First, yours again for a point of comparison:
As I'm sure you, personally, know Wulf, the thing about RAW vs JPEG is that which is better depends heavily on what was captured in the first place. I'm going to (somewhat) arbitrarily define the quality of the jpeg to be measured by how closely it renders the image to "reality" in the subjective mind of the photographer. Therefore, I'm not going to focus on things like noise reduction, etc, but rather it's ability to be psychic. Perfection, for a jpeg engine, is achieving a "window like" view into reality. The reason for doing that demonstrates why I personally prefer to shoot in RAW+ mode. If a JPEG's perfection is defined by not needing alterations, it's weaknesses are defined by how much recovery can be made for common errors. In other words, if it chooses incorrectly, how well can you fix it? .
With that in mind, I've prepared a few example situations.
First, is simply pushing a single stop. The goal here for me was avoiding blown highlights, unrealistic "haze" on blacks, and color skews. RAW is on top:
The result, for me, is a bit of a wash. On this particular image, there weren't a lot of highlights to protect, the dark area in the wheel well gets a bit hazy, and you start to lose contrast on the dirty hood, but overall it's not significantly different.
However, what if the white balance IS wrong, and you didn't happen to catch it? Say, for example, you're shooting broad daylight shots, turn to your right, see the car, and use your flash to fill without adjusting white balance? To protect the highlights, you wisely decided to expose to the left. Now, you need to correct the temperature cast, and push it a stop. What I did was adjust the color on the RAW, then adjust the jpeg color to match, then pushed both and hand adjusted the curve to protect as much as possible on both. With the JPEG, I had to balance extremely carefully before blowing highlights, over saturating and giving a weird look, losing blacks to haze, and in general just having to walk an extremely narrow line, but I think I did it successfully:
Overall, on this one for effort RAW wins easily, but the JPEG was still able to do it, just with more fudging. I do think I was unable to prevent a certain washed out look to the JPEG, but I think I could fix that given more time and effort.
Finally, what if it was even more wrong than that? What if the exposure is more or less correct, but you want more dynamic range, AND the white balance was even more off. Let's say the greens and blues threw off the sensor and made it think it was a sunny day, but you were actually exposed to a rainy day but shooting into the shadows of some trees?
To me, personally, the RAW wins hands down. I processed these as similarly as I could, but do to the nature of this I had to made trade offs based on my perception of what looked better I had to balance light and dark, how much highlight to protect, how much saturation looked bad to me, etc. I objectively feel the RAW shot has a certainly reality to it while the JPG shot just has a 'processed' feel to it. I'm not sure how to describe it more objectively than that, and I'm not sure how to fix it in the jpeg. Darkening the jpeg just lost too much to the black level, and lost too much detail, so I had to come up with a balance that to me seems over exposed compared to the RAW options.
Without knowing how the scene REALLY looked, those assumptions may be completely right or completely wrong, and it's impossible to tell.