Going to just try and interject facts here without any side tracking. The simple question is what should the bolt action rifle be doing for aimed fire. That's its purpose on the battlefield. I have pulled the accuracy data and audited it at the behest of this thread.
We can look at expected dispersion of bolt action rifles in terms of either MOA (minutes of angle) or mils. MOA = 2πR÷21,600 at a given range.
Lee-enfield rifles (those not worked over by the armorer for accuracy) were expected to achieve 2-2.5 MOA accuracy with stock ammunition, or a grouping of 3" at 100 yards
with 10% of any production lot tested out to six hundred yards, where 90% of the shots had to fall in a two foot circle.
On the .303 ammo side, a lot was passed for service if it achieved less than 6" at 600 yards. Thus the minimum standard for .303 ball is about 2.67 MOA.
By comparison match grade 7.62mmx51 ammo today comes in at about 1 MOA or less.
A lee-enfield can shoot sub MOA on a good day with support and with good ammunition and a tuned weapon, as can the K98 or MAS, but as its the norm we need to use, it is reasonable to expected 2 MOA from this rifle or really any properly manufactured bolt action rifle.
As for the K98 and MAS-36 there simply isn't much to choose from in terms of a different accuracy. All these rifles can also shoot sub MOA but a 2 MOA results for stock ammo/rifle/average shooter is reasonable. All can achieve sub MOA if using a rested firing position and tuned ammo/ optics.
So in a quick review of the data and converting to mils - all bolt action rifles should be .59 mils without any further human effects (suppression, breathing control, rate of fire, shooter training and quality considerations etc) That's normal dispersion with stock rifles and ammunition for an average shooter should be about that.
Bolt action SNIPER rifles using selected ammunition and an armorer tuned weapons should be at about .3 mils assuming a generic high power optical sight.
The difference in the bolt actions is really in the rate of fire. Lee Enfield is the clear winner there as it has a shorter lighter bolt and a magazine capacity delta. In a mad minute, the Lee-Enfield will trounce both K98 and MAS-36.
Rates of fire - aimed shots per minute to include reload time
Lee Enfield 24-25 (record rate of fire was 37 rounds/minute)
Far too much quoting of specific instances from the internet using tuned rifles, match quality ammunition, etc. What matters to the game is what the average enlisted shooter could do with his weapon. So - that's the proper data going forward for rifle accuracy and rate of fire. It's backed up by a mountain of practical shooting data for all three weapons, and in some instances, the same weapon is various calibers. I've personally fired two of these three weapons and I also find the data in alignment with my personal experience as an infantry officer. I trained snipers once upon a time so while not a world expert I have a wealth of practical knowledge in judging this for what's reasonable. Same same for the 1903 Springfield.
Thanks to all for bringing it to attention...next look will be on rifle sway and offset in raising to a firing position, but be aware that that has a lot to do with you exhaustion state and the load you are carrying. Again range fire is WAY different than advancing under fire with a load and trying to stay alive long enough to shoot. Rate of fire and suppression also counts. There SHOULD be randomization on raising to a firing position in a real field environment...the only question is whether or not there is too much or too little ATM. Quoting youtube videos of bench rested sitting shooters is simply NOT what is achieved by infantryman in real world combat environments.