Originally Posted By: Audi Junkie
Extended braking distances seems to increase the likelihood of crashing into an object. -Fact. Worse protection at a given impact speed. -Fact. Crashing into a stationary object seems to model a single vehicle crash pretty well. -Fact. Same with roll-overs. -Fact.
Anyone who knows anything about statistics knows about probability too. The effect is not cumulative, everytime I walk out the door the odds of an accident is the same, it's not "inevitable". Knowing the statistics too, you should know that these numbers need to be re-compared after scaled against the general public. To simply say, "you are 2 or 3x more likely to die in a car" is misleading. You also need to consider your OVERALL chances and divide by that.
Frankly too, I don't think Mooving2 is here to discuss the issue, rather to hammer anyone who won't submit to his single interpretation of cherry-picked data.
Since a wiki reference was requested...
http://en.wikipedia.org/wiki/Category:Cognitive_biases
This is getting easier and easier. Here goes:
1.
Originally Posted By: Audi Junkie
Extended braking distances seems to increase the likelihood of crashing into an object. -Fact.
Audi Junkie- now that makes sense, but do you know to what extent it increases the likelihood of accidents in the real world for the vehicle types we are discussing? For example, the extended braking distance could make a 30% difference in accident rates between SUVs and cars, or it could only make a 0.1% difference in accident rates. Any data, or just guesses as usual?
2.
Originally Posted By: Audi Junkie
Worse protection at a given impact speed. -Fact.
a) Wait, let me get this straight and reword this in the context of this conversation to see if I can make any sense of it. You seem to be saying: (an SUV which has) "extended braking distance" (relative to a car) implies that (the SUV) has "worse protection at a given impact speed", right? Please correct me if I'm wrong here.
b) And you say this based on...what exactly? It seems to me that would depends on how the "extended braking distance" affects accident rates in the real world in SUVs vs. cars. Any data on this, or just guesses? (See #1).
c) In addition, you seem to relate "extended braking distance" to "worse protection at a given impact speed". I do not understand this. First of all, the protection you speak of would come likely come from a difference in impact speed or collision avoidance, not from a constant "given impact speed" you talk about. Now, if we were comparing 2 vehicles, identical in all ways except that one had "extended braking distance" relative to the other, then I could sort of understand what you're saying- the car with the "extended braking distance", would probably be more likely to get into accidents (how much more likely, you don't seem to know), and it might provide "worse protection" only because the impact speed would be expected to be higher when some accidents did occur (though you can't seem to quantify this "worse protection", either). The problem is, this isn't a case with 2 identical vehicles we're talking about here- it's SUVs vs. cars, so the "worse protection at impact" statement makes no sense. An SUV's size and weight are different than a car, and SUVs also have a lower fatality rates overall, so if an SUV's "extended braking distance" caused it to crash into another car at, say, 40mph, vs. a better braking car which crashed into another car at, say, 30mph, the SUV may still provide better protection at impact speed. Any data to show otherwise? Any data to back your statement about "worse protection"? You may argue that, in some cases, the SUV may get into an accident that better-braking car may avoid. Which again brings us back to (#1). Any data on this, or just guesses?
3.
Originally Posted By: Audi Junkie
Crashing into a stationary object seems to model a single vehicle crash pretty well. -Fact.
Fact? Or baseless statement?
a) Crashing into a stationary object models a single vehicle crash very well? Since I can only assume you're referring to the IIHS tests I've brought up again and again, I'm not sure how many times I have to post this, but now that I'm quick-quoting, I could go on forever. Here goes again:
Originally Posted By: moving2
IIHS explains "The barrier's deformable face is made of aluminum honeycomb,
which makes the forces in the test similar to those involved in a frontal offset crash between two vehicles of the same weight, each going just less than 40 mph.
Test results can be compared only among vehicles of similar weight. Like full-width crash test results, the results of offset tests cannot be used to compare vehicle performance across weight classes. This is because the kinetic energy involved in the frontal test depends on the speed and weight of the test vehicle, and the crash is more severe for heavier vehicles. Given equivalent frontal ratings for heavier and lighter vehicles, the heavier vehicle typically will offer better protection in real-world crashes."
Translation: the IIHS is clearly stating here that both full-width and offset frontal crash tests are meant to simulate a crash between two vehicles of the same weight, NOT a single vehicle crash. I'll even paste just the relevant part so you can see it alone:
Originally Posted By: moving2
IIHS explains "The barrier's deformable face is made of aluminum honeycomb,
which makes the forces in the test similar to those involved in a frontal offset crash between two vehicles of the same weight
I'll say it again for emphasis: The IIHS is clearly stating here that both full-width and offset frontal crash tests are meant to simulate a crash between two vehicles of the same weight, NOT a single vehicle crash. You seem to contradict the IIHS' statements about their own test, claiming it models a single vehicle crash, when the IIHS says it models a two-car crash between vehicles of the same weight. Care to make a fact-based argument on this, or just more guesses and incorrect assumptions?
b)
Originally Posted By: moving2
IIHS has explained it to you, I've explained it to you, and the fact still remains:
Because crash test ratings cannot be compared across vehicle weight classes,
nothing you have posted so far speaks to the relative safety of a Hummer vs. a car.
This is supported by IIHS' own statements above. Do you not agree with the IIHS' own statements regarding their crash tests?
Can you argue any of your points regarding the relative crash safety of any Hummer vs. a car using actual crash/fatality rate data, or only with more of your guesses and incorrect assumptions?
4.
Originally Posted By: Audi Junkie
Same with roll-overs. -Fact.
Fact? Or baseless statement?
a) Again, let me get this straight and reword this in the context of this conversation to see if I can make any sense of it. You seem to be saying: "Crashing into a stationary object seems to model roll-overs very well", am I right? Please correct me if I'm wrong here.
b) Do you even know how rollover testing is done? The IIHS does a roof crush test, while the NHTSA does static and dynamic rollover tests.
http://www.consumerreports.org/cro/cars/...lover-tests.htm
None of this involves crashing into a stationary barrier.
c) So, when you say "same with rollovers" referring to crashing into a stationary barrier, you do realize this is not what the NHTSA or IIHS consider a good measure of probability of vehicle rollover, right? This simply isn't how they test rollovers. Care to make a fact-based argument on this, or just more guesses and incorrect assumptions?
5.
Originally Posted By: Audi Junkie
Anyone who knows anything about statistics knows about probability too. The effect is not cumulative, everytime I walk out the door the odds of an accident is the same, it's not "inevitable".
I would hope everyone understands that accidents are not "inevitable". However, I don't know about you, but when I think about vehicle safety, I'm not thinking of the 99% of the time that I'm driving along perfectly safe and impact-free. I'm thinking about what will happen if I get into an accident, however small the probability that I will get into one. That's what crash testing is for, right? It assesses what may happen on those days when the unexpected happens- when you're hit by another vehicle in the front/side/rear, when you crash into a barrier, when you rollover, etc.
6.
Originally Posted By: Audi Junkie
Knowing the statistics too, you should know that these numbers need to be re-compared after scaled against the general public. To simply say, "you are 2 or 3x more likely to die in a car" is misleading. You also need to consider your OVERALL chances and divide by that.
a) Since you seem to know so much about probability and statistics, maybe you can explain this statement, because it makes no sense to me. Hypothetically, let's say the data shows the fatality rate in cars is 3x more than the fatality rate in SUVs. Now let's consider "probability" and "overall chances", as you say. Let's say the probability of getting into an accident at all is 1/1000. Now let's "divide by that", as you explain. OK, so 3x the fatality rate divided by (1/1000) = 3000. So what does this tell us again? Please do the math for us here and explain what your results mean.
b)
If you're referring to the overall probability of getting into an accident, that's already factored in- the IIHS fatality rates I presented earlier are expressed "per million registered vehicles", so how do these numbers not take into account "overall chances" again? Please explain this to us.
c) This topic seems to relate to that "fractions of a percent difference" you referred to in an earlier post:
Originally Posted By: Audi Junkie
Spyder's observation that we are talking about fractions of percents difference is right on point.
And I talked about this in my response, yet another post you chose to ignore. Instead of typing it all up again, I'll quick-quote it here to see if you can respond to it this time:
Originally Posted By: moving2
Spyder and Audi Junkie- I'm not sure I understand the references to "statistically small variable", "improvements in safety" (Spyder) and "fractions of percents difference" (Audi Junkie). If it's a "statistically small variable", and if we're talking "fractions of a percent" difference, then the "improvements in safety" must not mean much to you. Go ahead and look all the way back to the beginning of the IIHS study data (1978) and you'll see that, even then, car fatalities were about 155 / 1,000,000. While this is about 2.5x greater than the 2008 number, it is still 0.0155% of all registered cars vs. 0.0065% of all registered cars in 2008. If it is just a "fractions of a percent" difference that you see in this data, you might as well go back to a 1978 vehicle since the difference is "insignificant". And here's where your implications don't hold up- this isn't the percentage of all accidents involving a vehicle type that are fatal, it is the percentage of the total number of registered vehicles that are in accidents involving fatalities. As a percentage of all the vehicles registered, I'd expect the portion of vehicles involved in accidents to be small, and the portion of fatalities to be much smaller. This doesn't mean the small percentages are insignificant, however- I could see your point if the numbers were expressed as a percentage of all accidents, but not when they are expressed as a percentage of all registered vehicles- of course the percentages will appear small this way.
The point I'm trying to make is that it would be erroneous to use these small percentages to say that, in a car accident in modern vehicles, there is a tiny chance of there being a fatality, and further erroneous to then say the difference between cars vs. SUVs is insignificant. That's because, again, this data is not expressing the fatality percentage as a portion of all accidents, but instead it's expressed per million registered vehicles, so the tiny percentages alone are expected, and do not imply that the difference is insignificant.
So what was that you were saying about "fractions of a percent" difference again? Please explain. And, if you believe that, why aren't you driving a 1978 vehicle again?
7.
Originally Posted By: Audi Junkie
Frankly too, I don't think Mooving2 is here to discuss the issue, rather to hammer anyone who won't submit to his single interpretation of cherry-picked data.
a) Single interpretation? Cherry-picked data?
Throughout this thread, I've asked you over and over again to make your argument using facts and data, and try as you may- from crash test video, to stopping distances, to incorrect assumptions, to baseless statements labeled "facts" without supporting data, you have yet to come through in a way that proves any point at all about the relative crash safety of cars vs. SUVs when it comes to injuries/fatalities using crash/injury/fatality data.
b) Here's more data you chose to ignore and the conclusions I've drawn from it.
If the data is "cherry picked", please explain how it is cherry-picked, and provide us with data that proves your point. If you disagree with my interpretation, please provide us with your own interpretation, and make sure it's also backed by the data. Here goes:
i) Fatality rates (overall):
Cherry picked? Do you have different data? Do you have an alternate interpretation?
ii) Fatality rates (broken down by vehicle size):
Originally Posted By: moving2
According to the same IIHS data:
1. Small, midsize, AND LARGE cars all have an overall death rate HIGHER than even SMALL SUVs, not to mention midsize, large, and very large SUVs.
2. Small, midsize, AND LARGE cars have an overall death rate that is about TWICE that of even MIDSIZE SUVs, not to mention large and very large SUVs.
3. ONLY "Very Large" cars have an overall death rate comparable to SUVs- and that rate is matched or beaten by midsize, large, and very large SUVs.
Cherry picked? Do you have different data? Do you have an alternate interpretation?
Audi Junkie- your problem seems to be with data and facts. An argument unsupported by data is just that, an unsupported argument. And facts do not become facts simply because you claim them to be so, as shown by your "facts" which I've debunked here.
With all the crash/injury/fatality rate data available on the NHTSA and IIHS websites, you just cannot seem to construct a fact-based argument, backed by data, that supports your position. So instead, you resort to spouting off one-liners, incorrect assumptions, guesses, and videos without ANY supporting data that proves your point when it comes to injury and/or fatality rates in cars vs. SUVs.
And when I present IIHS data over and over again, you simply ignore it time and time again, claiming you're "not interested". Well, if you're going to keep posting, you're obviously interested, just not interested in responding to facts that don't fit your preconceived notions. So how about responding to the data for once? I've numbered my points above. I'm not expecting you to respond to any of them. In front of whoever on the forum is left reading this, let's see if you can stand behind your position and respond to a post for once in this entire thread. 
