Papafox's Daily TSLA Trading Charts

[Papafox]

What can I say? Today was much like the previous two trading days: buyers were willing to bid TSLA higher, but an obvious game of "whack-the-mole" any time TSLA ventured into the green prevented any real rallies from taking place. With the exception of the first half hour when shorts manufactured a mandatory morning dip and the last half hour when longs overpowered the "whack-the-mole" capping of excursions into the green, the rest of the day was a game of defense played by short-sellers. I see such games as encouraging because they suggest that the shorts lack the strength to push TSLA lower without help from bad news or negative macros.

This weekend, Tesla now enters its second programed production pause of Q2, with the goal being changes to the line that will enable 5,000/wk M3 production in June with burst modes of 6,000/wk. Should Tesla achieve this goal, the current SP will be a thing of the past and Tesla will begin an era of sustained profitability. Should the production targets lag into July, I think the market will forgive Tesla for this short delay if it continues to show steady progress towards reaching the goal in July. Within two weeks we should see substantial production increases.

For the week, we see significant evidence that Model 3 has reached a sustainable production rate of 3500/wk. That is 70% of the 5,000/wk goal for June and a huge increase over the past few weeks. TSLA closed at 278.85 today, up 2.03 from last week's 276.82. For longs I say, "Have a good weekend!" For shorts, I say, "Tick, tick, tick."

Conditions:
* Dow down 59 (0.24%)
* NASDAQ up 9 (0.13%)
* TSLA 278.85, up 1.00 (0.36%)
* TSLA volume 3.8M shares
* Oil 67.56, down 3.15 (4.45%)

 

[Papafox]

Today was a big down day for the broader markets (especially the Dow) and yet TSLA closed up more than $4. Such a performance clearly shows strength. Let's hope that Consumer Reports finds the Model 3 over the air update fixes the braking issue and brings Model 3 into the "recommend" category. That act would propel TSLA higher.

In today's trading, the shorts began the day with a mandatory morning dip (what a surprise!) and then when it was defeated early on that defeat led to an exuberant climb. Shorts sold 35,000 shares in one minute at 10:08 to stop the rally, but it resumed and hit 286 before running out of rocket fuel. The dip at noon was a NASDAQ dip that TSLA responded to. At closing, over 100,000 shares were sold that minute, which suggests that shorts were either covering or rearming. Short activity dropped to 52% today, and so I think we're seeing the beginning of a trickle of covering.

More to come...

Conditions:
* Dow down 392 (1.58%)
* NASDAQ down 37 (0.50%)
* TSLA 283.76, up 4.91 (1.76%)
* TSLA volume 5.6M shares
* Oil 66.89, up 0.16 (0.24%)
* Percentage of trading by shorts: 52%

 

[adiggs]

Today's short percentage of trading (In descending order):
* AMZN 55%
* NFLX 49%
* FB 46%
* AAPL 30%
* F 26%
* GM . 21%

Clearly companies such as Amazon and Netflix get lots of attention themselves from the shorts, whereas Ford and GM are in a whole different category.

I just realized that between this post and a previous post, I don't have the technical answer to my question about "short percentage of trading".

I can think of two definitions, but these two definitions lead to two very different interpretations and will meaningfully impact your interpretation @Papafox.


First a note - every trade has a buyer and a seller. The traditional trade is one party buy-to-open and the other party sell-to-close. The selling party is closing their position in the security, and the buyer is opening a corresponding position in the security.

But you also have the ability sell-to-open (short seller), and buy-to-close (the second half of the transaction, for a short seller).

There are two more possible combinations - buy-to-open paired with sell-to-open, and buy-to-close paired with sell-to-close.


The first definition, and this makes the most sense to me, is the % of trades (share weighted) that involve a "sell-to-open" transaction (but does not include the "buy-to-close" transactions). This would give us a direct measure of the number of new shares "created" in the market by short sellers (while ignoring the shares "destroyed" by the short sellers as they exit positions).

The second definition would be the % of transactions that were either "sell-to-open" OR "buy-to-close", and would tell us something about the activity levels of shorts in the market, but wouldn't tell us whether net shares were increasing or decreasing (if all the shorts were engaged in "buy-to-close", that would look like a lot of activity, when what we're seeing is shorts leaving the market).

Actually a third definition I can think of is the first definition with buy-to-close shares subtracted out. But I don't know how that would turn into a % of shares as reported.


I looked back at the two websites and I haven't yet found something where they talk about the specifics - just the % of short activity. I'm going to do some more digging.

 

[adiggs]

I just realized that between this post and a previous post, I don't have the technical answer to my question about "short percentage of trading".

I can think of two definitions, but these two definitions lead to two very different interpretations and will meaningfully impact your interpretation @Papafox.


First a note - every trade has a buyer and a seller. The traditional trade is one party buy-to-open and the other party sell-to-close. The selling party is closing their position in the security, and the buyer is opening a corresponding position in the security.

But you also have the ability sell-to-open (short seller), and buy-to-close (the second half of the transaction, for a short seller).

There are two more possible combinations - buy-to-open paired with sell-to-open, and buy-to-close paired with sell-to-close.


The first definition, and this makes the most sense to me, is the % of trades (share weighted) that involve a "sell-to-open" transaction (but does not include the "buy-to-close" transactions). This would give us a direct measure of the number of new shares "created" in the market by short sellers (while ignoring the shares "destroyed" by the short sellers as they exit positions).

The second definition would be the % of transactions that were either "sell-to-open" OR "buy-to-close", and would tell us something about the activity levels of shorts in the market, but wouldn't tell us whether net shares were increasing or decreasing (if all the shorts were engaged in "buy-to-close", that would look like a lot of activity, when what we're seeing is shorts leaving the market).

Actually a third definition I can think of is the first definition with buy-to-close shares subtracted out. But I don't know how that would turn into a % of shares as reported.


I looked back at the two websites and I haven't yet found something where they talk about the specifics - just the % of short activity. I'm going to do some more digging.

I may have the answer to my question.

This site:
http://regsho.finra.org/DailyShortSaleVolumeFileLayout.pdf

Describes the format for the daily file, and the definition implies to me definition 1 - shares in transactions in which the selling party is sell-to-open.

The single row of data for each day captures the number of shares in sell-to-open (my interpretation) and the total number of shares traded in the stock during the day, and that easily leads to the % of short volume. I wish the FINRA definition was more explicit on this point.

Note that even on a day with >50% of shares shorted, the net number of shares short in the market can decrease. At the extreme, all of the transactions with a sell-to-open can be paired with a buy-to-close, yielding a net change of shares of 0 in these transactions. Meanwhile, some fraction of the sell-to-close crowd can be paired with a buy-to-close, yielding an overall net reduction in shares in total from these shares, leading to an overall no change or even reduction on a day with >50% of activity involving shorting shares (the shorts are trading with themselves, to a large degree ).

(and of course, with >50% of transactions involving shorted shares, its easy for the total beneficially owned shares to increase).


EDIT: Another interesting tidbit from here:
http://www.finra.org/sites/default/files/NoticeDocument/p120044.pdf

The daily short sales % is limited to market hours, 9:30am to 4pm EST. So after hours today isn't rolled into tomorrow, and pre-market - to the degree you can successfully trade - doesn't show up in these numbers.

 

[Papafox]

I may have the answer to my question.

This site:
http://regsho.finra.org/DailyShortSaleVolumeFileLayout.pdf

Describes the format for the daily file, and the definition implies to me definition 1 - shares in transactions in which the selling party is sell-to-open.

The single row of data for each day captures the number of shares in sell-to-open (my interpretation) and the total number of shares traded in the stock during the day, and that easily leads to the % of short volume. I wish the FINRA definition was more explicit on this point.

Note that even on a day with >50% of shares shorted, the net number of shares short in the market can decrease. At the extreme, all of the transactions with a sell-to-open can be paired with a buy-to-close, yielding a net change of shares of 0 in these transactions. Meanwhile, some fraction of the sell-to-close crowd can be paired with a buy-to-close, yielding an overall net reduction in shares in total from these shares, leading to an overall no change or even reduction on a day with >50% of activity involving shorting shares (the shorts are trading with themselves, to a large degree ).

(and of course, with >50% of transactions involving shorted shares, its easy for the total beneficially owned shares to increase).


EDIT: Another interesting tidbit from here:
http://www.finra.org/sites/default/files/NoticeDocument/p120044.pdf

The daily short sales % is limited to market hours, 9:30am to 4pm EST. So after hours today isn't rolled into tomorrow, and pre-market - to the degree you can successfully trade - doesn't show up in these numbers.

Adiggs, I too want to be sure we're interpreting the FINRA.org numbers correctly. For this reason, earlier today I called FINRA but could not get the question answered. I then left a text question with FINRA to get clarification on the numbers used. Let me hopefully get back to you tomorrow if FINRA is quick in answering the question.

 

[adiggs]

And if you're REALLY intrigued by short sales patterns, the same organization publishes a trade level detail each month for the previous month's short sales trade volume. Transactions, shares shorted in round and mixed lots, and price. So August trade level detail file is published by the end of September.

If somebody were so inclined, they could go back in time and look through that transaction level detail for any particular days, and see if the presumed short trading patterns link up with the data to a greater or lesser degree. And maybe find additional patterns in that trading data.

 

[adiggs]

Adiggs, I too want to be sure we're interpreting the FINRA.org numbers correctly. For this reason, earlier today I called FINRA but could not get the question answered. I then left a text question with FINRA to get clarification on the numbers used. Let me hopefully get back to you tomorrow if FINRA is quick in answering the question.

I saw their phone number on one of the pages - I had the thought of calling them as well; I'm glad you called, because a more definitive answer than "this is how I read / interpret it" seems like a good idea

 

[Papafox]

Back to Tuesday's trading...
Looking at the technical chart, you can see that TSLA is zeroing on an upward run on a day with bad macros, which would imply that there's still plenty of steam behind this rally. This week, news allowing, it may run up as far as the upper bb, which stands at 310 and some change. The interesting thing about 310 is that this is a buy point for many conservative investors who want to be sure that the crazy short-selling dip is finally over. Technical traders have said that if TSLA strongly rises above 310, there's very little resistance holding it back from 330 or 340, so if we see some good M3 production numbers start coming out, all of this goodness is well within striking distance. I see TSLA's rise today as just the beginning of a bigger news, given the .strength on a bad macro day.

Now, take a look at the percentage of trading by shorts. I think we're starting to see limited covering by shorts, and it makes sense now for the smarter shorts to give up their positions since there is virtually no evidence that TSLA is going to descend from here. Look at the horrendous FUD that came out in the last month, which now will likely be constrained by Elon's efforts to call out the inaccurate "journalists". Look, too, at the total number of shorts. There's simply no momentum right now in those numbers growing, and in fact the reverse is true. Add less FUD and the beginning of a decline in short holdings and you have a crack in the dike that even a Dutch boy with large fingers can't stop.

For all these reasons I continue to believe that TSLA is ready to recover lost territory, provided that the news from the Consumers Report brake issue is good and Model 3 production numbers respond favorably to the recent production pause. I would imagine that M3 production numbers over 4,000/wk, which are entirely possible soon after the pause is complete, will be a catalyst for positive stock price action with TSLA.

 

[Papafox]

I may have the answer to my question.

This site:
http://regsho.finra.org/DailyShortSaleVolumeFileLayout.pdf

Describes the format for the daily file, and the definition implies to me definition 1 - shares in transactions in which the selling party is sell-to-open.

The single row of data for each day captures the number of shares in sell-to-open (my interpretation) and the total number of shares traded in the stock during the day, and that easily leads to the % of short volume. I wish the FINRA definition was more explicit on this point.

Note that even on a day with >50% of shares shorted, the net number of shares short in the market can decrease. At the extreme, all of the transactions with a sell-to-open can be paired with a buy-to-close, yielding a net change of shares of 0 in these transactions. Meanwhile, some fraction of the sell-to-close crowd can be paired with a buy-to-close, yielding an overall net reduction in shares in total from these shares, leading to an overall no change or even reduction on a day with >50% of activity involving shorting shares (the shorts are trading with themselves, to a large degree ).

(and of course, with >50% of transactions involving shorted shares, its easy for the total beneficially owned shares to increase).


EDIT: Another interesting tidbit from here:
http://www.finra.org/sites/default/files/NoticeDocument/p120044.pdf

The daily short sales % is limited to market hours, 9:30am to 4pm EST. So after hours today isn't rolled into tomorrow, and pre-market - to the degree you can successfully trade - doesn't show up in these numbers.

I had a talk with a FINRA representative today regarding the May 2018 Reg SHO Daily Files data. I just got off the phone with a FINRA employee, trying to get a better feel for what exactly is included in "short volume" and "total volume". These FINRA numbers are the ones that www.shortvolume.com uses for their daily charts. Both numbers have to do with the sell side of a transaction. If a short-seller is buying to cover, that transaction is NOT included in short volume. I am told that short volume includes sales that have a short-seller modifier attached on the transaction that goes to the tape. I asked if the short volume was only "sell to open" transactions, and he said it was simpler than that. It was only sales that had the short-seller modifier attached on the tape version of the transaction. I was told that sometimes it becomes complicated because in instances in which a bank initially does a transaction with itself as a buyer or seller and then later resolves the transaction on the market, several transactions can take place and only one goes to the tape to prevent inflated volume. The takeaway is that we're only looking at the sell side of transactions. No buying to cover by short-sellers is included in the "short volume" number.

I also asked about why FINRA's volume was quite a bit less than the volume NASDAQ reports for a stock every day. I was told that FINRA is looking at only institution to institution transactions and if I wanted the full volume for a stock I would need to gather information from all 12 exchanges. I am not particularly knowledgeable about exchanges and please take my retelling of this answer with a grain of salt because I am capable of making an error.

 

[adiggs]

I had a talk with a FINRA representative today regarding the May 2018 Reg SHO Daily Files data. I just got off the phone with a FINRA employee, trying to get a better feel for what exactly is included in "short volume" and "total volume". These FINRA numbers are the ones that www.shortvolume.com uses for their daily charts. Both numbers have to do with the sell side of a transaction. If a short-seller is buying to cover, that transaction is NOT included in short volume. I am told that short volume includes sales that have a short-seller modifier attached on the transaction that goes to the tape. I asked if the short volume was only "sell to open" transactions, and he said it was simpler than that. It was only sales that had the short-seller modifier attached on the tape version of the transaction. I was told that sometimes it becomes complicated because in instances in which a bank initially does a transaction with itself as a buyer or seller and then later resolves the transaction on the market, several transactions can take place and only one goes to the tape to prevent inflated volume. The takeaway is that we're only looking at the sell side of transactions. No buying to cover by short-sellers is included in the "short volume" number.

I also asked about why FINRA's volume was quite a bit less than the volume NASDAQ reports for a stock every day. I was told that FINRA is looking at only institution to institution transactions and if I wanted the full volume for a stock I would need to gather information from all 12 exchanges. I am not particularly knowledgeable about exchanges and please take my retelling of this answer with a grain of salt because I am capable of making an error.

Very helpful, and the explanation about the complication actually makes sense to me. It was something I bumped into yesterday as I was reading, started to try and explain, and then decided that we'd all be better off without the explanation

(Ok - I can't help but try. Sometimes when somebody issues an order to sell 100 shares, what happens on the back end is more than somebody else buys 100 shares. The more complicated version might be that my broker FIRST issues a sell order of their own for 100 shares in order to find a buyer for the 100 shares. When that buyer is found, then the broker buys my shares from my in order to immediately rebalance their books back to 0 net shares. For a brief period of time, the broker was net short the stock - that intermediate transaction is part of the mechanics of the market function, and DOES NOT show up in the information we see.

So now that you've read this parenthetical, you can also promptly forget about it)


Thanks for the additional detail @Papafox. That means that the direct interpretation of say 60% short volume is exactly what I have been naively assuming - 60% of shares traded on such a day, involved a seller that was selling to open a position (short seller modifier on the tape).

We can't make the next leap, and assume that the open short sale interest increases by that amount - any individual trade that involves a short seller can be paired up with a buy-to-close, thereby keeping the total beneficial ownership in the stock unchanged.

 

[neroden]

Wow. Thanks for the explanation of what the statistic means. 60% is an insanely high percentage of trading by sell-to-open orders. I suppose TSLA longs don't trade as often.

This does imply to me that it won't take much for the stock to shoot up. If the short-sellers run out of ammo, the selling interest drops dramatically, but the buying interest doesn't change -- supply and demand immediately drives the price up.

Pity we don't have insight into the buy-to-close statistics.

 

[Papafox]

Today was a big up day for the broader markets and TSLA benefitted, especially with its own good news added. The broader markets peaked around noon but TSLA kept climbing, likely fueled by the news that Consumer Reports gave Model 3 a "recommend" after the over the air brake fix reduced stopping distances by about 20 feet. Consumer Report's Jake Fisher was blown away by how quickly and successfully the brake issue could be fixed and deployed on vehicles in service. All in all it was a great day for Tesla because the benefits of Tesla innovation really shined today.

Looking at percentage of TSLA selling involving short-selling (I can now say this with confidence), the downward trend of the past several days was broken and that number jumped up to 56% today. Looks like shorts aren't giving up quite yet. We saw 127,000 shares trade hands in the final minute, which strikes me as likely due to shorts recharging their ammo after an active day of trying to cap or walk down the SP this afternoon.

Looking at the technical chart, you can see that TSLA closed above both the 50 day moving average and the mid bollinger band. The stock is still about $18 below the upper bb, so with some more good news there's headroom for additional gains in the short run.

Tomorrow is supposed to be the last day of the Fremont factory Model 3 production line shutdown, and within a week or so we should get a chance to see what the line can run at with the new changes in place. I'm guessing that output above 4K/wk will be enough in early June to push the SP higher. Last night on the market action thread I suggested that the shorts have effectively lost the race to generate another deep dip prior to good news about Model 3 sending TSLA higher. Once TSLA starts climbing above 310, more conservative buyers will consider the deep short-induced dip to be over and will consider jumping in just as shorts who opened positions in the 280s and below feel the pain and some begin to bail. Today's Consumer Reports news helped noticeably. Now it's up to the Model 3 production line. Can't wait.

Conditions:
* Dow up 303 (1.26%)
* NASDAQ up 66 (0.89%)
* TSLA 291.72, up 7.96 (2.81%)
* TSLA volume 7.3M shares
* Oil 68.05, down 0.16 (0.23%)
* Percentage of TSLA selling by shorts: 56%

 

[defc0n]

Once TSLA starts climbing above 310, more conservative buyers will consider the deep short-induced dip to be over and will consider jumping in just as shorts who opened positions in the 280s and below feel the pain and some begin to bail.

That's one thing (and the most important right now, in my opinion) that I agree with you on. I'll be watching what happens between 310-312 like a hawk. After 312 to me means it's off once again to the races.

The million dollar question for me is how high it will go in the medium term. But I'll start worrying more about that later.

 

[cousinHub]

Question : could the AGM coming soon (5th of June) also have an impact for whatever reason?

Specifically, does someone know if large institutional shareholders, who may have lendend shares, are calling them back a few days ahead of the AGM ? To be able to vote maybe ?

Let me give you an example: let's say I am a big institutional investor, I own 1 Million shares for the long term, as a core position in my fund. I tell my broker that he can lend these shares to the shorts community (this way, I'm earning an extra % on my core position). The broker does indeed lend these shares to a hedge fund who shorts them in the market. All the people taking the other side of this trade (buying) are now long 1 Million shares in agreggate.

Who can vote in the AGM ? I would guess it is all the people who just bought the shares from the hedge fund, right?

If I am the institutional investor, owning the initial 1 Million shares, and if I want to vote in the AGM, I need to 'call them back' a few days ahead of the AGM, right ?

Next question (if I'm right in my reasoning) : how many days ahead ?

 

[adiggs]

Question : could the AGM coming soon (5th of June) also have an impact for whatever reason?

Specifically, does someone know if large institutional shareholders, who may have lendend shares, are calling them back a few days ahead of the AGM ? To be able to vote maybe ?

Let me give you an example: let's say I am a big institutional investor, I own 1 Million shares for the long term, as a core position in my fund. I tell my broker that he can lend these shares to the shorts community (this way, I'm earning an extra % on my core position). The broker does indeed lend these shares to a hedge fund who shorts them in the market. All the people taking the other side of this trade (buying) are now long 1 Million shares in agreggate.

Who can vote in the AGM ? I would guess it is all the people who just bought the shares from the hedge fund, right?

If I am the institutional investor, owning the initial 1 Million shares, and if I want to vote in the AGM, I need to 'call them back' a few days ahead of the AGM, right ?

Next question (if I'm right in my reasoning) : how many days ahead ?

You are correct - when I lend out my shares, the specific right that goes with the shares I lend out is the right to vote those shares on any shareholder votes. There will be a record date for determining who votes each share of stock, and if I want to vote, I'll need to recall my shares ahead of that date so that I am the owner of fully paid and not lent out shares if I want to vote.

I don't know how far in advance.

 

[Papafox]

At the annual meetings I have attended, there have from time to time been proposals that challenge the status quo but never get much headway. I suspect if management thought that one of the proposals was a real threat, they'd call the institutional investors and request that they help out. I have heard of no share recalls and I suspect there's no need for them. With issues that are not likely to inspire more than 40% of investors to vote on an issue, Elon can override the rogue group with his votes alone. Anything is possible, but I think we would have seen or heard something by now if management was seriously concerned about any of these proposals.

 

[cousinHub]

Thanks for the explanation.

Still, what is the record date in this case ? someone knows ?

And another question if I may, about "share recalls": if an institutional investor lends a few thousand shares for, let's say, 3 months. Can he call them back suddenly (before the end of the agreed period, without penalty) ?

 

[mongo]

Thanks for the explanation.

Still, what is the record date in this case ? someone knows ?

And another question if I may, about "share recalls": if an institutional investor lends a few thousand shares for, let's say, 3 months. Can he call them back suddenly (before the end of the agreed period, without penalty) ?

April 12th (5 second Google search, or ir.tesla.com)

AFAIK, there is no time frame associated with a sell to open. Shares can be recalled at anytime. However, the pool of shares at a brokerage is large, so the exact source of the initial shares (which person) is not critical. If the borrowed shares were requested back, they could be replenished by other shares at the brokerage (up to the max shortable level they set forth). So the requester gets their shares and the short position stays open.

 

[cousinHub]

Thank you @mongo.

As you can see in my previous message, I did go to the Tesla IR website and found the AGM page Shareholder Meeting | Tesla
but I didn't see the small print below the image & I didn't know that it was the official 'record date':

"Note: If you plan to attend the meeting in person, please bring your photo ID, and either the notice or card addressed to you on behalf of us or your broker or your brokerage statement as of April 12, 2018. Please see the materials you received for details. We will not admit anyone, including guests of shareholders, who does not have these documents."

(I was also wondering if you need to take into account the settlement date or not (t+2) => people who bought the shares on April 12th can vote or not ?

You said 'they could be replenished by other shares at the brokerage' => Yes indeed I can understand this but it seems like the stock is "hard to borrow", which would mean that it would not be so easy eventually, isn't it ?

PS: I'm just trying to learn the inner-workings of all this, sorry if this is not the right thread to ask these type of questions.

 

[neroden]

Thank you @mongo.

As you can see in my previous message, I did go to the Tesla IR website and found the AGM page Shareholder Meeting | Tesla
but I didn't see the small print below the image & I didn't know that it was the official 'record date':

"Note: If you plan to attend the meeting in person, please bring your photo ID, and either the notice or card addressed to you on behalf of us or your broker or your brokerage statement as of April 12, 2018. Please see the materials you received for details. We will not admit anyone, including guests of shareholders, who does not have these documents."

You said 'they could be replenished by other shares at the brokerage' => Yes indeed I can understand this but it seems like the stock is "hard to borrow", which would mean that it would not be so easy eventually, isn't it ?

Yeah, but TSLA isn't that hard to borrow right now. When it starts to be hard to find shares, the brokerage (a) raises interest rates charged to short-sellers, inducing some of them to close their positions, and (b) starts paying interest (more and more) to people who lend their shares out, inducing more people to lend their shares out.

SolarCity got up to exorbitant interest rates: you could earn 12% lending it out. At that rate, a lot of people will lend their shares. Right now, TSLA lending isn't paying much of anything.