2026 Toyota RAV4 GR Sport PHEV

[Blueprint]

[RRocket]

I don't understand why the safety is only 7.5 out of 10.

How much more safety systems do you expect (or need) a car to have?

I have the same safety suite in my BZ and you'd REALLY have to be screwing around to have an accident. Like almost deliberate. The car will stop itself from hitting people and moving and stationary objects. See video below. If it can do all that, what more is expected and what's lacking?

Although this version hasn't been tested yet, the previous version was 5 star both for NHTSA and Euro NCAP.  The 26 Lexus NX (same platform) is a top pick IIHS and 5 star NHTSA.

https://youtu.be/Mkz1Q6OT89U?t=145&si=thdePlzya81yLqya

[Blueprint]

I don't understand why the safety is only 7.5 out of 10.

How much more safety systems do you expect (or need) a car to have?

I have the same safety suite in my BZ and you'd REALLY have to be screwing around to have an accident. Like almost deliberate. The car will stop itself from hitting people and moving and stationary objects. See video below.

Although this version hasn't been tested yet, the previous version was 5 star both for NHTSA and Euro NCAP.  The 26 Lexus NX (same platform) is a top pick IIHS and 5 star NHTSA.

https://youtu.be/Mkz1Q6OT89U?t=145&si=thdePlzya81yLqya

We don't have directives on these scores, but my approach is always one of caution when no testing is available from either IIHS or NHTSA. I default to 7.5 in these cases if the aids are behaving fine. Otherwise it's like grading a kid before he takes his exam because his older brother did good the year before.

One exception - the Prelude got a higher score despite not having been reviewed by IIHS or NHTSA since I took care of the crash tests myself 

[RRocket]

We don't have directives on these scores, but my approach is always one of caution when no testing is available from either IIHS or NHTSA. I default to 7.5 in these cases if the aids are behaving fine. Otherwise it's like grading a kid before he takes his exam because his older brother did good the year before.

It's not his older brother...it's the same person who's a year older. Who's historically done well on these very same exams. Aced them, actually. So while we don't want to presume, I think it's safe to believe the RAV4 isn't going to go appreciably backwards in terms of crash testing. And it's posh brother (NX) lends truth to this.

So when it gets 4-5 stars, you'll change the rating to 9-10? Or it will forever be 7.5 regardless of what happens? Because if you don't go back and change it, I feel your methodology is flawed.

I'm continually perplexed by these completely arbitrary and journalistically questionable methodologies..

[Blueprint]

We don't have directives on these scores, but my approach is always one of caution when no testing is available from either IIHS or NHTSA. I default to 7.5 in these cases if the aids are behaving fine. Otherwise it's like grading a kid before he takes his exam because his older brother did good the year before.

It's not his older brother...it's the same person who's a year older. Who's historically done well on these very same exams. Aced them, actually. So while we don't want to presume, I think it's safe to believe the RAV4 isn't going to go appreciably backwards in terms of crash testing. And it's posh brother (NX) lends truth to this.

So when it gets 4-5 stars, you'll change the rating to 9-10? Or it will forever be 7.5 regardless of what happens? Because if you don't go back and change it, I feel your methodology is flawed.

I'm continually perplexed by these completely arbitrary and journalistically questionable methodologies..

The previous gen RAV4 got no awards and more significantly marginal ratings from the IIHS on the updated side impact test:

https://www.iihs.org/ratings/vehicle/toyota/rav4-4-door-suv/2025

There's a strong chance Toyota revised the structure significantly with the 2026 to improve crash performance. Plus, note the "acceptable" rating for automated braking. Had I based my score on the 2025, it would have been a 6.5, so I'm giving the new one the benefit of the doubt.

As for the rating remaining there, the text states it has not been tested yet. When me or a colleague reviews the 2027, the scoring will reflect on the results of said testing once it's done.

In my "real" field, there's something called "engineering judgement" for grey areas. It's what I apply too in my "other" field 

[GreatBigAbyss]

^^^ After looking at the iihs safety ratings for the previous Rav, I understand and agree with what you're doing r.e. the 7.5 safety rating. 

[RRocket]

^^^ After looking at the iihs safety ratings for the previous Rav, I understand and agree with what you're doing r.e. the 7.5 safety rating. 

So the IIHS carries more weight than NHTSA results which were 5/5?

For example, if he'd (hypothetically) drop it a point because of a "marginal" from IIHS, would it regain that point by getting 5/5 from NHTSA? Is NHTSA results excluded entirely then?

And I don't point this out just for the Rav4, I see inconsistencies like this for other vehicles frequently. I've seen vehicle A with less total rating than vehicle B...but vehicle B was more highly rated (or the same) for safety by the reviewer. It seems arbitrary.

I totally get the "personal" ratings (how it "feels", drives, etc) and how those could be scored.  I'm less clear on scores using outside data.

Also, his last review of the Rav4 (2022) gave safety 8.0. So it's hard to understand how even using the "old" crash data he'd (hypothetically) end up with a 6.5....1.5 points lower than the old 2022 model.

[Blueprint]

^^^ After looking at the iihs safety ratings for the previous Rav, I understand and agree with what you're doing r.e. the 7.5 safety rating. 

So the IIHS carries more weight than NHTSA results which were 5/5?

For example, if he'd (hypothetically) drop it a point because of a "marginal" from IIHS, would it regain that point by getting 5/5 from NHTSA? Is NHTSA results excluded entirely then?

And I don't point this out just for the Rav4, I see inconsistencies like this for other vehicles frequently. I've seen vehicle A with less total rating than vehicle B...but vehicle B was more highly rated (or the same) for safety by the reviewer. It seems arbitrary.

I totally get the "personal" ratings (how it "feels", drives, etc) and how those could be scored.  I'm less clear on scores using outside data.

Also, his last review of the Rav4 (2022) gave safety 8.0. So it's hard to understand how even using the "old" crash data he'd (hypothetically) end up with a 6.5....1.5 points lower than the old 2022 model.

In 2022, the RAV4 was an IIHS "Top Safety Pick". Relative to that model year, and the testing methods of that model year, it did great, hence the 8.0. When the IIHS moved to its "updated lateral impact test", where the rig is sized and weighed like an SUV, test results went sideways. Manufacturers that crave IIHS awards, like Subaru and Mazda, went straight to their engineers and introduced under-the-skin changes. Toyota did not with the previous RAV4, so relative to its competition, its results got downgraded. NHTSA is useful, but not as detailed, but I'm aware of some IIHS biases (against smaller cars) in their communications.

So it's all relative to when the score is attributed - like water quality test at a beach. The test results are valid for that specific moment, until the next test is done.