308 Win was introduced three-quarters of a century ago and bullets have changed vastly in that time. Our understanding of optimal rotation (hence twist rates) has also moved on mostly thanks to work by Bryan Litz and his company Applied Ballistics.
Until fairly late in the 20th century, bullet component and manufacturing quality were very poor by today's standards from most makers with many having inconsistent thickness / weight jackets and similar issues. Such 'out of balance' examples were very spin-rate sensitive as they veered off the aimed trajectory in any circumstances, but even more so as rotational speeds increased. A good example is the issued ammunition provided in British and GB Commonwealth countries' 'Target Rifle' shooting (Fullbore/Palma is the US equivalent) which was normally shot with NATO-specification 7.62mm arsenal ammunition.
Among this ammunition's many faults were poorly made 144-150gn FMJBT 30-cal bullets, and some production lots were really dire. Yet competitors shot with this stuff over 200 to 1,000 yards! To make the trajectories more consistent, slower twist barrels were adopted, the norm being one turn in 14-inches in the 1970s/80s. Many range tests proved conclusively that this very slow twist rate gave better (well, less bad really) results than faster twists especially at long range.
Today's commercial match bullets, especially Bergers with their super-well made and consistent J4 jackets, can be
much spun faster than stability models say are actually needed without affecting precision. Fairly recently Applied Ballistics found that increasing rotational speeds to move the resulting Sg (Coefficient of Stability) value higher from the previously recommended 1.4 to 1.5 reduced bullet in-flight drag and allows the full design BC value to be obtained. in 308/7.62 military use, the US Army decided some years ago that the 175gn Sierra MK used in its match/sniper rounds has an optimum barrel twist rate of one in 11.25-inches at 7.62 MVs in the worst possible ambient conditions and has adopted that in its sniper / designated marksman etc rifle specifications, which in turn saw some commercial rifle manufacturers adopt it for some models.
As others have said, many disciplines have seen a steady drift to longer very streamlined bullets that have to be rotated faster to be fully stabilised - the 20 or so years of 308 Win in F/TR have seen 155gn bullets be largely replaced by much longer 185-230gn models, the Berger 200gn 200-20X model being the most popular. These require one in 10 rates minimum, but 9 or 9.5 have become widely adopted. Bryan litz is currently working on the long-range effects of reducing twist rates even further, to one in 8 or faster in 308 for instance. Interest and participation in very long-range shooting at beyond 1,000 yards has also pushed bullet stability issues, of which rotational speeds are one part, into greater prominence.
The sole precision discipline that has stuck to the slowest twist possible mantra is short-distance benchrest, but BR is a highly specialist field where effects are measured in thousandth-inch group size changes.
To have a play with the effects of the interaction of bullet models' characteristics and twist rates, visit here:
https://bergerbullets.com/twist-rate-calculator/
select a 30-cal bullet and typical MV for its weight and enter different pitch rates.