4 ms·

Bad drivers should pay more for car insurance.

tristor 3y ago
True. The problem is that telematics can’t tell you who is a bad driver and who isn’t. Instead they approximate bad driving by rate of acceleration, brake pressure, and speed. To these systems anyone who drives the way you are taught to drive in any serious course outside high school would be flagged as a bad driver, despite statistically have lower accident rates. The things that actually make people bad drivers are things like driving unpredictability, failing to signal, driving under the influence of drugs or alcohol, and not paying attention / distracted driving. These are the things that cause most crashes. They also can’t be directly measured with telematics and none of the measurements we have is a good proxy. So most folks would agree that bad drivers should pay more for insurance, but it doesn’t follow that sending telematics to insurance companies is a useful or accurate way to do that. It’s certainly not worth the privacy trade off.
- pnut 3y ago
  Whether due to someone being distracted, under the influence, a crappy driver, or a spirited, trained driver, I'd say someone driving fast and accelerating/decelerating hard is inappropriate behaviour on shared transportation infrastructure, and don't have a problem with the net being cast wide enough to discourage the practice.
  - hansvm 3y ago
    It's too clumsy of a cudgel. The sensors themselves are imperfect, the thing you actually measure is probably the thing easiest to measure rather than actually pushing the tire grip toward its limits, even if you actually measure the right thing (hard acceleration/deceleration) you miss the fact that even an ordinary trip around town is often safer with more acceleration than a naive model would predict (e.g., short on ramps, or just that your speed when close to merging should probably be a bit higher than your intended merge speed in most cars), and actions like hard braking can rightly indicate appropriately slowing down for unexpected construction. Misaligned incentives aren't something to be trifled with (something something Goodhart's Law). Maybe cutting all the hard-braking at every stop sign would be worth it, but if even a few people chose to merge more slowly onto the freeway that would greatly increase accidents and have ripple effects in rush hours and traffic patterns.
  - tristor 3y ago
    > I'd say someone driving fast and accelerating/decelerating hard is inappropriate behaviour On what basis? You probably have something different in your mind than what I am actually referring to. I'm not talking about people skidding to a stop or laying a stripe of rubber when they take off, and short of that, there is basically no increase in risk from accelerating or decelerating more rapidly than a grandmother hypermiling a prius, which is how the insurance companies want you to drive. Regarding speeding, in most cases speeding is required to /not/ cause accidents. The correct and safest behavior is to observe the road conditions and your surroundings and drive predictably within the limits of those conditions. That means driving the speed of traffic, not what it says on the sign. Sometimes that speed is lower, sometimes that speed is higher. Sometimes the speed is different in different parts of the country or city. What's most important is that other drivers can understand and predict your behavior on the roadway, and that you can do the same for others. It's /far/ more dangerous to do 65 in a 65 when the prevailing speed is 80 than it is to do 80, but combining speed data with GPS data will still get you penalized by the insurance company for doing 80.
    - sfn42 3y ago
      That's bullshit. Higher speed generally means less margin for error in unexpected circumstances and more serious consequences when an accident does happen. And if you frequently have to brake hard that's a pretty good indication that you're driving recklessly - following too close, driving too fast, not being ready for predictable events such as a pedestrian crossing the road etc. Some times it's necessary. But if it's necessary every time you get in your car you're not a safe driver - practically every one of those hard decelerations occur because you are trying to avoid an accident. It follows that if you were to have a small lapse of attention or judgement, you wouldn't have been able to avoid the accident. If you give yourself more margin for error you rarely have to brake hard. Keep a good distance to cars in front, slow down in unpredictable situations, respect the speed limit because it was set for a reason. https://www.iihs.org/topics/speed https://www.iihs.org/topics/speed
      - tristor 3y ago
        > That's bullshit. This could only be said by someone that isn’t experienced driving on US Interstates. If what you say is true, why don’t we do 20mph everywhere? Could it be that the safe speed to drive is conditional, like I stated? Could it be that what I wrote is no bullshit after all?
- BurningFrog 3y ago
  This argument assumes that insurance companies can't do math and/or don't care that they're losing money on this. I'm 100% convinced they have both the math skills and the greed to know this is profitable for them. I'm less sure it's actually happening. Only proof I've seen is the internet comment a few levels up claiming it.
  - tristor 3y ago
    I never said it wasn't profitable. I said it wasn't accurate. Insurance companies aren't concerned with the accuracy, they are concerned with the profit. It effectively provides some "data" to give them license to increase rates, whether its accurate or not. As long as their rates outpace their claim payouts, they'll be profitable. It's also definitely happening, not only do insurance companies buy telematics data from automotive companies, they also provide discounts (sometimes) for customers that voluntarily provide telematics data using devices which connect to the vehicle OBD2 port. One example: https://www.progressive.com/auto/discounts/snapshot/ https://www.progressive.com/auto/discounts/snapshot/
    - BurningFrog 3y ago
      So you think it's accurate in predicting which customers will generate bigger or smaller claims, but not in who is a bad or good driver? To me those are functionally the same thing.
      - tristor 3y ago
        No, I think it gives a veneer of data to the insurance company demanding more money, irrespective of any correlation to claims rates.
        BurningFrog 3y ago
        Seems simpler to just raise the rates.
Spooky23 3y ago
What’s a bad driver? This stuff isn’t exactly peer reviewed data, and insurance companies don’t publish the telematics data they produce or how they make decisions. They are funny with how they evaluate certain things. If you have 3 accidents in a certain period, they’ll drop you regardless of fault because you’re unlucky. But they’ll also penalize you if brake heavily and avoid accidents.
- hansvm 3y ago
  The only winning move is not to play ;) I'm half serious. All else equal, if you drive less you'll have fewer problems that lead to insurance penalties.
  - Spooky23 3y ago
    Honestly, don’t be a dope is the optimal move. I’ve driven for 25+ years, and barring some stupid teenage male stuff and a few deer, have done so without incident.
    - hansvm 3y ago
      So, a bit worse than the average driver? Surely there's some better evidence you can throw at us ;) I'm kind of on your side. Not being a dope decreases the crash rate by 90% or so per mile last I checked ("crashes caused by driver error"). Drive enough and you'll still become a statistic though.
- r_klancer 3y ago
  Well, statistics matter here and one thing insurance companies know about is statistics. If you brake hard once or twice, no biggie. But, for sure, my prior is that someone who frequently brakes hard on the road is someone who gets themselves into jams they shouldn't have been in in the first place. Maybe it's a matter of failing to anticipate what's coming up ahead. Maybe it's a matter of where, when, and how often they drive. Both indicate risk that, arguably, should be priced appropriately. Same for health behaviors. Dropping you is a thing that happens when the expected value of your insurance payouts exceed the expected value of the premiums collected. With granular data from telematics, an insurer [*] could notify you to change your game long before they end up concluding, "Well, none of the data we have on this person differentiates them from the last 3 guys to cost us a million plus in settlements and claims, so sayonara" [*] thinking of TFA here I mean a suitably regulated insurer, not one that is somehow sneakily accessing telematics data without any transparency
bobim 3y ago
We all have our moments, and we have the right that it’s forgotten when it happens. Data collectes is forever.
_ea1k 3y ago
A big input for what defines "bad" is excessive speeding. Going 75 in a 45 is excessive speeding. But what happens when your daily commute takes you through a former construction where the computer thinks the speed limit is still 45?

True. The problem is that telematics can’t tell you who is a bad driver and who isn’t. Instead they approximate bad driving by rate of acceleration, brake pressure, and speed. To these systems anyone who drives the way you are taught to drive in any serious course outside high school would be flagged as a bad driver, despite statistically have lower accident rates. The things that actually make people bad drivers are things like driving unpredictability, failing to signal, driving under the influence of drugs or alcohol, and not paying attention / distracted driving. These are the things that cause most crashes. They also can’t be directly measured with telematics and none of the measurements we have is a good proxy. So most folks would agree that bad drivers should pay more for insurance, but it doesn’t follow that sending telematics to insurance companies is a useful or accurate way to do that. It’s certainly not worth the privacy trade off.

pnut 3y ago
Whether due to someone being distracted, under the influence, a crappy driver, or a spirited, trained driver, I'd say someone driving fast and accelerating/decelerating hard is inappropriate behaviour on shared transportation infrastructure, and don't have a problem with the net being cast wide enough to discourage the practice.
- hansvm 3y ago
  It's too clumsy of a cudgel. The sensors themselves are imperfect, the thing you actually measure is probably the thing easiest to measure rather than actually pushing the tire grip toward its limits, even if you actually measure the right thing (hard acceleration/deceleration) you miss the fact that even an ordinary trip around town is often safer with more acceleration than a naive model would predict (e.g., short on ramps, or just that your speed when close to merging should probably be a bit higher than your intended merge speed in most cars), and actions like hard braking can rightly indicate appropriately slowing down for unexpected construction. Misaligned incentives aren't something to be trifled with (something something Goodhart's Law). Maybe cutting all the hard-braking at every stop sign would be worth it, but if even a few people chose to merge more slowly onto the freeway that would greatly increase accidents and have ripple effects in rush hours and traffic patterns.
- tristor 3y ago
  > I'd say someone driving fast and accelerating/decelerating hard is inappropriate behaviour On what basis? You probably have something different in your mind than what I am actually referring to. I'm not talking about people skidding to a stop or laying a stripe of rubber when they take off, and short of that, there is basically no increase in risk from accelerating or decelerating more rapidly than a grandmother hypermiling a prius, which is how the insurance companies want you to drive. Regarding speeding, in most cases speeding is required to /not/ cause accidents. The correct and safest behavior is to observe the road conditions and your surroundings and drive predictably within the limits of those conditions. That means driving the speed of traffic, not what it says on the sign. Sometimes that speed is lower, sometimes that speed is higher. Sometimes the speed is different in different parts of the country or city. What's most important is that other drivers can understand and predict your behavior on the roadway, and that you can do the same for others. It's /far/ more dangerous to do 65 in a 65 when the prevailing speed is 80 than it is to do 80, but combining speed data with GPS data will still get you penalized by the insurance company for doing 80.
  - sfn42 3y ago
    That's bullshit. Higher speed generally means less margin for error in unexpected circumstances and more serious consequences when an accident does happen. And if you frequently have to brake hard that's a pretty good indication that you're driving recklessly - following too close, driving too fast, not being ready for predictable events such as a pedestrian crossing the road etc. Some times it's necessary. But if it's necessary every time you get in your car you're not a safe driver - practically every one of those hard decelerations occur because you are trying to avoid an accident. It follows that if you were to have a small lapse of attention or judgement, you wouldn't have been able to avoid the accident. If you give yourself more margin for error you rarely have to brake hard. Keep a good distance to cars in front, slow down in unpredictable situations, respect the speed limit because it was set for a reason. https://www.iihs.org/topics/speed https://www.iihs.org/topics/speed
    - tristor 3y ago
      > That's bullshit. This could only be said by someone that isn’t experienced driving on US Interstates. If what you say is true, why don’t we do 20mph everywhere? Could it be that the safe speed to drive is conditional, like I stated? Could it be that what I wrote is no bullshit after all?
BurningFrog 3y ago
This argument assumes that insurance companies can't do math and/or don't care that they're losing money on this. I'm 100% convinced they have both the math skills and the greed to know this is profitable for them. I'm less sure it's actually happening. Only proof I've seen is the internet comment a few levels up claiming it.
- tristor 3y ago
  I never said it wasn't profitable. I said it wasn't accurate. Insurance companies aren't concerned with the accuracy, they are concerned with the profit. It effectively provides some "data" to give them license to increase rates, whether its accurate or not. As long as their rates outpace their claim payouts, they'll be profitable. It's also definitely happening, not only do insurance companies buy telematics data from automotive companies, they also provide discounts (sometimes) for customers that voluntarily provide telematics data using devices which connect to the vehicle OBD2 port. One example: https://www.progressive.com/auto/discounts/snapshot/ https://www.progressive.com/auto/discounts/snapshot/
  - BurningFrog 3y ago
    So you think it's accurate in predicting which customers will generate bigger or smaller claims, but not in who is a bad or good driver? To me those are functionally the same thing.
    - tristor 3y ago
      No, I think it gives a veneer of data to the insurance company demanding more money, irrespective of any correlation to claims rates.
      - BurningFrog 3y ago
        Seems simpler to just raise the rates.

Spooky23 3y ago

What’s a bad driver? This stuff isn’t exactly peer reviewed data, and insurance companies don’t publish the telematics data they produce or how they make decisions. They are funny with how they evaluate certain things. If you have 3 accidents in a certain period, they’ll drop you regardless of fault because you’re unlucky. But they’ll also penalize you if brake heavily and avoid accidents.

hansvm 3y ago
The only winning move is not to play ;) I'm half serious. All else equal, if you drive less you'll have fewer problems that lead to insurance penalties.
- Spooky23 3y ago
  Honestly, don’t be a dope is the optimal move. I’ve driven for 25+ years, and barring some stupid teenage male stuff and a few deer, have done so without incident.
  - hansvm 3y ago
    So, a bit worse than the average driver? Surely there's some better evidence you can throw at us ;) I'm kind of on your side. Not being a dope decreases the crash rate by 90% or so per mile last I checked ("crashes caused by driver error"). Drive enough and you'll still become a statistic though.
r_klancer 3y ago
Well, statistics matter here and one thing insurance companies know about is statistics. If you brake hard once or twice, no biggie. But, for sure, my prior is that someone who frequently brakes hard on the road is someone who gets themselves into jams they shouldn't have been in in the first place. Maybe it's a matter of failing to anticipate what's coming up ahead. Maybe it's a matter of where, when, and how often they drive. Both indicate risk that, arguably, should be priced appropriately. Same for health behaviors. Dropping you is a thing that happens when the expected value of your insurance payouts exceed the expected value of the premiums collected. With granular data from telematics, an insurer [*] could notify you to change your game long before they end up concluding, "Well, none of the data we have on this person differentiates them from the last 3 guys to cost us a million plus in settlements and claims, so sayonara" [*] thinking of TFA here I mean a suitably regulated insurer, not one that is somehow sneakily accessing telematics data without any transparency

bobim 3y ago

We all have our moments, and we have the right that it’s forgotten when it happens. Data collectes is forever.

_ea1k 3y ago

A big input for what defines "bad" is excessive speeding. Going 75 in a 45 is excessive speeding. But what happens when your daily commute takes you through a former construction where the computer thinks the speed limit is still 45?