Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register

EV-CPO.com CPO Consolidator support thread

This site may earn commission on affiliate links.
It is odd as writing code to rip a web site must surely cost more than paying a few dollars to gain access. And I appreciate you've done a lot of work, as have all the consolidators, so yes, having your work stolen can't be any fun.
I’m just curious how the companies that were ripping off Hank’s data compilation could profit from it? My mind can’t grasp that. It seems that this would only be useful for people like us, looking for a good deal on a used Tesla that fits our particular wants. Is it for reselling? That doesn’t seem right because third party resold Teslas are usually slightly cheaper (but without extended warranty).
 
I’m just curious how the companies that were ripping off Hank’s data compilation could profit from it? My mind can’t grasp that. It seems that this would only be useful for people like us, looking for a good deal on a used Tesla that fits our particular wants. Is it for reselling? That doesn’t seem right because third party resold Teslas are usually slightly cheaper (but without extended warranty).

The firms I know about (there are probably more that I don't know about) were downloading the current and past status of the inventory to plug into their investment models. These are highly sophisticated hedge fund firms that have analysts whose only job it is to know EVERYTHING there is to know about a company in their portfolio, and whether to advise their clients to increase or decrease their positions in certain stocks (or do it for them for managed funds). Hedge funds pay huge sums of money for data in order to gain even the slightest advantage over any other company. Clearly the number of cars that Tesla is adding to, and removing from, their inventory is a critical number to understand for these models. That's how they profit off of the information on EV-CPO.

Here is an sample analysis I did for one such firm that paid for me to create this kind of information:

upload_2019-3-7_21-21-27.png
 
  • Informative
Reactions: dhanson865
The firms I know about (there are probably more that I don't know about) were downloading the current and past status of the inventory to plug into their investment models. These are highly sophisticated hedge fund firms that have analysts whose only job it is to know EVERYTHING there is to know about a company in their portfolio, and whether to advise their clients to increase or decrease their positions in certain stocks (or do it for them for managed funds). Hedge funds pay huge sums of money for data in order to gain even the slightest advantage over any other company. Clearly the number of cars that Tesla is adding to, and removing from, their inventory is a critical number to understand for these models. That's how they profit off of the information on EV-CPO.

Here is an sample analysis I did for one such firm that paid for me to create this kind of information:

View attachment 384051

So why would they not go to the sites that have the hidden stock? You only list Teslas publicly listed stock, and while you make a strong argument for a buyer that if Tesla list it, its more likely to be real and available, the other sites list stock which all have valid Tesla URLs, some have valid locations, and we suspect that Tesla massage their public inventory listings a lot to give the appearance of low stock (for instance they have heavily reduced inventory listings at times only for them to reappear).

Maybe they do go to all the sites, but I if I was looking for a market edge, I'd want to know more about the hidden stock than the public stock. I've seen some of the other sites on Linkedin give some rough, high level data stats, maybe their business model isn't to profit from buyers (few others have a subscription model), they sell their data into the trade with analysis on the data such as the rate of vin changes (like the bloomberg M3 tracker), demographic distribution around the world, model distribution, mean time to sale etc?
 
It appears there is a glitch in the website where the Florida cars are listed as Corte Madera, CA. I got excited that so many were in my area, but alas, they're not.

Yes, late last night Tesla started reverting the location back to regions instead of specific locations, and I haven't had a chance yet to catch up with it. It should be corrected later today.
 
So why would they not go to the sites that have the hidden stock?

There's a well known term in Computer and Data Sciences: Garbage In, Garbage Out.

If I were managing a billion dollar hedge fund, my investment models wouldn't take any data that's even the least bit suspect, unknown or from hacked sources. These firms can tell quality data sources from the chaff.
 
There's a well known term in Computer and Data Sciences: Garbage In, Garbage Out.

If I were managing a billion dollar hedge fund, my investment models wouldn't take any data that's even the least bit suspect, unknown or from hacked sources. These firms can tell quality data sources from the chaff.

You are very critical of your competition, and naive of hedge funds, presumably as a sponsor on here you can comment freely whereas I presume they don't pay. But if I had a choice of 5000 valid vins globally, each where the Tesla website responds to the link with a detailed specification and where the car is in the world, and I can understanding the time between Tesla releasing the VIN and the car disappearing to an owner, and the level of discounts that greater number of cars was listed at, and where those vins must represent a sizeable proportion of the Tesla MS and MX production, or I can look at a relatively small subset of cars in the world that reflects a manipulated representation by Tesla to try and give a false impression, I would take the larger dataset. Each to their own. Either way, ripping of your data is wrong.
 
  • Like
Reactions: Mikeisnomech
, and naive of hedge funds,

Really? How do you get that? I've worked with hedge funds and investment companies, and they take their data very seriously. To do otherwise is putting billions of dollars of their customers' money at risk.

f 5000 valid vins

There's no guarantee they are "valid VINs". They're VINs, yes, and they are technically "valid". But in many (most) cases, those cars aren't even built yet. Or in pre-or-mid-production. Or built and sitting on a lot. Or built and being shipped out. That's the problem with that data, There's no way to tell which ones are actual built cars, or what the disposition is. Garbage in.

the Tesla website responds to the link with a detailed specification and where the car is in the world,

As I've noted many times, the location is meaningless if the car isn't even built yet, or sitting in a lot in Fremont, CA awaiting shipment.

and I can understanding the time between Tesla releasing the VIN and the car disappearing to an owner,

You're making a lot of assumptions there. The time between a VIN and configuration being created (and a webpage being available) has no relation to the car being built, available for sale, shipped to the destination, or sold (or just unlisted for sale, those are different things).
 
My work computer has 3X 4K monitors but your Web site always complains about opening the window in landscape mode :p My browser window on a 4K display probably has more horizontal pixels than most browser windows open in full screen mode on lower resolution displays... Can your code only show this warning for 1080P and lower resolution displays?

Ok, please try it now. Thanks for the feedback.
 
  • Love
Reactions: PhilDavid
So the browser window height (~1944) being more than the browser window width (~1260) is what was triggering the message. I don't have a 4k display, so I can quite visualize what that would look like, but the code was just looking for displays where the height was greater than the width (like most cell phones in portrait mode).
 
So the browser window height (~1944) being more than the browser window width (~1260) is what was triggering the message. I don't have a 4k display, so I can quite visualize what that would look like, but the code was just looking for displays where the height was greater than the width (like most cell phones in portrait mode).

That explains it... For my software development, I just switched to a 3X 4K 60hz display setup and I can't imagine how I managed to work before... Try it sometime! Also just setup 2X Stream Decks for all my shortcuts and macros and it's been great for productivity.

Have you talked to the Only Used Tesla folks to see if you can include their listings as well? They have a great Web site but the interface is a bit silly with no way to sort listings based on when listed or the listed amount...

Keep up the great work!
 
Thanks for your support!

Another update: Recently Tesla has started reporting the price differently in the full data stream versus the individual car detail pages. The stream was not including the 1200$US destination fee, but the code in the car detail page was. So for instance, VIN 5YJSA1E25KF305536 was coming across with an $89,000 (correct) price, but when cross-checked against the detail page, the price was $90,200 including the destination charge. The end result was that the site was sending out constant alerts with a whipsaw $1200 increase followed by a $1200 decrease. That's now fixed so the repeated alerts should be stopping now.

My first thought when I saw this oddity was that Tesla was A/B price testing in different regions..so the same car would be priced differently. Luckily it wasn't that (yet).

edit: It looks like they made this change a few days ago. This is what it looks like:
upload_2019-3-11_0-38-4.png
 
Last edited:
  • Informative
Reactions: SmartElectric
P85D looked to be a great price but was a U.K. listing in GBPounds but I was searching USA
5YJSB8H43FF096936
Model S P85D 5YJSB8H43FF096936 | Tesla UK

Yes, there are now two locations, "London" and "Birmingham" that appear like they are in the US, but of course, really in the UK (it's a quirk in the data feed from Tesla).

Unfortunately, I'm in Honduras for two weeks, and the internet down here is dreadful. I'll look into fixing this as soon as I return. Thanks for understanding.