seeking Alpha:

First, we suspect most early shorts in the 27 to 34 range had no cover initially , while most early purchasers of 2015 calls were simply longs, either adding leverage or happy paying a premium to keep capital off the table. Many banks don't understand the stock.
Then, after Musk said earnings would be a reality, with positive indications from Europe, volumes particularly calls /puts ramped up, as well as call premiums. That is where we think shorts began to try to cover aggressively. It looks that in the last few weeks, a lot of shares purchased by buying calls with negligible premiums at or near expiry which would be exercised. Purchasing in this way has a slight added cost , but avoids pressuring the stock price.
Third, it is likely shorts might sell puts simultaneously to substantially reduce the cost of managing 'out' or of getting caught.
None of that would change the thrust of your math however, that there are not enough outstanding calls to cover that much of the shorts.
Given what appeared to be the original 'binary short thesis' (=TSLA to zero), what has surprised the most has been the continued increase in the short position well after the election. We'd expect that drops steadily in the next few reports.
Final observation, is that TSLA is ultimately worth only what someone is willing to pay for it, but what has really happened is that 'ultimately' now seems to involve a much more prosperous lifespan than shorts had expected

 

[luvb2b]

Here's a similar theory posted on Seeking Alpha:

this theory that you linked is unrelated to the 3 stooges theory, which is very specifically addressing options that trade but then vanish from the open interest reports.

we are indeed seeing huge size in the series i mentioned today. will try to post a fuller update later.

 

[luvb2b]

huge size again!

as expected, monster volume trading thru the phlx in a bunch of strikes. i estimate total over 45,000 contracts this way again, representing 4.5 million shares. i think they have to stay under 5.5 million shares because that would trigger an sec filing (over 5% of the outstanding shares).

so more or less, they are maxxing out what they can do.

may 35 call vol 2,854 oi 427
may 39 call vol 1,416 oi 208
may 40 call vol 6,000 oi 1,486
may 41 call vol 2,000 oi 535
may 42 call vol 3,600 oi 860
jun 32 call vol 1,736 oi 268
jun 33 call vol 2,008 oi 304
jun 34 call vol 1,102 oi 151
jun 35 call vol 3,550 oi 575
jun 36 call vol 2,878 oi 489
jun 37 call vol 2,020 oi 310
jun 38 call vol 5,956 oi 979
jun 39 call vol 2,800 oi 675
jun 40 call vol 2,000 oi 8,947

i also see these guys in some further expirations now, sep, dec, and jan. too many strikes to post all, but how about this biggie:

jan 30 call vol 2,800 oi 676

with the exception of the jun 40 calls, all the volume is proportional to the open interest.

should be some rebalancing that market makers have to do tomorrow, although i wonder if they haven't figured it out enough to start rebalancing today. i guess we'll see how firm the stock is through the morning.

 

[Johan]

Follow up question: If the basic assumption here is that the two big stooges (who are trading all these deep in-the-money options with eachother every day) are doing this to stay short on TSLA but gradually switching their stock to options to be short on, to avoid lending charges, then at some point they will have exchanged their entire actual stock position to an equivalent, still short, position on a combination of calls and puts right? And there is still no sign of them slowing down the pace which must mean that they had a very large short position in actual TSLA stock to begin with? Now, they are trading in big volumes in the may calls. So in effect unless TSLA drops significantly in time for the may options expiration they will be sitting on a lot of worthless calls? My point being they are staying short only on option instead of stock (thereby avoiding high lending charge rates). But unless TSLA pulls back quite a lot they will be sitting with quite some losses come may?