We've had it down to -11C with no loss in house temperature so far (which would mean we're achieving less than 30kBTU/h heat loss heating 6300sqft of conditioned space).

The other bad assumption we made was that the backup heat can heat the loop for the hot water tank, when in fact it can only heat for space heating.

but that also means my domestic hot water can only be set to 45C/113F or lower if I am trying to avoid electric booster heat

Same for the ICF, if we take raw materials it only adds up to R25 or so, but field testing is closer to R38.

@Rob/Heat Loss Calc - I think ultimately it depends on the software that you use.  If the software factors in thermal bridging, then you'd fudge upwards the R value of SIP/ICF/whatever.  If the software doesn't factor in thermal bridging, then you wouldn't fudge for ICF, and you'd fudge down a framed wall (more so for traditional framed, less so for SIP).  I know what you're getting at, but depending on the software, sometimes you have to "fool" the software.  For typical 2X framed wall, the wood would represent well more than 10% of the conductivity area, so it would make a material difference in the heat load calcs, and with R1/inch for wood and R3.8 +/- per inch for EPS/Cellulose, it can be enough to make equipment sizing matter.  I do agree the infiltration estimate is overblown, but the infiltration numbers off the heat loss calcs, even in its entirety, doesn't offset the difference we are seeing in the real world, so there must be another factor at play.

@Rob/Passive Solar - The passive solar is helping significantly, and we're very fortunate to be in an area where smart meters were mandated, and our local utility gives access to hourly usage of hydro, that we can compare against weather station hourly statistics for sun and temperature, and since we're currently still in construction and haven't moved in yet, we've been able to get decent statistics for heat loss alone, not factoring in BTU gains from equipment and activities like computers and cooking, and the body heat that contributes.  All that being said, the passive solar isn't accounting the whole story, because we can look at night time statistics where temperature is coldest and heat loss is greater, with no passive solar effect.  We don't have much surface mass inside (no concrete or slate floors) that can help retain heat; the sun is basically hitting cork floors and drywall, so the carry-over effect isn't very pronounced.  Some of the initial numbers I'm looking at is hinting the passive solar gains are signifcant during the day (cutting heating demands by 1/4 to 1/2 @ 20F weather) but that the carry-over fizzles out about 3-4 hours after sunset, i.e. around 8-9pm the electricity usage climbs up to a point where it's fairly steady after that throughout the night.  Where I'm basing my claims of the variance of the heat loss calc to actual is in the morning the house is maintaining the set indoor temperature, where it has gone through the coldest temperature and where the solar gains have been "used up", and the BTU output of the Altherma can be correlated by the loop temp and the overnight outdoor ambient temperature.

@Rob/Methodology - I'm pretty much a laymen here, my background is not from construction, so do correct me where you see fit; I'm here to learn.  Not enough knowledge to be an expert, just enough knowledge to be dangerous. 

@BadgerBoilerMN/Infiltration - I think that was changed to average after a conversation between my designer and the Town's plan examiner.  The original heat loss calc was done with tight, but it was changed by the time we revised it after the Town had issues with our heat loss calc.  Likely they're not buying into the tightness of the house, as I recalled I had to sign off personally on a letter for the windows I'm using for them to allow the high window R value.  The one thing we constantly ran into was the Town throwing things back at us because they haven't seen it before, so everything needed an Ontario Minister's Ruling (in order to use the product) or engineer's stamp.  Stucco TAFS was a perfect example, where products existed for TAFS directly on ICF, and where the town insisted we put another inch of EPS with a drainage plane behind the EPS like an EIFS install, even though there's no wood behind it.  In the end it was many thousands of dollars later for us to bring an engineer out many times to inspect during the application so they can sign off on the install.  In hindsight I should have just added the layer, since the engineer's cost was almost as much as the extra layer of EPS.  SIPs were another example, where we were very limited in the choice of manufacturers; there's very few manufacturers that do have an approval for use in this wonderful province of ours. 

@BadgerBoilerMN/Wrightsoft - That's what my designer used, and I'm pulling numbers there.  I had used some online heat loss calculators (such as the one at builditsolar.com), since I don't have a license for it.

Victor
www.ecobuilthome.ca
A 4350 sqft Net Zero Energy inititative