I have been following this lateral load provision since if first came to the 09 IRC. This article from late 2009 discusses how it got in the IRC and it's implications.
New Code for Resisting Lateral Loads - Codes And Standards, Structure, Engineering, Building Science - Professional Deck Builder Magazine
I mentioned before, it was the committee response for "study of the band joist connection" that drove this provision. When the public commenter brought the lateral load detail in response, they brought the following argument:
Deck failures do occur where the deck is attached to the rim joist for lateral loads, but the rim is not adequately anchored into the floor system. Positive anchorage of the deck joists to the floor framing addresses this potential failure. The figure is based on a similar figure from FEMA 232.”
The argument from the proponent to get this in the IRC was about concern for the rim joist connection to the floor system. This is what I describe in my video.
Now we have testing that says neither the ledger connection to band joist or band joist connection to house failed at loads 4 times that greater than occupants can generate.
This is what I am trying to get your attention to. If there are other lateral load concerns (like nails in hangers), let's address them. But let's not have any more research, products, or alternatives based on code provisions included in the IRC that have been subsequently found inaccurate. If I have a deck with a ledger that is 30 ft. long, I've got at least 23 joists, 1500 lb hold downs on only two joists doesn't seem like an appropriate design. Let's talk about what is. Perhaps screws in hangers...
In the meantime, we should remove what we know is not necessary and provide more freedom to research and innovation to not be based on 1500 lbs in two locations. And let's relieve the decking industry a little. It's not all about the cost of the hold downs, it's the cost of their installation. In regions with finished basements or second floor decks it adds an interior remodel aspect of the job. With engineered floor joists, (six feet in of blocking) that can be a big deal.
--Of the four lateral load articles I reference, all of them were guided in some way by the current code language of two 1500 lbs connections. Future research should not be bound by that any longer.
--ICC ES has already worked on acceptance criteria for alternative methods for getting two 1500 lb concentrated load connections. Why? Why do we want this in the code until 2018 to continue to mislead innovation to creating a product equivalent to an IRC load of 1500 lbs that turns out is not really the target. Why? We need a clean slate.
I think I've provided enough information here to show that this subject does need further discussion and development. We do not have the answer yet, but it is not 1500 lbs in two locations according to the most recently released information.