A few thoughts:

Different data from different perspectives:

Bullet companies have the luxury of testing and then listing data for specific bullets, or a limited number of specific bullets they make in that weight.

Powder company data has to encompass a broad range of bullets in that weight or in some cases weight range. Even if you restrict the data to jacketed bullets, there is variation in lead core hardness, jacket thickness and hardness, and bearing surface, all of which have impacts on pressure. If you restrict it to cast lead bullets you still have differences in alloy hardness, diameter, lube effectiveness and bearing surface that all effect both pressure and potential for leading.


Limited space in manuals:

There are practical limits to manual size and both bullet and powder companies want to include the latest gee whiz powders being marketed. When it comes down to choosing to repeat old data or list new data, the new data wins most of the time.

That means data on old go to powders like Bullseye, Red Dot, Unique, etc can be hard to find in newer manuals.


Powder formulations do change over time:

The National Center for Forensic Science maintains a database on powder chemical composition over the years that is useful in analysing gunshot residue. It's also interesting for a handloader who wonders if today's power is the same as that can made 20 years ago, and in turn how useful that data may be from 20 years ago.

For example you'll find six variations of Unique.

National Center for Forensic Science

Despite variation in the chemical formulation the over all burn traits should be very similar, but the take away is to use old data fairly conservatively by starting about 10% low and working up to in your firearm with your components.


Test standards have evolved:

Back in the day pressure was measured with a copper crusher. Since it relied on deformation of a copper cylinder, which also involved a brief time element while the copper deformed the peak pressure measured wasn't the same as it is now with much more responsive piezoelectric measurement of pressure.

That change resulted in new maximum average pressure standards for some loads, usually being adjusted downward and old manuals (mid 70s and before) may not reflect those changes. For example, the Hornady 3rd edition published in 1973 is regarded by some folks as having some fairly hot loads, and that's supported by comparison with load data in later manuals where the max loads for some of the cartridge, powder and bullet weight combinations are significantly lower, even with the same test firearm used in both manuals.


Charges listed in manuals are nominal charges:

Canister grade powders used in handloading are far more consistent than the bulk powders used by ammunition manufacturers and military arsenals. However, they still have some variation.

There are some powders that are acknowledged to be the "same" powder just with different labels. H110 and Win 296 is one of the better known examples. HP38 and Win 231 is another well known example. However if you look at a particular manual and you look at a particular page that lists both powders for the same bullet, it's not uncommon to find a .5 to .7 grain difference in the maximum loads listed on that page. If they are the same powder why is that? It's because they were randomly selected and packaged from different lots of powder that have normal lot to lot variation.

Your take away from that needs to be that if you have two cans of H110 from different lots, you should expect to potentially see the same variation in lots.

Again, this is why manuals almost always advise starting 10% below and maximum load and working up.


Surplus and bulk powders are not the same as canister powders:

What really makes me shake my head is when some internet expert starts citing a specific powder charge of a military powder like WC844 or WC 846 to duplicate a military load. They do that because they see a specification for a nominal load of powder and think it means something far more precise.

How it really works is that Dupont will ship a 10,000 pound lot of "WC 844" to an ammunition plant to produce 5.56x45 NATO ammunition. When the lot arrives, ballisticians will work up a load specific to that 10,000 pound powder lot to get the required velocity within the maximum pressure limits specified by the military.

For example with one lot of WC844 for use in M193 27.0 grains might produce 3,200 fps while another lot of WC 844 may produce 3,185 fps with a 28.0 grain charge. That's a significant difference. The end result is that with that 10,000 pound lot, you'll end up with about 2.4 million rounds of M193 that are loaded with a specific charge weight that may not be the same as other M193 loaded from a different lot of powder. Those different lots of M193 will however have the same average velocity.

This gets us to another scary issue with surplus pull down powders. If a company is pulling bullets on surplus ammo and then dumping the powder, you never know quite what you are going to get. It it all came from the same 24 million or so rounds loaded from that single lot of powder, you are good to go. But that's not likely and when powder from different lot is mixed, you're now clearly in crash test dummy territory with an unknown powder that may or may not be homogenous in the keg. If it isn't in a factory sealed keg from DuPont, I'll pass on WC844 and WC846.

WC844 is also a special case as the "WC844" specification was developed specifically for M193 and it was taken from one end of the much wider WC846 specification used for 7.62x51 M80 ball ammo. In other words if you think 1 grain variation is something, wait until you see a 1.5 or 2.0 grain variation in the larger round. But none the less you see some fool posting an exact charge weight to replicate M80 ball.

It gets worse when the internet expert states that WC844 is the same as H335 and that WC846 is the same as BLC(2). It's more accurate to say that WC844 is similar to H335 and that WC846 is similar to BLC(2) but that's as far as it goes. H335 and BLC(2), as canister powders do adhere to tighter specifications.

To make the point about the wide range found in a bulk powder consider that WC844 was just a subset of WC846, and then see where H335 and BLC(2) are at on a powder burn rate chart.

On one particular chart H335 is listed as #81 right after Benchmark, N133 and IMR 3031. BLC(2) is listed as #102, in between W748 and Leverevolution and H380. In between H335 and BLC(2) you'll find IMR 4895, IMR 4064 and IMR 4320. No one considers those to be the same powders, but you'll find an internet expert claiming an exact charge for WC846 when it spans that entire burn range. They'd never suggest a single correct charge that could be used for IMR 4895, 4064 and 4320, but they'll do it for WC 844 or WC846 because they don't know what they are talking about.

If you've googled some burn rate charts or compared charts in different manuals you've probably also discovered that no two powder burn rate charts ever agree completely.

The take away here is that if you decide to start buying and shooting surplus powders you need to know what you are doing and you need to be very conservative in your approach to load development.