clydeiii
Member
We need more screenshots to get to the bottom of this. What smartphone OS are you on and what version of the Tesla app? I’m iOS 16.3.1 and Tesla app 4.20.69-1691.
-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
FSD beta lowers safety score
- Thread starter nickbhavsar
- Start date
legendsk
Member
In my case, it was just there when I woke up one morning.
Android: 4.20.69-1691-c9b4........
SS V2.0
Tesla HW3 2022.45.15
Last edited:
What version of the safety score are you in? You can check the version in the app. They have changed various things in 1.2 and now 2.0 (not everyone has been upgraded to the latest). For example, 5s instead of 3s grace period after disengagement to count faults; deduct points for driving between 10p-4am; braking in yellow light no longer counted for version 2.0; only your drives are used to calculate the score, you can no longer using FSD drives to increase the score as in 1.1. If you want to keep the score high, use FSD whenever you can.
You be notified, I got upgraded to 1.2 one day with email also, and the score is reset to 100.
I read somewhere that the safety score always internally uses 'medium' (why would you let the user reduce that sensitivity). So perhaps you might want to set it to the other way around 'early' so sometimes the ding will let you take action that might prevent the medium sensitivity warning to go off.
I agree that there are false or unjustified warnings. I get one in a fixed location when there are cars parked on a curved road, as I approach it doesn't know that the road curves and those stationary cars will not be collided with. (relevant to the FSD discussion, speed detection from vision is not as good as radar and so if it misclassifies a parked vs moving car this false warning will happen).
Othertimes it gets a warning when I am fully aware of the car in front and I'm slowing down with plenty of time.
That isn't working for me, if you get a 'forward collision warning' (the only thing that hits my score) in the 1 mile you drive manually near home vs the 20 you drive on FSD/AP, it makes the 'warnings per mile' statistic much worse. So I dilute it with manual freeway driving, where AP/FSD is probably safer than humans, and the score improves. So it's encouraging less safe behavior. AP/FSD is less safe than humans on streets probably. (In any case, it should be warnings per minute not mile).
Fortunately for me this doesnt have any financial impact as it's not used in the insurance cost, but it would be very annoying if it did.
Yeah, CA doesn't allow that. One time I had an FCW in the few miles that I drove and that trip score is 65 and dropped my overall score by several points from 100, then I have to drive hundreds of miles to get back up to 99. For me, driving manually on the highway here is futile, I got F2C all the time because of busy traffic and just no way to keep the distance far enough, so using FSD/NOA 100% of the time. The bad thing about using FSD all the time is numbing your responses, especially in emergency situation or simply because I forgot I am not in FSD mode.
clydeiii
Member
I don’t think your premise is true. The 20 miles on FSD/AP should count as perfect miles.
They are miles, but not perfect or flawed. They don't contribute to your score outside of adding miles.
That's how it's always been.
Safety Score Beta | Tesla Support
The Safety Score Beta is an assessment of your driving behavior based on several metrics called Safety Factors. These are combined to estimate the likelihood that your driving could result in a future collision. We combine your daily Safety Scores (up to 30 days) to calculate the aggregated...
clydeiii
Member
Exactly, they assist in drowning out the bad miles.They are miles, but not perfect or flawed. They don't contribute to your score outside of adding miles.
That's how it's always been.
Safety Score Beta | Tesla Support
The Safety Score Beta is an assessment of your driving behavior based on several metrics called Safety Factors. These are combined to estimate the likelihood that your driving could result in a future collision. We combine your daily Safety Scores (up to 30 days) to calculate the aggregated...www.tesla.com
They don't though.
That's exactly what others are saying. If you have 1 mile with close following, then put 1,000 miles on AP...you can have 100% close following. Unless this changed, this was proven 1,000 times over when people were gaming SS for FSD Beta.
clydeiii
Member
lol.
Not according to Tesla and pretty much everyone else. AP miles do not count to your safety score...good or bad.
clydeiii
Member
The biggest method people used to game SS back in 2021 was to get on AP on the highway and drive hundreds of miles back and forth until your score went up to the desired number.
clydeiii
Member
I have and posted it on Facebook. I drove 386 miles...for 3 miles I was bumper to bumper in 60mph traffic...some of that I was pretty close. I had 86% close following.
Again, this was with the older SS, but nothing has changed in the verbiage on that matter.
clydeiii
Member
You had 86% close following for those 3 miles and then you had 383 miles that didn't have close following so you're fine.
Here's the OG "game the SS" thread on TMC: Guide To A Perfect Safety Score for the FSD Beta Queue, or Tesla Insurance reasons. (Whether you like it or not)
Biggest piece of advice? "The goal is to let your car drive as much as possible. 90%+ of my drives are on AP/NoA. Aim for that %."
Here's the OG "game the SS" thread on TMC: Guide To A Perfect Safety Score for the FSD Beta Queue, or Tesla Insurance reasons. (Whether you like it or not)
Biggest piece of advice? "The goal is to let your car drive as much as possible. 90%+ of my drives are on AP/NoA. Aim for that %."
I think what you are confused about is that Tesla gives you a score for a single day. If you have bad miles on that day and then drive on AP the rest of the day...the bad miles will be heavily weighted on the day....BUT if you drive 3 days in a row all on AP with nothing bad, the days are averaged together weighted by total miles...so yes over the course of 30 days AP miles with no incidents with non-AP miles will level out the bad, but not on a single day or drive.
There are maybe 1,000 examples of this being true and Tesla explains that's how it's structured.
clydeiii
Member
The OP in the "game the SS" thread above explains your exact issue:
The struggle here that causes people to complain "The score is broken! I didn't follow closely 30% of my drive blah blah blah," is that the time/% it considers is ONLY when there is actually a car ahead of you, in the 1-3 second range. so if you have a bad following moment, and then there is never a car ahead of you while going >50mph for the rest of your drive, your score is screwed. If you are closely behind someone for 3 seconds, and then at the proper distance for 7 seconds before they are out of your lane or farther ahead of you, then this score is only based on 3 seconds of driving close, out of the 10 seconds total of being a behind that car at a proper distance. Boom. 30% close following for your entire drive. even if you drove 1000 miles.
clydeiii
Member
Similar threads
