2017 Investor Roundtable: TSLA Market Action

[Waiting4M3]

My math is wrong. IF cars are delivered 15 days after production, this would mean VIN 259 came off the line sept 8, thus making production for the first week of Sept 129. That's not bad, but who's counting?

It also takes a few days to ship it out to Austin. M3 may also need a longer QA process compared with MS/X, so 15 days delay between production and delivery in Austin could be plausible. Still, hoping to see higher VIN next week.

 

[RiverCard]

Here's more on the FSLR matter (that "upside" is more than double what it is now)

View attachment 249186

Is the solar part of Tesla in a different position than FSLR (at least until the Buffalo plant is producing in volume)?

"The U.S. government imposes antidumping and countervailing duties on solar cells manufactured in China and/or Taiwan. Based on determinations by the U.S. government, the antidumping and countervailing duty rates range from approximately 33%-255%. Such antidumping and countervailing duties are subject to annual review and may be increased or decreased.

In addition, the U.S. government imposed additional tariffs on solar modules manufactured in China (with solar cells manufactured in other countries) and solar cells manufactured in Taiwan. In early January 2015, the U.S. government announced its affirmative final determinations in both the countervailing duty and antidumping cases against China and in the antidumping case against Taiwan.

Since these tariffs are reflected in the purchase price of the solar panels and cells, these tariffs are a cost associated with purchasing these solar products. In the past, we purchased a significant number of the solar panels used in our solar energy systems from manufacturers that were based in China. We continue to be affected by these tariffs/duties and any changes to them as many of the solar panels we currently purchase contain components, including solar cells, from China and Taiwan.

If additional tariffs are imposed or other negotiated outcomes occur, our ability to purchase these products on competitive terms or to access specialized technologies from countries like China and Taiwan could be limited. Further, foreign suppliers in other countries could also be the subject of these or future trade cases. Any of these events could harm our financial results by requiring us to account for the cost of trade penalties or to purchase and integrate solar panels or other system components from alternative and potentially higher-priced sources.

In September 2014, we acquired Silevo, a solar panel technology and manufacturing company with manufacturing operations in China. In May 2015, we were contacted by the U.S. Customs and Border Protection (CBP) to provide information regarding our importation of solar panels from China. Based upon the information we provided, CBP agreed that our solar panels were not subject to the scope of antidumping and countervailing duty Orders on Crystalline Silicon Photovoltaic Cells/Modules from China (Orders); however, CBP instructed us to obtain a determination from the Department of Commerce (DOC). In June 2016, the DOC issued its final determination finding that Silevo’s solar panels, manufactured in China, are within the scope of the antidumping and countervailing Orders. We are currently challenging that final determination in the Court of International Trade (CIT). However, until this issue is resolved, we have not and do not intend on importing Silevo solar panels produced in China for our U.S. operations without depositing the applicable antidumping and countervailing duties."

 

[GoTslaGo]

I don't quite understand what you're suggesting is going on with the convertibles. Could you explain?

The last 10Q still had around 50-60k of the convertibles still unconverted (IIRC). TSLA had to account for these on their books. These would be the last batch of the super cheap convertibles from the DOE loan that were left. I figure that those are the ones that Fidelity owns (interestingly I own the convertible fund and it showed up there on the last statement I got from Fidelity--honestly I never really noticed until recently).

I would have assumed they would have slowly petered them out, but you never know. I just recalled that the last 10Q stated that 1.2 million (?) of the convertibles were exercised in June--presumably at the previous peak. So presumably the last batch may have been converted the last couple of days. Numbers that @zdriver saw would fit (he saw a huge 50k order come in to sell).

It would be no big deal to sell at market because their basis was so much lower than any SP we have now, even 350 (just like how Tesla execs who get shares allocated for meeting targets likely do the same--they got the shares for "free", so anything is gravy). I don't know the details of how the converts work--can you convert a little at a time or have to do it in a big batch? Does the fact that they have to go to Tesla to convert the shares limit how often they would do this? I don't know.

Nonetheless, presumably that many being sold as a large order would bring the SP down a lot (since I recall most people/funds/hedges try to sell in 100 share lots to keep from being too noticeable and move the market). This in turn would make paranoid traders... more paranoid, and they would start to sell, add in the FUD, etc... so on and so on... (the thinking would be: someone big is selling big, thus something bad is afoot... we must sell too! Then add in the algobots who are programmed to do the same).

Obviously I have no idea if this is correct, but would be happy if those convertibles finally get executed and "disappear". By the way these were the same issues that killed SP last year after SCTY/TSLA merger was announced. It showed up on FMR LLC Edgar .... or more appropriately, a large batch disappeared after the 3rd quarter from FMR's holding.

 

[geneclean55]

My math is wrong. IF cars are delivered 15 days after production, this would mean VIN 259 came off the line sept 8, thus making production for the first week of Sept 129. That's not bad, but who's counting?

Lots off guesswork obviously, but 129 isn't far off @austinEV estimate

Another important parameter is the starting point. I just grabbed 5/week to be low but here is 10: Note it fits the July output better and fits the 2 points better:

View attachment 247291

Which is also a more reasonable 68% per week. It also shows that 15/day is marginally behind, but not really since that is like a day +/-.

 

[Intl Professor]

Buy and hold investors are laughing at us.

No, just not worried by the dips and worrying about your losses.

 

[zdriver]

I would have assumed they would have slowly petered them out, but you never know. I just recalled that the last 10Q stated that 1.2 million (?) of the convertibles were exercised in June--presumably at the previous peak. So presumably the last batch may have been converted the last couple of days. Numbers that @zdriver saw would fit (he saw a huge 50k order come in to sell).

Dark pool net selling today alone was around 135k shares, plus another net selling of ~30k shares on liquidity. (20k short covering near EOD) If memory serves, net dark pool selling the last few days was around 290k shares. Keep in mind this is almost certainly a very conservative number, as just being tape reading software, it can't correctly identify every dark pool trade, so it errs on the side of caution.

 

[Krugerrand]

Ah. Short term call plays, very tricky.

See, I bought 200 shares of stock which I intend to hold long-term: I'm an accumulator of TSLA and will be until at least 2019, so it's a question of "accumulate" or "wait to accumulate". (I also sold some $400 puts but they don't expire until November; we'll see how they turn out, but I'm actually hoping to get assigned.) So basically I just missed out on buying the same amont of stock for $6000 less. I should clarify that this purchase increased my overall position by less than 3%, and all my other lots are still deeply "in the green".

I don't particularly regret it -- I'm sleeping better having a slightly larger position going into next year. But I do have to remember in future that if part of my brain is saying "No, really, this feels like a local top, you'll probably have a better buying opportunity next week", that part is probably right. It was saying this when I bought at $279 in late 2014 out of FOMO, too, so it's sent the right signal twice, but I listened to a different part of my brain. Lesson for the future. It also said this at the last $380 top and I got that one right, buying at the bottom of the next dip.

You might want to talk to someone about all those voices.

 

[bdy0627]

Ok, did some screenshots of the last 6 days of trading. Click to embiggen.

First up is Algo liquidity buying/selling, aka market orders, as I understand it. I personally correlate this with *urgent* MM delta hedging of calls/puts. Note how it follows price, making step changes when price makes larger changes. Beyond MM algos, this likely also includes more aggressive trading algos.
On interesting spike was late last Friday, which is to be expected given effects of gamma so close to expiration.
View attachment 249311

Next up is dark pool liquidity. I find this one interesting as it shows very clear buying and selling spikes, including 3 different days of late day short covering, done at times of high liquidity such that price is not impacted as much by covering. These spikes are most likely from trading firms, not institutions selling positions.

View attachment 249313

Next is Market Pressure. This is the one I didn't listen to. I loaded up on weekly calls on the dip Monday when this spiked, but I should've been out of weekly calls by noon yesterday when this accumulator wasn't improving and still down trending. Furthermore, I should've been out of medium term calls (Oct, Nov, Dec, J18) this morning when things kept going downhill, rather than staring at it in disbelief. (or whatever I was doing). But really, I should've bought weekly puts yesterday when MP kept heading down, along with all the selling on buy/sell...

View attachment 249314

Next is algo buy/sell. This tries to identify algo related trades - trades that are put out on the order book as limit bid or offers. This is another major directional indicator for me during the day, and another I ignored this week. Tuesday is a common look for downgrade days. The spike at the end of today is common on days like this, I'm guessing mean reversion algos - that range commonly trades back out (is sold) if the beginning of the next trading day doesn't make rapid upward progress, so if we're down Monday, that could add a bit of insult to injury.

View attachment 249317

Finally, dark pool buy/sell. This one is commonly always down, so I usually don't pay much attention unless it's doing anything out of the ordinary. There's always institutions selling, especially in an uptrend, and they typically do it carefully for obvious profit reasons, and I suspect many institutions accumulate shares using buying algos, so their activity may not even show up here. Out of the ordinary, I would classify as the last 3 days. Monday and Tuesday look pretty typical for this accumulator, as does last Friday, maybe with the exception of that sell spike. The buy spike yesterday before close was a bit interesting, I'd probably classify that as short covering by a trading firm.

View attachment 249318

I'm leaving raw tick out because it's basically useless over long time periods. I only use it for context intraday, no more than 5-15 minutes at a time.

Finally, if this software looks helpful, do consider trying it out for a month. Well worth it if you swing or daytrade much. HFTAlert (@HFTAlert) | Twitter

I was surprised after the big drop over the last few days that shorts didn't appear to cover near close, thereby causing the share price to climb a bit. The likelihood that the late chasing shorts are going to get reamed is pretty high. I think smarter shorts will be looking to take profits soon with the stock bouncing up and down so hard. Once it reverses, it's going to get ugly quickly for them.

 

[Navin]

Besides the wild drop in the stock price - nothing changed from the tesla story that existed on Monday when the stock was at an all time high. Therefore - I bought more Friday. Don’t think we will get to 1500 in sept for model 3 but - it’s not a demand issue. It’s a supply issue that will eventually resolve itself. We are literally at that inflection with tesla when *sugar* is about to get real. Don’t let the shorts and Wall Street monkeys shake the shares from you. Trade em well.

 

[kenliles]

You might want to talk to someone about all those voices.

I think he has,
but that one doesn't have a name yet--

 

[bdy0627]

Dark pool net selling today alone was around 135k shares, plus another net selling of ~30k shares on liquidity. (20k short covering near EOD) If memory serves, net dark pool selling the last few days was around 290k shares. Keep in mind this is almost certainly a very conservative number, as just being tape reading software, it can't correctly identify every dark pool trade, so it errs on the side of caution.

Could you give a quick explanation on the differences in these? Dark pool selling is sort of behind the scenes, right? What's the purpose of it? Is liquidity simply market selling? Thanks for all this info.

 

[brian45011]

The last 10Q still had around 50-60k of the convertibles still unconverted (IIRC). TSLA had to account for these on their books. These would be the last batch of the super cheap convertibles from the DOE loan that were left. I figure that those are the ones that Fidelity owns (interestingly I own the convertible fund and it showed up there on the last statement I got from Fidelity--honestly I never really noticed until recently).

I would have assumed they would have slowly petered them out, but you never know. I just recalled that the last 10Q stated that 1.2 million (?) of the convertibles were exercised in June--presumably at the previous peak. So presumably the last batch may have been converted the last couple of days. Numbers that @zdriver saw would fit (he saw a huge 50k order come in to sell).

It would be no big deal to sell at market because their basis was so much lower than any SP we have now, even 350 (just like how Tesla execs who get shares allocated for meeting targets likely do the same--they got the shares for "free", so anything is gravy). I don't know the details of how the converts work--can you convert a little at a time or have to do it in a big batch? Does the fact that they have to go to Tesla to convert the shares limit how often they would do this? I don't know.

Nonetheless, presumably that many being sold as a large order would bring the SP down a lot (since I recall most people/funds/hedges try to sell in 100 share lots to keep from being too noticeable and move the market). This in turn would make paranoid traders... more paranoid, and they would start to sell, add in the FUD, etc... so on and so on... (the thinking would be: someone big is selling big, thus something bad is afoot... we must sell too! Then add in the algobots who are programmed to do the same).

Obviously I have no idea if this is correct, but would be happy if those convertibles finally get executed and "disappear". By the way these were the same issues that killed SP last year after SCTY/TSLA merger was announced. It showed up on FMR LLC Edgar .... or more appropriately, a large batch disappeared after the 3rd quarter from FMR's holding.

At the end of June, $60.17 million of 2018 notes remained unconverted. At 8.03 shares/$1,000 note, that's about 483,200 shares. The normal early conversion process is:

- the holder gives Tesla notice,

-Tesla has a few days to tell the holder how the value above conversion at par ($124.52) will be settled between cash and shares; the principal is settled in cash.

-the value to be paid to the holder is then determined by using the Volume Weighted Average (share) Price over the next 20 trading days.

The hedge parties are exposed between $124.52 and $184.84, above that the risk (and dilution) reverts to Tesla. Holders can convert after 3/1/18 (without going through the valuation process) and the notes mature on 6/1/18.

The 5/31/17 agreement to exchange 1.16 million shares for $144.8 million in 2018 notes short-circuited the valuation process. It's unclear how the obligations of the hedge parties were resolved. If there is a similar agreement for the remaining ~$60 million, an 8k will be filed.

Conversion at par share prices for the 2019/2021 Notes is $359.81 and for the 2022 Notes is $327.50. Above these levels, holders have a near risk-free interest arbitrage play by shorting the common shares. None of these issues have qualified for early conversion.

 

[vgrinshpun]

Conversion at par share prices for the 2019/2021 Notes is $359.81 and for the 2022 Notes is $327.50. Above these levels, holders have a near risk-free interest arbitrage play by shorting the common shares. None of these issues have qualified for early conversion.

I do not understand how this would be an attractive strategy.

For the original holder of the note shorting at price higher than the conversion price of $359.81 (2019/2021) or 327.50 (2022) would guarantee profit of the difference between price at which short is initiated and conversion price, at the expense of reducing the interest by the interest charged for shorting (which actually can potentially exceed the interest of the notes) and by giving up any upside from the SP at which short was initiated. This does not seem to fit the oulook on the future of TSLA that an original conversion note holder would have.

Can you please explain your thinking?

 

[KLC13]

Interesting M3 video. I'm not a tesla owner, is this similar to the controls on the S or X?

(VIN# 259)

 

[EnzoXYZ]

Looks like Software for the 3 is still work in progress. Makes sense to pull from car of the year.

I want my 3 now!

 

[geneclean55]

My math is wrong. IF cars are delivered 15 days after production, this would mean VIN 259 came off the line sept 8, thus making production for the first week of Sept 129. That's not bad, but who's counting?

Actually that car belongs to RiggerJon poster from M30C.
Looking through his posts, he was told on Aug31st that he would receive his car between Sept14-21. He received it either Sept 21 or 22.
Also, his posts strongly suggest that it was manufactured sometime during first week of September.

 

[Lump]

Towards the beginning of the video (.30 mark) something about a 12v battery issue is quickly mentioned, whats that about?

 

[KLC13]

Towards the beginning of the video (.30 mark) something about a 12v battery issue is quickly mentioned, whats that about?

I am not sure, but as @EnzoXYZ pointed, it seems the M3 is still WIP (although minor issues but still need to be ironed out). I was surprised that if there are 300+ M3's on the road out there then why we haven't seen a complete walkaround, drive and review of the car by anyone yet.

Has Tesla signed any agreements that prevent employees from doing such reviews yet? It would be easy to enforce this agreement if you are employed by the company and know the consequence

 

[erthquake]

Towards the beginning of the video (.30 mark) something about a 12v battery issue is quickly mentioned, whats that about?

I was wondering the same thing, but it sounds like it's just a software issue. And the Austin car was one or two software updates behind the SoCal non-employee owner. 2017.34 versus 2017.36. So it may already be fixed.

 

[Lump]

Looks like Software for the 3 is still work in progress.

Coming up on 5 years & the Model S software is still work in progress.