According to prof. Nutt (the guy behind the big famous cannabis meta study) it increases the risk of an accident with about 30% (about the same as driving in the dark)
Alcohol in comparison increase the chance of crashing by some 2700%..
They obviously must have averaged out the intoxication somehow to get these figures.
So yes, you shouldn't do neither, but we are talking different ballparks of dangerous.
A main finding is that even though your cognitive abilities are impaired by cannabis, people that drive while intoxicated by cannabis think they are more impaired than they acutally are, thus driving more carefully which compensates to a certain degree slower reaction time etc.
With alcohol the effect goes the other way, people overestimate their ability to drive.
If you combine the two drugs you are even more likely to crash.
Funny story about Nutt is that he was was the UK's chief drug advisor until he published the study (ordered by the government), but they didn't like his evidence based findings so instead of adapting their policies to the findings, they sacked him. The War on Drugs don't stand up to scrutiny! :)
On the top of my head I think they ranked 20 of the most common drugs and found that alcohol was most dangerous (when you combined individual damage and damage on society), cannabis was in the middle of the pack, and ecstacy, LSD and mushrooms at the bottom.
The study did not take into account the medicinal value of the drugs, if you'd done that it would have been even more flattering for the above mentioned drugs.
And since this was a meta study, it also didn't exclude some old studies later disproven that found the same substances to be way more harmfull than they actually are, so in reality these substances are probably even less dangerous than his findings.