What’s the best way to get a good score if current score is bad?

[DaveG_NJ]

A 59 sounds like a normal driver to me.

My first score, being mindful about following distance and hard braking and hard turns scored an 83. I can get perfect scores on braking and turning by pretending I have a hot cup of coffee, filled to brim, sitting without a cover on top of my dashboard.

What I cannot avoid is safe following distance dings. In NJ and Philly, you leave a decent gap and like clockwork, you'll get passed and that same car will cut in front of you. You'll have to drop back and another car will cut it. Pretty soon, you're back where you started your drive. I think the only hope is to stay off multi-lane roads, or use AP as much as you can. I wish we could see how AP scores - it definitely brakes MUCH harder and takes turns MUCH faster than I do in pre-Safety Score driving.

 

[HX_Guy]

A 59 sounds like a normal driver to me.

My first score, being mindful about following distance and hard braking and hard turns scored an 83. I can get perfect scores on braking and turning by pretending I have a hot cup of coffee, filled to brim, sitting without a cover on top of my dashboard.

What I cannot avoid is safe following distance dings. In NJ and Philly, you leave a decent gap and like clockwork, you'll get passed and that same car will cut in front of you. You'll have to drop back and another car will cut it. Pretty soon, you're back where you started your drive. I think the only hope is to stay off multi-lane roads, or use AP as much as you can. I wish we could see how AP scores - it definitely brakes MUCH harder and takes turns MUCH faster than I do in pre-Safety Score driving.

The following distance is only scored at speeds above 50...so best to either stay below that on surface streets or use AP at all times.

 

[KDeanCville]

The first very important question is: Are you using this on Google Sheets or Excel? I found out early on that one particular function is different in Excel and the spreadsheet doesn't work on Excel. If you're using it on Sheets than we can dig deeper.

FYI - @n.one.one - I think I found the issue with the worksheet (for me). To clear out the sample numbers and replace them with my own, I selected the range of cells and deleted them all. However, the data columns are not contiguous. There are columns in between with formulas, so I suspect when I deleted the cells, I also deleted some of your formulas (hidden). When I later just cleared and copied one column at a time, it worked fine. Thx

 

[derotam]

Also apparently it only count Following Distance if you are driving over 50 MPH and not on Autopilot.

The following distance is only scored at speeds above 50...so best to either stay below that on surface streets or use AP at all times.

So how does driving under 50mph affect the cumulative score? Do all under 50mph miles just average in zeros or do they just not count at all?

 

[n.one.one]

So how does driving under 50mph affect the cumulative score? Do all under 50mph miles just average in zeros or do they just not count at all?

All miles count for something. How they affect your score depends on many variables including speed, AP vs. non-AP, and most importantly whether those miles are on a day when you have errors. So...if you drive 100 miles on AP and have a perfect drive you get a 100 score for that 100 miles. However, if you're finishing that 100 miles, reach the end of the ramp at 10MPH (with AP now off) and you jam on the brakes you have created an error event and that perfect 100 miles turns into a terrible 100 miles. You can still correct that day by doing some soft braking on the same day.

At the end of the day you get a rounded score on the Tesla app. Let's say for argument sake you got a 98 score for 100 miles on day one. Now you go out on day two and score 100 for another 100 mile day...your cumulative mileage weighted score is now 99. You can't simply average those two days because the daily scores are mileage weighted but in this example there is no difference.

 

[Silicon Desert]

It depend how many miles you have driven while being scored and how much time you can dedicate to some error free highway driving on AP to average your overall Safety Score up. You can calculate how many miles you need using my spreadsheet. Unfortunately I don't believe there is any way to reset your score.
Spreadsheet link:

Tesla Safety Score - Tracking & Planning v10

Safety Score This is set to read only. To make your own sheet go to the file menu, 'Make a Copy' and open in Google Sheets (not in Excel). No share requests please. There are hidden columns with some of the working formulas so be careful what you change or move. <a href="https://teslamotorsclu...

docs.google.com

Just to say that spreadsheet is pretty impressive and reasonably quite accurate. I know the details of the code, yet didn't think someone could figure it out to this degree of accuracy.

 

[jpfive]

Buy a new car and create a new Tesla account?

And pay for FSD again...

 

[derotam]

All miles count for something. How they affect your score depends on many variables including speed, AP vs. non-AP, and most importantly whether those miles are on a day when you have errors. So...if you drive 100 miles on AP and have a perfect drive you get a 100 score for that 100 miles. However, if you're finishing that 100 miles, reach the end of the ramp at 10MPH (with AP now off) and you jam on the brakes you have created an error event and that perfect 100 miles turns into a terrible 100 miles. You can still correct that day by doing some soft braking on the same day.

So for examlpe the hard braking

All miles count for something. How they affect your score depends on many variables including speed, AP vs. non-AP, and most importantly whether those miles are on a day when you have errors. So...if you drive 100 miles on AP and have a perfect drive you get a 100 score for that 100 miles. However, if you're finishing that 100 miles, reach the end of the ramp at 10MPH (with AP now off) and you jam on the brakes you have created an error event and that perfect 100 miles turns into a terrible 100 miles. You can still correct that day by doing some soft braking on the same day.

At the end of the day you get a rounded score on the Tesla app. Let's say for argument sake you got a 98 score for 100 miles on day one. Now you go out on day two and score 100 for another 100 mile day...your cumulative mileage weighted score is now 99. You can't simply average those two days because the daily scores are mileage weighted but in this example there is no difference.

Yes I agree that they all count for something. I am mainly concerned with the Unsafe Following because that is the one that is a little wacky and I don't understand how it works so I can't try to mitigate the 15% from my drive this morning(mostly on AP). All the other metrics are pretty easy for me to understand, its just this unsafe following that I can't figure out. It looks like even while on AP you can get hit for unsafe following which is dumb, especially if you have a following distance set to 7.

I did use that spreadsheet which is interesting, but it doesn't help me understand the unsafe following metric. Though according to it, I need about 250 miles to get to 99% and 1500 miles to get to 100% as long as I zero out most everything and keep the unsafe following down below 10%.

 

[DaveG_NJ]

If your score is low and you want to start over again, you can opt out and back in. It may not immediately reset everything, so if it doesn't work the first time, opt out and in again. The first time or two I did it, my old scores were preserved. On the third attempt, everything reset to zero (scores, mileage and days). I went for a drive around the block and have a score of 100 with a mileage of 1 with only one day of driving history. So now I need 99 miles and 6 more days of history (and a perfect score, of course).

 

[n.one.one]

So for examlpe the hard braking


Yes I agree that they all count for something. I am mainly concerned with the Unsafe Following because that is the one that is a little wacky and I don't understand how it works so I can't try to mitigate the 15% from my drive this morning(mostly on AP). All the other metrics are pretty easy for me to understand, its just this unsafe following that I can't figure out. It looks like even while on AP you can get hit for unsafe following which is dumb, especially if you have a following distance set to 7.

I did use that spreadsheet which is interesting, but it doesn't help me understand the unsafe following metric. Though according to it, I need about 250 miles to get to 99% and 1500 miles to get to 100% as long as I zero out most everything and keep the unsafe following down below 10%.

The general consensus as well as my personal tests indicate that Unsafe Following can't be corrected while on AP. Your best bet is to get on a highway over 50 MPH, then follow a truck/vehicle with a 1 to 3 second gap between you. To measure the seconds watch for the lead vehicle to reach a point (sign, pole, etc.) and start counting one-one thousand, two-one thousand, etc. Try to keep that time about 2 seconds but no less than 1 second and no more than 3. I suggest a truck since they can't stop very quickly. Keep in mind this will only change your unsafe following score on the same day you incurred the ding.

 

[derotam]

The general consensus as well as my personal tests indicate that Unsafe Following can't be corrected while on AP. Your best bet is to get on a highway over 50 MPH, then follow a truck/vehicle with a 1 to 3 second gap between you. To measure the seconds watch for the lead vehicle to reach a point (sign, pole, etc.) and start counting one-one thousand, two-one thousand, etc. Try to keep that time about 2 seconds but no less than 1 second and no more than 3. I suggest a truck since they can't stop very quickly. Keep in mind this will only change your unsafe following score on the same day you incurred the ding.

Yeah I came to the same conclusion about AP. I have done exactly as you have described for determining time distance. I think there is some weirdness in that metric calculation though.