FSD button is here!

[101dals]

Looks like Tesla sold almost 250,000 cars last quarter. I've got a long wait until my Brooklyn driver dodging score of 93 gets called.....

 

[Yelobird]

That's not the point I was making. All owners paid the price of admission when they entered and exit a Tesla dealership. They are entitled to anything and everything this car has to offer. Unlike, Mercedes, Cadillac, Lincoln, etc., you are NOT buying a finished product as its been evolving for how many years? Promises made to buyers should be kept and NOT held out there like a carrot on a stick and watching the stick grow longer by the YEARS. There is nothing done or safe about these cars or any other. People will drive their cars as they will and if you have a Mercedes and put on cruise control you're responsible for using it. Same goes for a Tesla; you are the responsible party for how it operates and if you crash your car you crashed YOUR car not 1000's of others. We are NOT a collective here, we are ALL individuals with our check books out paying our OWN bills. You really think that a finished version of this program hasn't been out and there and has NOT been being tested for the last few years? Who are those beta testers? Engineers at Tesla? Elon? The keeper of all EV knowledge and know how? We will never be done and never be safe so get off one of your Tesla high horses.

And like your MKz you’ll be the first in line for your “entitled“ hand out when you crash. With some simple research pre-purchase you would know this is a car/company/technology that is continuously evolving. Want what your entitled to, buy a Honda. Tesla is so lucky to have you as a customer… Never seen a Tesla dealership as you mention but you should probably just return it sooner then later as you will certainly never be satisfied. Said from my high horse of observation.

 

[Carbonfiber]

So what happens next after having this safety score for a week? I feel this is just a way for Tesla to buy more time on this FSD fiasco

 

[Geech72]

Yesterday I went on trip of 130 miles which was 99% interstate only. I used autopilot for 100% of my highway miles, yet I was dinged for following too closely on 29.7% of the miles traveled for a score of 89. I guess that means that Tesla believes that their own autopilot is unsafe? I’m going to stop worrying about the beta and my safety score, and just drive the car as I want to drive it.

 

[spsheridan]

Yesterday I went on trip of 130 miles which was 99% interstate only. I used autopilot for 100% of my highway miles, yet I was dinged for following too closely on 29.7% of the miles traveled for a score of 89. I guess that means that Tesla believes that their own autopilot is unsafe? I’m going to stop worrying about the beta and my safety score, and just drive the car as I want to drive it.

What setting did you use for following distance/time?

 

[mackgriggs]

Yesterday I went on trip of 130 miles which was 99% interstate only. I used autopilot for 100% of my highway miles, yet I was dinged for following too closely on 29.7% of the miles traveled for a score of 89. I guess that means that Tesla believes that their own autopilot is unsafe? I’m going to stop worrying about the beta and my safety score, and just drive the car as I want to drive it.

So....anything that happens on autopilot doesn't count towards your safety score. Like, if autopilot does a hard brake, it doesn't count against your score. However...the inverse is true. Anything good that happens on autopilot doesn't count either.

Lets take your case. You drove 130 miles, mostly on autopilot. For the very brief time that you were not on AP, lets say...1 mile, you happened to be driving behind a car, or maybe a car cut in front of you. For the time that you were within 3 follow seconds of that car, 29.7% of that time was closer than 1 follow second.

Normally this isn't a problem. The issue is.....for that day you had NO other car following interactions while not on AP, so it gave you a score based on ONE interaction, which unfortunately was a bad one. So now you have a bad safety score for that day, because Tesla is only scoring you based on a very brief moment when you were not on AP. This is further compounded that now that daily score is weighted by 130 miles, which does count the AP miles.

Basically...driving on Autopilot is a double edged sword. If you get a Foward collision warning in the morning, then yes, driving on autopilot for the rest of the day would be a good thing, because AP miles will dilute your FCW warning. However, if you get a hard braking incident in the morning, you better drive manually for the rest of the day so you can rack up plenty of "good braking" incidents during the day, to dilute your bad brake incident.

Its like this. If you drive 500 miles on AP, but only have 1 turn, you had better make ABSOLUTELY sure that your turn is a good turn because your whole trip score is going to be based on that 1 turn. If you do good...GREAT! You have 500 miles of a 100 score. If its not...you really just hurt your daily score, which is now weighted by 500 miles.

I personally drive on autopilot a lot, just because its easier for me to drive like that, but I have to make sure that while NOT on autopilot, that my driving is perfect. I have 5 days where I have perfect scores, just because there are very few graded interactions those days, but 50+ miles. Plenty of people take the opposite approach though, and not use Autopilot very often. Yes, they get dinged more, but they dilute those dings with "good interactions" that are being recorded.

 

[N1L CO2]

Folks, IMHO, this is a total BS game played by Elon to buy time! How dare they "judge" and rate peoples' driving accordingly to some dumb algorithm?! There was no mention of any stipulations when I bought FSD and now I have to deal with changing my driving habits to fit into some scoring bracket??? WTF!! I have been driving accident & ticket free for the last 15+ years. I soooooo regret purchasing FSD. I could have put that money into other investments and created some + cashflow. Ugh, what a waste!!! As much as I like the Y, I am so turned off by this "vaporware" that's supposed to have been here years ago.

 

[HeyEph]

Make sure you don't go through a car wash! I was dinged with 102 Forward Collision Warnings, and I was even in Car Wash Mode!

 

[Silicon Desert]

Yesterday I went on trip of 130 miles which was 99% interstate only. I used autopilot for 100% of my highway miles, yet I was dinged for following too closely on 29.7% of the miles traveled for a score of 89. I guess that means that Tesla believes that their own autopilot is unsafe? I’m going to stop worrying about the beta and my safety score, and just drive the car as I want to drive it.

Set the following distance to 6 or 7.

 

[Silicon Desert]

Folks, IMHO, this is a total BS game played by Elon to buy time!

Are you also one of those that think the moon landing was staged back in 1969 No, it is not a BS game, but since it is not a perfect system, I can see why people have various theories.

 

[Geech72]

So....anything that happens on autopilot doesn't count towards your safety score. Like, if autopilot does a hard brake, it doesn't count against your score. However...the inverse is true. Anything good that happens on autopilot doesn't count either.

So....anything that happens on autopilot doesn't count towards your safety score.

Do you have any actual documentation that states this or is this an anecdotal observation? I don’t believe this statement to be true. When I look at the details of each trip the app shows that the trip consists of 130 miles, which would have to include all of the data related to the miles I drove while on autopilot. Also this one trip lowered my total safety score ( compiled over a few hundred miles) by 2 points. If only the few miles traveled while not on autopilot was the only data captured during this trip, then it would make up a great deal less than 1% of the total miles traveled and would have had an immaterial impact on my overall score. This was the only trip in my total data set in which I used autopilot, so if you theory is correct all of the miles prior would have been considered.

 

[UT4EVER]

Do you have any actual documentation that states this or is this an anecdotal observation? I don’t believe this statement to be true. When I look at the details of each trip the app shows that the trip consists of 130 miles, which would have to include all of the data related to the miles I drove while on autopilot. Also this one trip lowered my total safety score ( compiled over a few hundred miles) by 2 points. If only the few miles traveled while not on autopilot was the only data captured during this trip, then it would make up a great deal less than 1% of the total miles traveled and would have had an immaterial impact on my overall score. This was the only trip in my total data set in which I used autopilot, so if you theory is correct all of the miles prior would have been considered.

He’s telling you the truth. It was mentioned in the original release notes and FAQ. It clearly stated that any event while on autopilot would not be included in the score. And I have experienced this first hand…. I have had several hard braking and forward collision incidents while on autopilot yet my score is still at 100 after over 800 miles.

You might want to do your own research before you call someone else a liar.

 

[Geech72]

He’s telling you the truth. It was mentioned in the original release notes and FAQ. It clearly stated that any event while on autopilot would not be included in the score. And I have experienced this first hand…. I have had several hard braking incidents while on autopilot yet my score is still at 100 after over 800 miles.

You might want to do your own research before you call someone else a liar.

It was not my intention to call anyone a liar, nor did I actually do that. You might want to find a way to relax. Someone I don’t know posted something as fact on an internet message board. What I did was inquire about the veracity of the information; did they get it first hand from the manufacturer or just read it on another post. At any rate I’m not necessarily doubting the poster but I am doubting that it’s actually working that way in practice. Based on the number of miles I’ve driven since the start of the measurement it is mathematically impossible for my score to drop 2 points without factoring in all of the miles that I’ve driven using autopilot.

 

[mackgriggs]

Do you have any actual documentation that states this or is this an anecdotal observation? I don’t believe this statement to be true. When I look at the details of each trip the app shows that the trip consists of 130 miles, which would have to include all of the data related to the miles I drove while on autopilot. Also this one trip lowered my total safety score ( compiled over a few hundred miles) by 2 points. If only the few miles traveled while not on autopilot was the only data captured during this trip, then it would make up a great deal less than 1% of the total miles traveled and would have had an immaterial impact on my overall score. This was the only trip in my total data set in which I used autopilot, so if you theory is correct all of the miles prior would have been considered.

Sure.... Safety Score Beta

-------------
Tesla captures driving over all trips, where a trip consists of any driving (including reversing and Autopilot usage) occurring between the time the vehicle is powered on and able to be driven to the time the vehicle is powered off. Any driving and accrued mileage occurring while in service mode is excluded. Trips less than 0.1 mile will be excluded from scoring calculations. Driving on Autopilot will not be included in Safety Score calculations, but the miles driven while on Autopilot are included in the total.
-------------

So, Autopilot miles are used in two ways in the score:

1. FCW is per 1000 miles. Miles running in autopilot DO count towards that 1000 mile calculation.

2. Miles running in Autopilot count towards your daily miles driven, and is used in the daily miles weighted average.

And yes....your daily score from today can drop your score by 2 points, since its just a daily average.

For example....lets say you drive 5 days, at 100 miles each.

Day 1: Score: 100 miles: 100
Day 2: Score: 100 miles: 100
Day 3: Score: 100 miles: 100
Day 4: Score: 100 miles: 100
Day 5: Score: 90 miles: 100

Your new safety score: 98

New example:
Day 1: Score: 100 miles: 50
Day 2: Score: 100 miles: 50
Day 3: Score: 100 miles: 50
Day 4: Score: 100 miles: 50
Day 5: Score: 90 miles: 100

Your new safety score: 96.67 (rounded to 97).

 

[UT4EVER]

It was not my intention to call anyone a liar, nor did I actually do that. You might want to find a way to relax. Someone I don’t know posted something as fact on an internet message board. What I did was inquire about the veracity of the information; did they get it first hand from the manufacturer or just read it on another post. At any rate I’m not necessarily doubting the poster but I am doubting that it’s actually working that way in practice. Based on the number of miles I’ve driven since the start of the measurement it is mathematically impossible for my score to drop 2 points without factoring in all of the miles that I’ve driven using autopilot.

You still aren’t getting it. What he is telling you (truthfully) is that you are not being deducted points for incidents (unsafe following, etc) while on autopilot. You are only deducted points for incidents OFF of autopilot. The mileage driven would only be a factor in the equation…. Driving 1000 miles while 900 on autopilot will not be the same as driving 1000 miles with 100 on autopilot. It’s not that complicated to comprehend….

 

[Geech72]

Sure.... Safety Score Beta

-------------
Tesla captures driving over all trips, where a trip consists of any driving (including reversing and Autopilot usage) occurring between the time the vehicle is powered on and able to be driven to the time the vehicle is powered off. Any driving and accrued mileage occurring while in service mode is excluded. Trips less than 0.1 mile will be excluded from scoring calculations. Driving on Autopilot will not be included in Safety Score calculations, but the miles driven while on Autopilot are included in the total.
-------------

So, Autopilot miles are used in two ways in the score:

1. FCW is per 1000 miles. Miles running in autopilot DO count towards that 1000 mile calculation.

2. Miles running in Autopilot count towards your daily miles driven, and is used in the daily miles weighted average.

And yes....your daily score from today can drop your score by 2 points, since its just a daily average.

For example....lets say you drive 5 days, at 100 miles each.

Day 1: Score: 100 miles: 100
Day 2: Score: 100 miles: 100
Day 3: Score: 100 miles: 100
Day 4: Score: 100 miles: 100
Day 5: Score: 90 miles: 100

Your new safety score: 98

New example:
Day 1: Score: 100 miles: 50
Day 2: Score: 100 miles: 50
Day 3: Score: 100 miles: 50
Day 4: Score: 100 miles: 50
Day 5: Score: 90 miles: 100

Your new safety score: 96.67 (rounded to 97).

Thanks for posting that was helpful. the Details in the FAQ are not the clearest I’ve ever read:

“Your daily Safety Score is not impacted by the number of miles or hours you drive. We combine your daily Safety Scores (up to 30 days) into a mileage-weighted average to calculate the aggregated Safety Score, which is displayed on the main ‘Safety Score’ screen of the Tesla app.”

So it’s your interpretation that they’re calculating my average daily score using only non-autopilot miles, but using total miles driven that day to weight the daily score into the cumulative score? For example if I drive 130 miles total but 120 of em are on autopilot they take the score for just the 10 miles that are manual driving but weight it the full 130 miles when combining it with the other daily averages? That’s certainly an interesting approach. They say later in the FAQ that they include miles driven while on autopilot in the total displayed but they don’t say clearly if that’s the total they use to aggregate the daily results.

 

[everydaychris]

Yesterday I went on trip of 130 miles which was 99% interstate only. I used autopilot for 100% of my highway miles, yet I was dinged for following too closely on 29.7% of the miles traveled for a score of 89. I guess that means that Tesla believes that their own autopilot is unsafe? I’m going to stop worrying about the beta and my safety score, and just drive the car as I want to drive it.

Yeah
Same with me. Thought autopilot was golden so I drove in traffic with autopilot with 7 car distance. However with people cutting you off etc I came home and had 50% ding for traveling too close... WHY!?
I'm still trying to get a high score but I'm not going to stress about it as much

 

[mackgriggs]

For example if I drive 130 miles total but 120 of em are on autopilot they take the score for just the 10 miles that are manual driving but weight it the full 130 miles when combining it with the other daily averages?

Yes, only events that happen during the 10 miles off autopilot are used to calculate your safety score, except for forward collision warnings, which includes autopilot driven miles in its "fcw events per 1000 miles" calculation. And then yes, they weight the full 130 miles driven that day into the "daily miles weighted average" total. Its specifically says so in the FAQ:

------------
Miles driven while Autopilot is engaged is used to determine the Forward Collision Warnings per 1,000 Miles. Miles driven on Autopilot are also considered when calculating the mileage-weighted Safety Score.
------------

Yeah
Same with me. Thought autopilot was golden so I drove in traffic with autopilot with 7 car distance. However with people cutting you off etc I came home and had 50% ding for traveling too close... WHY!?
I'm still trying to get a high score but I'm not going to stress about it as much

You got the 50% ding for something that happened outside of autopilot. And....with 50%....I would assume it was for an event that didn't happen long. Similar to geech72, you probably had 1 "out of autopilot" following event (you were following a car within 3 seconds), and half of that time was within 1 second. For example, you just happened to be out of autopilot, and a car cuts in front of you for 2 seconds, but 1 second of that time was in the "danger" zone of being closer than 1 following second, so your WHOLE score is 50% "bad" for that trip.

 

[everydaychris]

Yes, only events that happen during the 10 miles off autopilot are used to calculate your safety score, except for forward collision warnings, which includes autopilot driven miles in its "fcw events per 1000 miles" calculation. And then yes, they weight the full 130 miles driven that day into the "daily miles weighted average" total. Its specifically says so in the FAQ:

------------
Miles driven while Autopilot is engaged is used to determine the Forward Collision Warnings per 1,000 Miles. Miles driven on Autopilot are also considered when calculating the mileage-weighted Safety Score.
------------



You got the 50% ding for something that happened outside of autopilot. And....with 50%....I would assume it was for an event that didn't happen long. Similar to geech72, you probably had 1 "out of autopilot" following event (you were following a car within 3 seconds), and half of that time was within 1 second. For example, you just happened to be out of autopilot, and a car cuts in front of you for 2 seconds, but 1 second of that time was in the "danger" zone of being closer than 1 following second, so your WHOLE score is 50% "bad" for that trip.

OHH. That's why! LOL. I was upset because it said autopilot doesn't count and yet I got screwed. Thanks for the clarification!

 

[Armando49]

I now have 6 to 7 days of Score days. Current Score of 89..
When can I expect to be allowed in, or FSD unlocked ?

has any one actually got FSD now after 7 days ?