I have two hundred cases for a 25-20 wcf. The case lengths vary a fair amount. most of them are just over the trim to length. About 20% are just a bit short. Can I trim them .001 less than trim to length? I'd like them to be all the same length; it makes adjusting the crimp much easier.

.001" is not going to matter to the case or your rifle. And since it headspaces off the rim it won't matter there either.
I've had my trimmer cut a bit more then it should have on more then one occasion and my cases ended up .005" short and zero problems.
Unless it's a case that head spaces off the case mouth I wouldn't worry about a few thousandths short

Unless of course you are building bench rest ammo or match grade stuff.

I believe that being short, but with unifrom length and square case mouths, it will be more accurate as the crimp will be more consistent. This is my first time using a roll crimp so I'm guessing, but it seems logical.

0.001" ain't no thang, 0.010" ain't no thang...even 0.100" ain't no thang in certain instances...especially in a rimmed cartridge and certainly a 25-20 WCF.

To keep it uniform, I would establish the shortest length and subtract a couple thou' and trim all the cases to this length, to be sure the all the cases a trimmed square. As long as the case neck is shorter by the amount the case grows at firing you won't have any problems. You might have to re-adjust your seating die if you crimp.

I haven't had any problems with accuracy with "short neck" cases as long as the load was worked up with uniform length cases...no matter how "short" they were...within reason...

You might have a problem with holding the bullet if the case neck gets "TOO" short...and that can be offset by sizing a few thou smaller...polishing down the sizing button or using bushing dies.

"I did some testing"...I know, people blow me off when I start with that...with 6mm Rem, 243 Win and 308 Win back in the day. The 243 and the 308 both have a rather short neck as does the 300 Sav and 300 WM, while the 6mm Rem has a slightly long neck in relation to the caliber.

I trimmed the 6mm Rem, original neck length ~0.350" down 0.150" to 0.200"...no problems with accuracy, bullet holding or anything else as far as I could tell...AND I made a chamber cast.

The total chamber neck length from shoulder/neck junction to lands was ~0.525". I took some '06 cases reformed them to 6mm Rem and trimmed to a neck length of 0.500". Now those were some "different" looking cases. What I did with those is another story.

I took the 243 and 308 necks down by 1/2 the origninal length and couldn't find anything really bad that couldn't be addressed or redressed...they just looked funny with the short necks.

Accuracy was still on par with the load developed with longer necks, loads didn't show any pressur problems, COAL was the same or slightly less depending on the bullet length, the added jump only impacted accuracy with the shortest bullets because the bullets couldn't be seated out to the lands...and not enough to bother with except with the 55-58 gr 6mm bullets where I lost 1/4" t 3/8" in group size.

I've also fiddled with straight walled cases...38, 357, 44 sp, 44 mag, 45 LC and 45 ACP, 444 M, 458 W, 45-70, 120...as far as lengths are concerned. I've made up some rat/snake/pigeion loads, light enought to do the job but not enough to destroy everything in sight. Good enough to take out the vermin and barely dent the barn wood. I could even resize recovered lead bullets and shoot them again or use plastic bullets that once were available.

Trimming used to be one of those anal things prior to the testing...now I just make sure the cases are trimmed to a uniform length and don't sweat the small stuff. There are a lot more things that impact accuracy and/or usefulness at a higher level like neck wall thickness uniformity or case weight uniformity, so I concentrate on those aspects.

The trim length is just to keep the case from growing out to where the throat begins and the chamber ends so you won't jam a bullet in the case by jamming it into the lands and cause pressure spikes.

I'm not sure if you need to worry about jamming problems with your cartridge, the 25-20 WCF is a pretty low pressure proposition, but jamming might cause accuracy problems.

Most chambers necks are plenty long, but to be sure, have a chamber caste made. I've had rifles with chambers necks so long they could be construed as "long neck/throated" similar to Weatherby's famous throating.

Long, round about way of saying "not to worry" in your application.

Luck

Few trimmers (if any?) will allow you to consistently trim within .001". The Wilson trimmers are pretty good, but it would be a struggle to keep trimmed lengths within .001". Some people trim on lathes, then maybe it might be possible? Most commercial "hobby type" trimmers won't even get close.

Worry not, though, as the learned members have posted above, .001" consistency is not required in trim lengths.

- mike

You might also check to see if you can find a taper crimper for that case...it might work better...I use taper crimpers in most of my straight walled cases if I can find one...trimming isn't a problem with taper crimpers.

I like taper crimpers much better for my 45-*** and pistols, and my lever guns.

Mike is right also...I use a Wilson trimmer with a micrometer adjuster and if my mind starts wandering so does the trim length.

I also did a short study on trim lengths...variations of 0.000" to 0.003" don't seem to make much difference even in my very accurate varminters...you need a HIGHLY accurate rifle, HIGHLY uniform shooting proceedure, extreme uniformity in ammo and NO wind to see any difference in that slight amount of length diffference and/or neck uniformity also, an that's only for competition shooters anyway.

You have to stay uniform within the parameters of the complete system your shooting. Splitting hairs doesn't work unless the shooter and shootee are up to it.

You achieve 90% of the potential accuracy and use with only 10% effort at the reloading end...getting that other 10% accuracy takes 90% more effort, in the TOTAL system, than is usually required in normal applications.

Luck

If you are going to crimp....make them all the same.