I have not found that with reloading and the OCW method with having numerous charge weights in sequence that all show a similar velocity /pressure and selecting the middle charge weight and tweaking the OAL I have improved the accuracy. Plenty of examples and evidence of this being the most economical and reliable way to develop an accurate load. Think why does GGG target ammo work so well and accurate in so many rifles if charge weight is the number one accuracy cause? No factory ammo has the powder charge weighed its all thrown straight into the case and so a charge that is optimum is whats needed.
Following this method I got a load for my Creedmoor that I can throw charges straight into the case (once the thrower is set up) and have an ES in single figures and a group under 0.4 inch at 122m.
This was my reasoning behind tweaking the OAL, maybe it's not worth trying but I may give it a go and see what results I get just out of interest.
There are some faults in the reasoning in this post.
1) A factory sporting cartridge is unlikely to be loaded to the optimum pressure level for accuracy- if it is in your rifle that's plain luck, whereas you are comparing this situation to that of one of your handload recipes whose charge has been determined using OCW. Sporting ammunition manufacturers usually load their cartridges up to something close to the SAAMI MAP level to produce a high nominal MV - that's what sells cartridges to the average user, so if the SAAMI MAP is say 60,000 psi, a typical factory sporting cartridge is loaded to 57,000-58,000 psi in an industry standard pressure barrel. That may or may not be anywhere near the OCW mid point in your rifle whose barrel and chamber are unlikely to conform exactly (or in some cases even closely) to the industry standard if mass produced.
2) The reason that the GB NRA's contract 308 Win ammunition works so well in a large number of rifles is twofold. First, it is loaded with the original 155gn 0.3083" dia. Sierra MatchKing, one of the most tolerant tangent ogive bullet designs known to man and rifle builder. Secondly, the other side of the partnership - the barrel - sees near 100% conformity to a small range of variables in Fullbore / Palma / GB & Commonwealth 'Target Rifle'. That is standard 'Palma' contour blanks that have been bored and rifled slightly undersize compared to the SAAMI 0.300/0.3080 bore / groove diameters making them marginally 'tight', and chambered with one of a very small range of chamber designs, in the UK the 'Bisley 150' chamber the norm for many years (and for all I know still the primary one in use today). Barrel lengths range from 30-32" but a very high percentage will be at 30-inches. The NRA contract's starting point is that the ammunition must work well precision-wise (in terms of mean radius group sizes) and deliver 2,925-2,950 fps with acceptable ES Values with this single proven bullet design in a range of existing privately owned rifles with various barrel round counts / states of wear. When the original RUAG ammunition was selected, the manufacturer had to provide samples which were blind-tested by a group of top GB TR shooters at Bisley against alternatives in a group of rifles using the most popular barrel makes and with various states of wear. (The blind testing saw an NRA 'assistant' load cartridges into each shooter's rifle so that the shooter didn't know which make / model was being used in any given shot and couldn't therefore be psychologically influenced.)
The same thing could be done in Scandinavian prone match shooting where everybody uses the Sauer STR 200 rifle with barrels from the same source in either 308 Win or 6.5X55mm. (But in three optional barrel lengths which is an interesting and complication factor to throw into the mix!)
3) Linked to 2) it has been said often enough and proven in military testing that somewhere around 80% of a cartridge's performance in precision terms is down to bullet design, dimensions and how well they match the barrel's interior, and manufacturing quality / consistency. If the bullets in the GECO ammunition are good and if they match the rifle very well, playing around with COALs may have a small effect on group sizes, but he chances of a major change are so limited that any such change will only show up in averaging a fair number of groups shot under identical conditions off a bench with each group consisting of at least five shots. (That's an issue in itself with normal load development - often people who choose one combination of load factors over another on the basis of one three round group is larger / smaller than another are fooling themselves - even single five round groups usually mislead badly.)
These days for the handloader, there is another issue - the use of 'aggressive' secant ogive bullet designs with very low Rt/R ratio values, also the Berger Hybrid design which IME contrary to Berger's claims can be very fussy indeed as to seating position / jump. Again, that is one reason why Sierra's old 155gn MK #2155 and 168gn big brother are such tolerant designs and work in so many different rifles - a high Rt/R ratio of ~0.9 and 7-calibre radius nose sections make for relatively high drag designs for long-range match shooting, but ones that are very tolerant of different chambers and throats.
Pulling cr*p bullets from cr*p ammunition and replacing them with good quality models ('Mexican reloads') is a very different proposition and often gives excellent results, albeit it is often so much work that it is usually a better use of one's time to use off the shelf components throughout from scratch.
Having said all that, as has been pointed out, if the individual knows what he or she is about in handloading and makes sure that the new COAL doesn't see bullets 'jam-seated' in the lands, there is no safety issue and if people wish to experiment in such ways, well good luck to them - it's their time, ammunition expenditure, and barrels that are being used up.