Range Loss Over Time, What Can Be Expected, Efficiency, How to Maintain Battery Health

[AlanSubie4Life]

Wish I'd discovered this article before deciding to purchase a Tesla. Posting for the benefit of others for whom range is important in their purchase decision process:

I'm sorry you missed this information. It's extensively covered here.

Tesla and Polestar achieve this primarily by using an additional 3 drive cycles its competitors do not to "earn" a higher EPA range.

Yes, that referenced article about the adjustment factor has been linked several times here, and prior to that, the exact method by which Tesla was increasing the rated range on their more recent 2021 models has been well described in this specific forum.

This extra 7% improvement on range over competitors who prefer to use the 0.7 factor is allowed by the EPA. Everyone has their own opinion on whether it's a good idea or not. I tend to think it leads to confusion and disappointment, and I think all EVs should be judged on exactly the same basis, and this latitude manufacturers have to use or not use the 0.7 factor disrupts that. But those are the rules.

You have to know exactly how the EPA numbers are calculated to be able to understand it and account for it (hence my "range optimism" category on my spreadsheet on the constants page). It trips up tons of Tesla owners.

Tesla vehicles are not even the most efficient at kWh/100 miles, per the article, raising questions about Tesla's potential to continue dominating the EV market as more established vehicle manufacturers with higher build quality and longer market experience move into the space.

This is a bit questionable. Note your 2021 AWD vehicle is not in the list!!! (Edmunds Tested: Electric Car Range and Consumption | Edmunds) The ones that definitely will be more efficient (Hyundai, Mini, Bolt, Kia) are not in the same class, arguably - and I would consider measuring their stopping distance as well (really low rolling resistance tires can be dangerously slippery).

Also note that the Model 3 Performance in the list is a 2018. When was this test conducted? Not really fair to compare a 2-3 year old test result with current test results. (The Model 3 efficiency has improved substantially over the last two years.)

I actually think that the 2021 Model 3 AWD, which would probably get 26kWh/100mi in this test, with a range of about 290 miles, is quite competitive when put alongside its competitors. (Mach-E apparently 33kWh/100mi but not sure if that is the best it can achieve or whether that was with sticky tires.)

Total range of 250 miles is what this vehicle can achieve when new, on relatively flat roads, in temperate conditions.

Yes, that's correct as an expectation, and well known here. I think with your vehicle with a little care you could probably do a little better than that, in ideal conditions. With a little traffic on the freeways, I'd expect you could get to nearly 300 miles (maybe 290?).

I take long road trips, and believed the vehicle would achieve about 300 miles between charges. It cannot.

No, it can't. Not reasonable to expect that.

I take long road trips, and believed the vehicle would achieve about 300 miles between charges. It cannot. This vehicle requires recharging every 200 miles or less on highways when driving at the speed limit of 75 MPH or a few MPH faster.

That's correct, on a road trip you should expect to charge every 150-200 miles. That's normal and it works really well - as long as you can access a charging station without waiting (much, much more important than the charging interval in my opinion).

Trips to farther destinations will likely require a recharge every 2-2.5 hours. Not what I thought I was buying, not how I like to drive, and not what Tesla markets aggressively in terms of range performance.

Yes, that's correct.

Why don't the EPA range numbers correlate with the EPA Consumption numbers? Is it due to degradation or the battery size used for calculation being different? Strange.

They drive the vehicles to 10 miles left. That makes these measurement subject to differences in how manufacturers provide margin at low energy. For Tesla (Model 3), this leaves about 7% (4.5% + 2.7%) of the battery energy remaining. No idea how it affects other manufacturers.

I think they should measure DC consumption somehow (also difficult because depends on how manufactures display it), and divide it into usable battery capacity provided by the manufacturer. Also subject to difficulties though. Their method is reasonable in terms of comparing real world ranges of vehicles directly - that comparison is valid, but it's not a great way to compare performance relative to the EPA rating.

In general, particularly on 2021 vehicles, I think the rated range numbers advertised will be very optimistic for Tesla. Mostly due to that 7% higher scaling factor. I don't think that's great for Tesla, but they're in it to sell vehicles. But I also think that fundamentally their efficiency is still extremely competitive vs. other vehicles - and for road trips, they have the Supercharger network, which is still likely superior to the other networks (try taking a road trip to the Dakotas in another type of EV), though the other networks are fortunately improving.

They sent an email and PDF showing motor usage represented 96% of total electricity usage.

Doesn't really answer the question of whether your vehicle has an undiagnosed rolling resistance problem, or whether your conditions were particularly difficult. But at least significant heat use can be ruled out (which isn't too surprising - we'd expect at most 1000W average or 16Wh/mi at 60mph average.)

Learning how road trips work in a Tesla is an experience. But if you were expecting to stop just once every 300 miles, you're definitely going to be disappointed. Expect 150-200 miles, with very quick stops (if there is no waiting) if you're using 250kW Superchargers and optimizing your charging strategy (10-70%).

I hope you give the vehicle a chance. It's really not that big a deal as long as charging stations are available without waiting when you get there. I certainly understand the disappointment at not getting what you expected, and I would say anyone getting a Tesla needs to do careful research to understand what they're actually getting.

 

[AAKEE]

I found myself using more or less exactly the EPA-range consumption yesterday.
My ’21 refresh performance have (the known) issues with the heat pump, and because of 1000km one way to the SC I havent been there yet. The heatpump works as is should sometimes and simetimes not, increasing the consumtion quite much in cold wx.

But yesterday it worked as it should, both ways to/from work. Its a 50km drive, 90km/h (and 110km/h for a short part), of which about 40 is highway, and 10 is outer parts of city. Theres about 20km of the 90km/h drive with road construction with 70km/h but mostly the speed will be around 80km/h. On total not a fast drive but I followed the traffic and even did overtake a couple of vehicles.
I have studded 235/40-19, probably not the most lightly rolling tyres. Morning temps was about -9C and -5C for the drive home.

After driving a while, I found the energy app reported range to be on pair with the battery range meter(usually set to % but this day I had it set to show range). A bit surprised that this could happen during winter. I didnt drive that slow. So, if needed I acould get the advertized (EPA) range, even in winter conditions.
(For Jk_tx: In EU we have the WLTP which gives 12% higher range values compared to EPA. These will be very hard to achieve, but I guess anyone knows WLTP values only are for comparison).

 

[KenC]

I found myself using more or less exactly the EPA-range consumption yesterday.
My ’21 refresh performance have (the known) issues with the heat pump, and because of 1000km one way to the SC I havent been there yet. The heatpump works as is should sometimes and simetimes not, increasing the consumtion quite much in cold wx.

But yesterday it worked as it should, both ways to/from work. Its a 50km drive, 90km/h (and 110km/h for a short part), of which about 40 is highway, and 10 is outer parts of city. Theres about 20km of the 90km/h drive with road construction with 70km/h but mostly the speed will be around 80km/h. On total not a fast drive but I followed the traffic and even did overtake a couple of vehicles.
I have studded 235/40-19, probably not the most lightly rolling tyres. Morning temps was about -9C and -5C for the drive home.

After driving a while, I found the energy app reported range to be on pair with the battery range meter(usually set to % but this day I had it set to show range). A bit surprised that this could happen during winter. I didnt drive that slow. So, if needed I acould get the advertized (EPA) range, even in winter conditions.
(For Jk_tx: In EU we have the WLTP which gives 12% higher range values compared to EPA. These will be very hard to achieve, but I guess anyone knows WLTP values only are for comparison).
View attachment 640061

Worse is the NEDC which is what China uses, and is about 43% higher than what the EPA uses.

 

[AlanSubie4Life]

After driving a while, I found the energy app reported range to be on pair with the battery range meter(usually set to % but this day I had it set to show range). A bit surprised that this could happen during winter. I didnt drive that slow. So, if needed I acould get the advertized (EPA) range, even in winter conditions.

Subtle but extremely confusing caveat: At 158Wh/km with 422km projected range, with battery gauge showing 422km, you'd actually only be able to travel 422rkm (displayed) *151Wh/rkm (displayed) /158Wh/km = 403km. And then you'd be at 0km. But then you'd still have 4.5% of your battery left, and yes if you're lucky you could drive ~422km. But most people like to stop at 0km or before! And at 158Wh/km you can only travel 403km if you start with 422rkm displayed and you stop at 0rkm.

Anyway, this is because each displayed km on the battery gauge only contains 95.5% of the rated constant energy of 158Wh/rkm. (Easy to verify, though there's also an additional ~1% loss factor I'm not going to discuss here!)

But to your point, yes: Certainly if you're willing to use your entire buffer, at 158Wh/km, you can do ~510km range that your Performance vehicle has. And your result is remarkably good for the conditions.

This does not contradict all the battery capacity calculations that have been covered elsewhere using this energy screen.

 

[jk_tx]

I'm sorry you missed this information. It's extensively covered here.



Yes, that referenced article about the adjustment factor has been linked several times here, and prior to that, the exact method by which Tesla was increasing the rated range on their more recent 2021 models has been well described in this specific forum.

This extra 7% improvement on range over competitors who prefer to use the 0.7 factor is allowed by the EPA. Everyone has their own opinion on whether it's a good idea or not. I tend to think it leads to confusion and disappointment, and I think all EVs should be judged on exactly the same basis, and this latitude manufacturers have to use or not use the 0.7 factor disrupts that. But those are the rules.

You have to know exactly how the EPA numbers are calculated to be able to understand it and account for it (hence my "range optimism" category on my spreadsheet on the constants page). It trips up tons of Tesla owners.



This is a bit questionable. Note your 2021 AWD vehicle is not in the list!!! (Edmunds Tested: Electric Car Range and Consumption | Edmunds) The ones that definitely will be more efficient (Hyundai, Mini, Bolt, Kia) are not in the same class, arguably - and I would consider measuring their stopping distance as well (really low rolling resistance tires can be dangerously slippery).

Also note that the Model 3 Performance in the list is a 2018. When was this test conducted? Not really fair to compare a 2-3 year old test result with current test results. (The Model 3 efficiency has improved substantially over the last two years.)

I actually think that the 2021 Model 3 AWD, which would probably get 26kWh/100mi in this test, with a range of about 290 miles, is quite competitive when put alongside its competitors. (Mach-E apparently 33kWh/100mi but not sure if that is the best it can achieve or whether that was with sticky tires.)



Yes, that's correct as an expectation, and well known here. I think with your vehicle with a little care you could probably do a little better than that, in ideal conditions. With a little traffic on the freeways, I'd expect you could get to nearly 300 miles (maybe 290?).



No, it can't. Not reasonable to expect that.



That's correct, on a road trip you should expect to charge every 150-200 miles. That's normal and it works really well - as long as you can access a charging station without waiting (much, much more important than the charging interval in my opinion).



Yes, that's correct.



They drive the vehicles to 10 miles left. That makes these measurement subject to differences in how manufacturers provide margin at low energy. For Tesla (Model 3), this leaves about 7% (4.5% + 2.7%) of the battery energy remaining. No idea how it affects other manufacturers.

I think they should measure DC consumption somehow (also difficult because depends on how manufactures display it), and divide it into usable battery capacity provided by the manufacturer. Also subject to difficulties though. Their method is reasonable in terms of comparing real world ranges of vehicles directly - that comparison is valid, but it's not a great way to compare performance relative to the EPA rating.

In general, particularly on 2021 vehicles, I think the rated range numbers advertised will be very optimistic for Tesla. Mostly due to that 7% higher scaling factor. I don't think that's great for Tesla, but they're in it to sell vehicles. But I also think that fundamentally their efficiency is still extremely competitive vs. other vehicles - and for road trips, they have the Supercharger network, which is still likely superior to the other networks (try taking a road trip to the Dakotas in another type of EV), though the other networks are fortunately improving.



Doesn't really answer the question of whether your vehicle has an undiagnosed rolling resistance problem, or whether your conditions were particularly difficult. But at least significant heat use can be ruled out (which isn't too surprising - we'd expect at most 1000W average or 16Wh/mi at 60mph average.)

Learning how road trips work in a Tesla is an experience. But if you were expecting to stop just once every 300 miles, you're definitely going to be disappointed. Expect 150-200 miles, with very quick stops (if there is no waiting) if you're using 250kW Superchargers and optimizing your charging strategy (10-70%).

I hope you give the vehicle a chance. It's really not that big a deal as long as charging stations are available without waiting when you get there. I certainly understand the disappointment at not getting what you expected, and I would say anyone getting a Tesla needs to do careful research to understand what they're actually getting.

Asked Tesla Service about an undiagnosed rolling resistance problem, and they wrote, "A rolling resistance issue is unlikely. The potential suspects would be alignment (I’ve performed an automated alignment analysis that reviews average steering angles to determine if steering is compensating for poor alignment) and did not find any issues. Low tire pressure would also be a contributing factor, but as long as you aren’t seeing any constant low tire pressure alerts, there is no indication that this is an issue. Another potential factor could be brake drag, but there would be other factors present such as excessive brake pad wear, abnormal brake noise, very hot wheels and odor of burning brake pads, etc. There likely is no rolling resistance issue as the vehicle speeds discussed above, which are well above EPA testing patterns, fully account for the increase in energy consumption."

I also asked about time at different speeds, and of the 189.4 miles they found 22 miles at 79 MPH (11%), 99 miles at 74-74 MPH (52%), and the remainder was mixed speeds (likely under 75 MPH as 30 miles each way have limits of 65-70 MPH). Per the service tech, "resulting in 63% of the drive spent at 74mph or higher. This alone, without any other external influences, would result in energy consumption higher than the EPA test cycles."

At this stage, I have no choice but to give the vehicle a chance. Bottom line is if my vehicle is driven at posted highway speed limits or 5 MPH above (which is my habit to not receive a ticket but arrive at my destination as soon as possible), this vehicle will achieve about 70% of advertised range when new (250 mile range but will need to be charged well before reaching that distance), in temperate weather, on relatively flat terrain.

Again, AlanSubie4Life, you are a saint of a resource on this forum: kind and informative. I'm frustrated. and you've been patient and very helpful. My sons' mother and I each bought Teslas (2021 M3 LR, 2021 MY) after research. Had we known our vehicles would only have 70% of advertised range when new instead of 90% (per Teslike and some other resources) at normal hwy speeds, we would not each have bought a Tesla. Likely we would have waited about 2 years when the battery tech allows for greater energy storage and more efficient usage. Thank you again for your kindness, AlanSubie4Life.

 

[AlanSubie4Life]

I also asked about time at different speeds, and of the 189.4 miles they found 22 miles at 79 MPH (11%), 99 miles at 74-74 MPH (52%), and the remainder was mixed speeds (likely under 75 MPH as 30 miles each way have limits of 65-70 MPH). Per the service tech, "resulting in 63% of the drive spent at 74mph or higher. This alone, without any other external influences, would result in energy consumption higher than the EPA test cycles."

That is a very nice and exhaustive analysis. With the model I provided earlier, I still believe those speeds do not account for a 280Wh/mi result (exclude climate control) at 75mph in this vehicle. Again, for parity with the (scaled) EPA test cycles, you need to get ~220Wh/mi, and if you want mile-for-rated-mile rolloff on the battery gauge, you need to get ~208Wh/mi on the trip meter. I'd expect more like 250-260Wh/mi at those speeds.

HOWEVER, at those speeds you are very sensitive to winds and as I said I looked up the winds and there was a generally south wind (and quite strong) during the time of your outgoing drive to the southeast, at least. That is what the observational data says (you can look it up; that is more reliable than observations in the car). Even a 10mph headwind (and it looks like it was more like a 20-30mph crosswind) WOULD explain the discrepancy. That’s like driving 85mph and at that speed 280Wh/mi is pretty reasonable.

Winds matter, a lot.

I think you should do another trial run (or better yet a real drive with a purpose) for yourself at some point in more optimal conditions, to make yourself feel better. Sure, sometimes you'll have headwinds. But sometimes you'll have tailwinds (but the winds hurt more than they help)!

Bottom line is if my vehicle is driven at posted highway speed limits or 5 MPH above (which is my habit to not receive a ticket but arrive at my destination as soon as possible), this vehicle will achieve about 70% of advertised range when new (250 mile range but will need to be charged well before reaching that distance), in temperate weather, on relatively flat terrain.

That’s about right for the expectation. However, I do not know why you say it needs to be charged well in advance. As long as weather conditions are not hazardous to life, it is perfectly fine to drive your car down to 5%, and the car has an excellent and accurate trip planner to allow this. You'll get just over 250 miles from 100% to 5% if you do better than 275Wh/mi (which you will, assuming good conditions), with an un-degraded battery. That's just math, there's no ambiguity or variability there. And: If you really have to push it from there, you've still got that 5%, and you've got another 4.5% (that you shouldn't attempt to use except in extreme emergencies, and you should drive very very slowly when using) below that!

I routinely drive above the speed limit by about 5mph (to keep up with the flow of traffic), and I've never had any issues with the distance between charges. I guess it's all about expectations. Would it be nice if I had the flexibility to go a little further? Sure. The only time it is really a problem is when traveling to remote locations without Superchargers. I had to be careful traveling to the North Rim of the Grand Canyon, for example. Charge up extra high, take some care to draft when convenient (especially when there is a headwind), etc.

I honestly think that as long as you're willing to optimize the charging strategy (5/10% - 70%) and you have 250kW chargers available (these are becoming more common and hopefully Tesla will make a point of replacing the old V2s), you'll find it's not that big a deal. Just enough time to use the bathroom and grab & eat a snack, maybe clean the bugs off your windshield, take a quick glance at your email/texts & respond, and then it's time to go. You can easily add ~220 rated miles in 20 minutes.

 

[MP3Mike]

Again, AlanSubie4Life, you are a saint of a resource on this forum: kind and informative. I'm frustrated. and you've been patient and very helpful. My sons' mother and I each bought Teslas (2021 M3 LR, 2021 MY) after research. Had we known our vehicles would only have 70% of advertised range when new instead of 90% (per Teslike and some other resources) at normal hwy speeds, we would not each have bought a Tesla.

Really? Teslike doesn't say you will get 90% at highway speeds: Teslike.com

It shows 81%, 254 miles, at 80 MPH with a 2020 Model 3 LR with 18" wheels with the aero hubcaps installed. (It doesn't seem that he has updated his resources for newer models.)

 

[AlanSubie4Life]

It shows 81%, 254 miles, at 80 MPH with a 2020 Model 3 LR with 18" wheels with the aero hubcaps installed. (It doesn't seem that he has updated his resources for newer models.)

Yes, since @jk_tx referred to this table, it seems to me he could have "reasonably" concluded that the range at 80mph would be 254mi*353/310 = 289 miles. I think that's likely just possible in ideal conditions (265Wh/mi would do it, which I'm not sure you'd quite get at 80mph, but I think it could be close if there is some light traffic to follow), but it neglects to account for having to drive all the way to 0%, and probably into the buffer, to get that range at that speed (that's what the table requires, which I think people don't realize).

Putting that aside, as you say, the table is out-of-date, and also the scaling doesn't work this way (~23 miles, very roughly (353-353/1.07) of the increase in range between 353 & 310 is likely not applicable, if you're not using the heat pump, in very hot (less likely to be of help, but interestingly the EPA data says it did) or very cold conditions). This is confusing to people, of course.

On the other hand, when using 250kW Superchargers and accounting for the lower Wh/rmi on the new vehicles, the maximum rate at which you add miles is MUCH higher now. Pretty sure you can get (roughly) about 250 rated miles in 30 minutes, as I've gotten 200 miles in ~20 minutes, and those were more energetic miles. Two distinct factors: 1) Different charging constants, so more miles per minute for a given wattage (same energy per minute of course, because that is the definition of power!) 2) Higher wattage Superchargers

 

[diamond.g]

Really? Teslike doesn't say you will get 90% at highway speeds: Teslike.com

It shows 81%, 254 miles, at 80 MPH with a 2020 Model 3 LR with 18" wheels with the aero hubcaps installed. (It doesn't seem that he has updated his resources for newer models.)

Yeah I would not go off this table, it hasn’t been updated since 2019 and the numbers are different now. No clue when Troy plans on updating it (he used to post here but was banned for reasons).

 

[Zoomit]

it hasn’t been updated since 2019 and the numbers are different now.

Not only that, but the numbers were incorrect in 2019 as well. Troy thought the EPA testing represented results without the Aero covers, which was not true.

It's easy to mistake precision for accuracy.

 

[AlanSubie4Life]

Not only that, but the numbers were incorrect in 2019 as well. Troy thought the EPA testing represented results without the Aero covers, which was not true.

It's easy to mistake precision for accuracy.

Yep, checking my table, story checks out! I figured at least the basic information would be correct. Guess not!

I kind of take back what I said earlier about the 7% factor. It does look like he's basing these tables on that HWFET test and scaling from there (not sure his model of that scaling though). That projected result wouldn't be affected by that scaling factor as long as he didn't change it. But in any case he's starting with the wrong data and it's not updated for 2021. The 2021 got 447 miles in this test compared to his table entry of 440 for 2020 (? He has 310 so I guess it's actually 2018 but he also has 2019 in there so who knows). But even that's not good enough, because he should really be normalizing for the battery capacity of that particular test article (in other words looking at efficiency), since there's ~1.5% variability from test article to test article in battery capacity - and that makes the resulting distance numbers misleading by 1-2% - and the error can cut either way. In the case of the 2021, it actually cut in its favor in terms of efficiency, since its battery was cut off 1.3kWh earlier, for the test vehicle.

All very easy for prospective buyers to follow. Haha.

 

[scottf200]

Re: Son's '18 TM3 - used to drive 110 miles per day but for several months has been able to work from home. The current battery is still a fair bit below average.

Earlier post: TeslaFi - Battery Degradation Reports (upload your data)

 

[PhantomX]

Apologize if this has been covered by previous discussions, but does charging the battery to 100% frequently voids the battery warranty? My Model 3 MR has lost 16% (222 miles rated range @100%) of its battery capacity, so I may need to charge closer to 100% more often than I had ever planned to for some of the long commutes.

 

[ggies07]

Apologize if this has been covered by previous discussions, but does charging the battery to 100% frequently voids the battery warranty? My Model 3 MR has lost 16% (222 miles rated range @100%) of its battery capacity, so I may need to charge closer to 100% more often than I had ever planned to for some of the long commutes.

Not sure about that, but wanted to add my data point that my '18 MR3 is 244 at 100%.

 

[PhantomX]

Not sure about that, but wanted to add my data point that my '18 MR3 is 244 at 100%.

Wow, I would love to get 244 @100%. It appears I have regained 1% of battery with 2021.4.11 update, from 222 miles to 225 miles @100%. At least now my 90% is above 200 miles. It was disappointing previously when I would get in the car and see 199 miles at 90%.

 

[ggies07]

Wow, I would love to get 244 @100%. It appears I have regained 1% of battery with 2021.4.11 update, from 222 miles to 225 miles @100%. At least now my 90% is above 200 miles. It was disappointing previously when I would get in the car and see 199 miles at 90%.

And here I thought 244 was sucky...really wish it was at least 250....

 

[PhantomX]

And here I thought 244 was sucky...really wish it was at least 250....

Agreed. I am still hoping it's a calculation error in BMS rather than actual capacity loss. Though I have pass the half way point on battery loss % for warranty claim, perhaps I should cheer for the degradation to continue at a faster speed so I can get a new battery

 

[AlanSubie4Life]

I am still hoping it's a calculation error in BMS rather than actual capacity loss.

It's very likely real; persistent estimate errors are rare (though they do occur of course). Easy enough to determine on a road trip. Of course, if you never have the opportunity to determine whether it's real, then it also doesn't matter at all to you; it has zero impact (except for resale value).

Lithium ion batteries lose capacity. It's normal and expected for all EVs, and will vary from vehicle to vehicle.

 

[PhantomX]

It's very likely real; persistent estimate errors are rare (though they do occur of course). Easy enough to determine on a road trip. Of course, if you never have the opportunity to determine whether it's real, then it also doesn't matter at all to you; it has zero impact (except for resale value).

Lithium ion batteries lose capacity. It's normal and expected for all EVs, and will vary from vehicle to vehicle.

Thanks. I have made a number of long road trips that I am convinced that I don't have the range the MR supposed to have. But those drives also weren't meant as a true range test, since I was going at the same speed as everyone else and I had time constraints. I do know the max I was able to comfortably extract out of the car was 180 miles, but again, I wasn't going for efficiency.

Question about the lithium ion battery losses. How does other manufacturers handle this? Reason I ask is the 2 BMW i3's I had before never showed loss in range over years I had them. And my friend just turned in his i3 after 2.5 years, and he also didn't observe any loss in range. Is BMW holding a large reserve at start and then slowly letting the car tap those reserve in order to maintain the available range?

 

[AlanSubie4Life]

I do know the max I was able to comfortably extract out of the car was 180 miles, but again, I wasn't going for efficiency.

I would focus more on energy that you are able to extract rather than distance. At this point, it is very likely reduced from what one would expect from an MR when new.

How does other manufacturers handle this? Reason I ask is the 2 BMW i3's I had before never showed loss in range over years I had them. And my friend just turned in his i3 after 2.5 years, and he also didn't observe any loss in range. Is BMW holding a large reserve at start and then slowly letting the car tap those reserve in order to maintain the available range?

I'm not sure how other manufacturers handle it. Certainly some handle it in the way you describe where there is a hidden buffer which gets used up by the fairly quick initial capacity loss, but I don't know anything about specific other manufacturers. Also, results will vary from vehicle to vehicle.

For my Spark EV, it's four years old and it has lost about 20% of its capacity as far as I can tell (you can monitor how much energy has been used since the last charge). Also, it typically displays about 64 miles on its GOM now rather than the ~80 it displayed when new. Of course, it also sits in the garage charged to 100% basically all the time (there is no feature to limit the max charge level).