Thanks for your effort.
Anything we can do to spread the truth on Tesla will lead to a fair valuation of the company and less volatility in the stock price.
-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
- Status
Thanks for your effort.
Anything we can do to spread the truth on Tesla will lead to a fair valuation of the company and less volatility in the stock price.
Can you share more about your VIN analysis? When in the process does Tesla assign a VIN and how long after is the car produced and then delivered? Does it vary by model? How long after an assigned VIN are European customers receiving their cars? Also, what's your best guess with Q3 production based on your VIN analysis, and can you share how you reach that number?
Just a reminder to everyone: Whatever you think the Q3 production figure is, you have to subtract the cars in transit to Europe before you have a number for deliveriese. (Or more precisely, the increase in cars-in-transit).
Also, I have seen no discussion of the train that was derailed in Texas, resulting in the totalling of a number of TMS that were on their way to Europe. I have no idea how many cars were affected - a couple of guys on the Norwegian forum were affected, but didn't seem like a huge number.
Elon has previously guided that they would exit 2013 with a production rate of maybe 24k/year (from memory). I think that doing that rate on average in Q3 would be a great result. That would mean a production of 6k cars (despite the first week of July being off). If EU-bound cars are 1k, then that would mean 5k deliveries. I would be impressed by that number. I think 6-6.5k figures being circulated here would be completely breakout results, and I am very skeptical of those levels.
EDIT: I think we will be quite astonished on the Average Sales Price (ASP). European cars tend to be very high specced on average.
Also, I have seen no discussion of the train that was derailed in Texas, resulting in the totalling of a number of TMS that were on their way to Europe. I have no idea how many cars were affected - a couple of guys on the Norwegian forum were affected, but didn't seem like a huge number.
Elon has previously guided that they would exit 2013 with a production rate of maybe 24k/year (from memory). I think that doing that rate on average in Q3 would be a great result. That would mean a production of 6k cars (despite the first week of July being off). If EU-bound cars are 1k, then that would mean 5k deliveries. I would be impressed by that number. I think 6-6.5k figures being circulated here would be completely breakout results, and I am very skeptical of those levels.
EDIT: I think we will be quite astonished on the Average Sales Price (ASP). European cars tend to be very high specced on average.
Mario Kadastik
Active Member
I think the train wreck is not a problem. You can consider them instant deliveries because insurance pays for them and Tesla had to remanufacture the car again. If the crash was for batteries only, well then the cars were delayed, but the cost in the crash is recovered so any profit from equipment on the train would be doubled.
smorgasbord
Active Member
It's almost a certainty that they skipped between the Sig and Production series - that was reported on TMC contemporaneously at the time back when almost everyone getting a Model S was on TMC. But, probably just about hundred, and like you say, doesn't really affect anything, especially if we look at the delta from last quarter.
I think that with production constraints, that's a reasonable point as they'd be more likely to insure at a value that covers lost profit. It would depend on premiums. In the future, if Tesla gets beyond constraints they'd lower their premium and only insure for cost.
Ok, I'll chip in my estimated deliveries for Q3: 6100. I have tracked VINs, not in terms of assignment, but in terms of actual deliveries, as I think this (surprise) captures deliveries better ;-)
Was very close on deliveries in Q2, but it is more uncertain now due to the EU pipe. I have actually subtracted almost 1000 cars net increase of the delivery pipeline in Q3 and some additional loaners.
Finally, it is rather complex to estimate actual average this Q due to the late arrival of VINs to EU. You see this clearly in the different cluster of data points in the graph, you also see clearly the batching going on (blue circles).
I'm fairly certain of 6100, as this seem to be a reasonable surprise to continue the momentum toward some 23000 deliveries for the year (what my data point at right now). Notice the progressively higher surprises Q to Q. You cant reasonably expect more progress than this. However, it might be even better if the pipe to EU is really well managed. I.e., the EU cars delivered in Q3 were (very strategically) mostly batched (i.e., built) in the middle of Q3, so it might be that fewer are in the pipe!
Also, I believe that the ASP will be great, my guess is 95kUSD, leading to non GAAP revenue of 664 m (GAAP 523).
For the bottom line I only think there will be very slight non-GAAP profit, and a GAAP loss on par with Q2, due to the rapid expansion of everything, and associated increasing costs. I also think Elon et al. were very cautious of us expecting to much of a profit for Q3 in the Q2 conference call. I think they will make it, but not by much. So, on average, unless there is surprise again in the revenue figures (I have been to conservative with these before) the market is now quite on par. (Yahoo finance say average (non-GAAP) revenue will be 533m). I don't expect a huge surprise this time, which is also reasonable, as the market now have had a full year of surprises to digest, they should be better at estimating Tesla's books by now.
/Buran
Additional legend for the figure:
Black vertical lines = Qs
Black horizontal lines =Guidance delivery
Green horizontal lines, past time = actual delivery
Green horizontal lines, future time = my estimates
Black trend lines: Top = naked eye linear regression based on my feeling of how much is end of the Q rush.
Black trend lines: Bottom = Same slope as top line, centered on EU deliveries
Black trend lines: Mid = Same slope as top and bottom line, placed at 2/3 US deliveries, and 1/3 EU deliveries (reported balance between them at TMC as far as I can judge)
Was very close on deliveries in Q2, but it is more uncertain now due to the EU pipe. I have actually subtracted almost 1000 cars net increase of the delivery pipeline in Q3 and some additional loaners.
Finally, it is rather complex to estimate actual average this Q due to the late arrival of VINs to EU. You see this clearly in the different cluster of data points in the graph, you also see clearly the batching going on (blue circles).
I'm fairly certain of 6100, as this seem to be a reasonable surprise to continue the momentum toward some 23000 deliveries for the year (what my data point at right now). Notice the progressively higher surprises Q to Q. You cant reasonably expect more progress than this. However, it might be even better if the pipe to EU is really well managed. I.e., the EU cars delivered in Q3 were (very strategically) mostly batched (i.e., built) in the middle of Q3, so it might be that fewer are in the pipe!
Also, I believe that the ASP will be great, my guess is 95kUSD, leading to non GAAP revenue of 664 m (GAAP 523).
For the bottom line I only think there will be very slight non-GAAP profit, and a GAAP loss on par with Q2, due to the rapid expansion of everything, and associated increasing costs. I also think Elon et al. were very cautious of us expecting to much of a profit for Q3 in the Q2 conference call. I think they will make it, but not by much. So, on average, unless there is surprise again in the revenue figures (I have been to conservative with these before) the market is now quite on par. (Yahoo finance say average (non-GAAP) revenue will be 533m). I don't expect a huge surprise this time, which is also reasonable, as the market now have had a full year of surprises to digest, they should be better at estimating Tesla's books by now.
/Buran
Additional legend for the figure:
Black vertical lines = Qs
Black horizontal lines =Guidance delivery
Green horizontal lines, past time = actual delivery
Green horizontal lines, future time = my estimates
Black trend lines: Top = naked eye linear regression based on my feeling of how much is end of the Q rush.
Black trend lines: Bottom = Same slope as top line, centered on EU deliveries
Black trend lines: Mid = Same slope as top and bottom line, placed at 2/3 US deliveries, and 1/3 EU deliveries (reported balance between them at TMC as far as I can judge)
Attachments
I feel compelled to mention the train derailed in Nebraska and not Texas. Although I could imagine most people in Norway wouldn't know the difference between Texas and Nebraska. For that matter many Americans who live on the coasts don't know the difference either.
Actually, many Norwegians would know (when you like in a small country, geography abroad becomes more important). However, the following article was posted on the forum along with the laments from those who lost their cars: http://www.usatoday.com/story/news/nation/2013/09/25/train-texas-amarillo-crash/2867843/
I assume that someone just google-newsed this and thought it was a good fit. Actually, I was wondering a bit about the train route, but just assumed there was some logic in going via the Gulf of Mexico or something like that.
Chickenlittle
Banned
Ah, I see. This is the derailment I had heard of:
http://columbustelegram.com/news/lo...cle_3e1fb644-23ae-11e3-8064-001a4bcf887a.html
That one actually makes more sense since it wouldn't make much sense for Teslas going from the San Francisco area to the port of New York to go through Texas. Actually, someone who would seem to be the husband of a Tesla employee posted a pretty convincing "fictional" story about electric motorcycles and the state of Bebraska and country of Borway on the Tesla forums that essentially confirmed that was the train. However, he has since deleted that post.
StapleGun
Member
Excellent data and analysis Buran!
I'm intrigued by the big clump of non-EU VINs in the 21000-23000 range right before the Q3 line. Looks like a dead give-away that they rushed out quite a few cars in the US just before the end of the quarter, but there's also a similar smaller clump of EU deliveries too. To me this all looks like a very calculated effort to maximize Q3 numbers. I agree with most of your estimates but I think 1000 cars in-transport is too high.
Also, I haven't heard much about the loaner situation lately but if they are still consistently sold out then it shouldn't have much of an effect on the quarterly numbers.
I'm intrigued by the big clump of non-EU VINs in the 21000-23000 range right before the Q3 line. Looks like a dead give-away that they rushed out quite a few cars in the US just before the end of the quarter, but there's also a similar smaller clump of EU deliveries too. To me this all looks like a very calculated effort to maximize Q3 numbers. I agree with most of your estimates but I think 1000 cars in-transport is too high.
Also, I haven't heard much about the loaner situation lately but if they are still consistently sold out then it shouldn't have much of an effect on the quarterly numbers.
RationalOptimist
Member
There's one wrinkle in using VIN data to estimate Q3 revenues that hasn't been talked about much. Most people seem to be making the assumption that VINs are issued just before production of the car. So if production + delivery = average of 3 weeks in US, they assume that VIN issued between first week of June and first week of Sept are an approximate proxy for Q3 deliveries (less some number for net increase in European cars in transit, which many are pegging at 1000.) This gives sales estimates of 7,000 or higher for the quarter, which suggests a massive earnings blowout.
But there's no guarantee that issuance of a VIN means production about to start. It may simply mean that a car is scheduled for production. If demand is increasing faster than production is scaling up, it may well be that the time period between VIN and production is increasing. That would mean that sales numbers will fall further and further behind VIN issuance. On Craig's thread over at Tesla Motors, there are signs that for both Americans and (especially) Europeans, the time between VIN issuance and production is lengthening to as much as 2 months. If so, that's an amazing story on demand acceleration, but may not fully translate to sales.
It's legitimate to be excited about the acceleration in demand and production... but we shd be wary in our assumptions of how many of those VINs will make it into Q3 sales. (I'm long Tesla, and still believe Q3 will be terrific. Just concerned that some are getting overexcited.)
But there's no guarantee that issuance of a VIN means production about to start. It may simply mean that a car is scheduled for production. If demand is increasing faster than production is scaling up, it may well be that the time period between VIN and production is increasing. That would mean that sales numbers will fall further and further behind VIN issuance. On Craig's thread over at Tesla Motors, there are signs that for both Americans and (especially) Europeans, the time between VIN issuance and production is lengthening to as much as 2 months. If so, that's an amazing story on demand acceleration, but may not fully translate to sales.
It's legitimate to be excited about the acceleration in demand and production... but we shd be wary in our assumptions of how many of those VINs will make it into Q3 sales. (I'm long Tesla, and still believe Q3 will be terrific. Just concerned that some are getting overexcited.)
Last edited:
Mario Kadastik
Active Member
I think you are basing this somewhat on what I wrote there and it's true. I got assigned a VIN on Oct 4th and according to Tesla it's likely my car will enter production end of December / early January. This means that there's 2-3 month difference between VIN assigned to actual production start for some EU deliveries. The same goes for a friend of mine who's ordered an S60 in May, got the VIN a few weeks before me and his estimate is end of November/early December production start. So at least some EU VINs will be assigned months before cars even enter production. Not sure how profound an effect it is though.
Mario Kadastik
Active Member
Well if all the investors take this is pure truth, then the play for earnings changes. We just have to hedge the day of earnings or 1-2 days before because such news should cause a serious run up pre-earnings so it'll be a buy the rumor sell the news type of play. Tesla will correct the understanding of people in the ER and once everyone settles back the run will continue. It's just a question mostly of wether the run up to earnings will be the main action or after ER. If you're already invested now, then it hardly matters which one will be realized
I would not be concerned about this at all. This is all but forgone conclusion that Q3 production is above 7,000 units as the average production in Q3 exceeded 600 cars/week. The unknown is total deliveries in Q3, as discussed extensively on this Forum.
Well, just consider the information that is out on this forum. In one of the deliveries threads there were reports about production reaching 600 cars/week by the beginning of Q3. Several posts also reported 700 cars/week starting first half of September. Given 12 production weeks in Q3, and making very conservative assumption of average production of 600 cars/week yields 7,200 cars produced in Q3.
- Status
Similar threads
