Hi Guys

About to install a new crankset and cassette on my 10 speed bike. Saw a video on YouTube from SRAM about chain installation which shows that for sizing you wrap the chain around the two largest cogs on the back and front and then under tension bring the end of the chains together and from there mark off what looks like two inners worth of extra chain links, and connect from there.
Is that still the right sizing method?

Yes.

Or you could use a chain length calculator, like this:

Javascript Bicycle Chain Length Calculator

If the new crank is the same specification as the old one (same chain ring teeth), then the new chain will be the same as the old chain. Just hold them up together to get the right length.

No both front and rear have change from 50 27 to 52 28.

The method of wrapping the two largest and adding 1" gives you the shortest length chain that's safe for your drive train. Any shorter and you risk damaging the drive train if you accidentally shift to the big/big combination.

But you can generally go longer.

To find the maximum length that allows all combinations to work, thread the chain through the RD and onto the small/small combination, then pull the chain together until the RD rotates enough that the lower loop clears the upper pulley and splice there.

Anything between the shortest and longest is OK, and some, including myself, prefer to go longer in case we ever need to cut out a damaged link and resplice.

I also like to add a couple of links for safety. Also useful if you change cassettes

Why would you think it is wrong? their online user manual recommends the same method. I had a slightly longer chain by using a different method and had problems we gear skip, I just resized by their spec and it works great now.

Is there a performance difference between the shortest safe length and anything longer?

I like the shortest chain.

The chain runs with less bend thru the jockey wheels, so theoretically less friction,

possibly less chain bounce going over bumps,

lighter weight,

and, with fewer links to lube, $ saved on chain lube.

What puzzled me about SRAMs approach of looping it over the two largest cogs and then adding an inch or so is that once you loop it thru the back deraileur jockey wheels...that's adding more than an inch of extra travel for the chain...how does that possibly work. I guess I should try it rather than visualizing it but it's a $50+ chain and I don't want to screw it up.

but then you get some added life from the longer chain spread out over more links. You also get reduced tension when cross chaining big+big so less friction there

Actually not really when the derailleur arm is pulled full forward as it should be for the large-large combination. And as was said, that is the shortest possible chain. But there are other methods. You can run the chain through the derailleur and over the large front and smallest rear. The right length is when the line between the two pulley axles is perpendicular to the ground in that combination with no extra links added. I have always used this method as it used to be the preferred method recommended by Shimano. Works for me.

the only thing with that method is you have be sure to be within the takeup capacity of the RD, otherwise they chain can end up too short if you are using it outside of the specifications

Looking at my bike (one of them), the pulley axles are perpendicular, etc. and the chain is the shortest possible- less than two extra links, so that doesn't seem like a safe method.

You need same length as old chain, plus 3 extra links, because 52-50=2 and 28-27=1.

The extra inch in B/B +1" isn't for the derailleur, it's to provided the necessary slack for the chain to lift off the sprocket and allow shifting. If you simply looped the two sprockets without the added slack, the chain couldn't come off, which was fine for your 20" coaster bike, but not for one that depends on that ability.

You don't later add any more when threading nit through the RD. Modern derailleurs off all brands have the ability for the idler wheel to rise above the straight line distance on the lower loop.

FWIW - Sram derailleurs are 100% insensitive to chain length and/or cage angle because the jockey (upper) wheel is the center of cage rotation and therefore never moves as the cage swings through it's range. That contrasts with Shimano and Campagnolo where the cage pivots between the pulleys and the jockey wheel rises and falls with cage angle changes. That rise and fall makes the RD more sensitive to chain length to achieve good balance for jockey wheel height with all chainrings.

Well it doesn't allow for crazy big largest and second largest cogs, which are where the danger lies. But that really isn't an issue on small arm derailleurs. If you are pushing beyond the recommended limit, however, and that is an issue for you, perhaps one of the other methods would work better. But if you confine yourself to the recommended maximum cog for a short derailleur, which is usually 28 teeth, I don't see how it could fail you.

My cassette is 11-28 (Sram RD, 34/48 rings). If I put on an 11-30 (done regularly w/ Shimano, I believe)

and used your method, I think it would fail.

But it is easy to know how to avoid the problem. Just add a full 1" link for two teeth more than the specified 28 tooth limit that the method works for, the difference between 28 and 30. Done and done.

It could fail if you have a large spread in the chainrings. I'm not a believer in the idea that there's only one way to do things, and, like you use a chain longer than the minimum. However unless I'm 100% sure that I won't have an issue, I take a moment to confirm that I'll be longer than minimum BEFORE I cut the chain.

BTW - there's a fringe benefit to cutting chains longer, and that's if you make a mistake, you can always cut a long chain shorter, but not the other way around.

One other note that relates to the forum. When working on my bike, or helping someone I know, I might use or suggest a method different from what I'd post on the forum. However, I never know who's reading a post or what skills they have so I make every effort to only suggest practices that are 100% bullet proof.