you are only partially correct . it is a guideline but it is more than that and yes it accounts for what we call sequence risk which means the order of gains and losses . by the way it was never meant to actually be a rule either , it was just some lab experiments to determine how low of a draw would make it through the worst of times , either that we actually had or the worst monte carlo simulations could come up with ..
think of it as building a house in a hurricane area . if i was building a house in nyc i would want to build my house to sustain the likes of hurricane sandy . that way i would be sure it could withstand the worst to date . while it still could get whacked by something stronger ,odds are pretty good it won't since sandy was such a rare occurrence itself .
so if you were a retiree starting out in 1907 ,1929,1937 or 1966 you retired in to the worst math for withstanding up to the worst case's in retirement history .
a safe withdrawal rate is one that would have endured all that history has thrown at you in the worst possible sequences . sequences meaning order of gains and losses .
an acceptable number was about 4% for a 50/50 mix . in fact to actually clear those time frames 100% you need to draw less than 4% or about 3.60% . 4% eliminates some of those worst date's .
the safe withdrawal rate is so conservative because it has to account for the worst sequencing in history that if we just eliminated those dates above the draw rate would be 6.50% .
90% of every rolling 30 year time frame here has left you with more than you started with at the end of 30 years with a 60/40 mix . 2/3's of the time it left you with more than 2x what you started with .
so if anything , trying to pass through those worst dates has been to conservative .
throw in another factor : life expectancy .
good planning says plan out to at least 90 or better 95 . even a couple has less than a 47% chance of one of them living that long so statistically you are planning for 30 years but reality say you will likely not be here for 30 years . that increases the success rate even more of not outliving your money .
next we throw human spending patters in . we tend to spend smile shaped .we spend more early on and less as we age until our 80's when healthcare goes up .
the end result is the things we no longer buy or do helps cover increases in what we still buy and do . that makes a lot less inflation adjusting needed than the every year increase the safe withdrawal rate figures .
so what you end up with is a number that ends up being far more conservative and in real life just keeps increasing it's probability of success through human actions .
so how does this help us in the future ?
well famed researcher michael kitces went searching for the common denominator to all the failures and what he found was this :
every single historical failure happened because the first 15 years of a 30 year retirement had the real return average fall below 2% . no matter how good things got after 15 years it was to little to late .
so now we can monitor things . if 5 years in to retirement we are still below a 2% real return , you may want to look at cut backs .
if you are way ahead you may want to take a raise
Here's an article on the bucket method. 
