Maybe this will help.
What is basically happening is the Zone System test changes the ratio between the speed point and the metered exposure from the ISO's Δ1.00 (10x) to the Zone Systems Δ1.20 (16x). It isn't so much about determining film speed.
The examples deal with exposure in testing and not exposure placement in use. Flare still has to be added to the mix.
This is 2/3 of a stop, already, towards ZS test results in lower speed than ISO test results. The film speed has not changed in this example.
2/3 stop is already gone. Just because of how you use the meter in Zone System. That is a big part of the reason.
Then, as Stephen says, flare has to be added to the mix. I also assert that for Zone System testing, you also develop for less time. This moves the 0.1 density further to the right (slower).
So while you could stop right here and see 2/3 of a stop is gone.
I believe other things move the speed difference between Zone System and ISO, left or right (faster or slower), more or less giving you a final result of Zone System having one full stop below ISO.
So once you get this one big one settled in your mind, we can start looking at the other little details. I like to finish with flare, because it's the hardest. But flare is a big one.
The range of EIs from most Zone System testing is generally from 1/2 to 1 stop slower than the ISO speed. I believe imprecision of using the camera's f/stop in ZS testing and experimental error are the major causes of the variance.
I was reading a bit today and looking at the the various graphs. It does all make sence, it just hasen't become second nature yet. I will need a bit of time before knowing this well.
My next question has been mentined already above. My next question was now concerning flare in the ZS and the shifting of the densities through the greater exposure.
Flare would now raise the densities on the toe. Are we then landing above the 0.1 speed point?
In this example, the curve on the left shows where Zone placement will fall when Zone I is placed at 0.10 over Fb+f and without factoring in flare. The curve on the right takes the 0.10 density placement of the curve on the left, but incorporates a 1 stop flare factor.
What do you think would happen to the Zone placement if instead of having to increase the EI by 2/3 to 1 stop, rate the film at the ISO and then factor in flare?
To clarify, a couple points:
There may not be a "real" speed difference, because fractional gradient and Delta-x say so. But because fixed density 0.1 is the Zone I target, there will be a numeric speed loss.
You lose this speed because you develop to approximately 0.58 gradient for Zone System versus approximately 0.62 gradient for ISO.
The difference between ISO and Zone System due to this fact... I agree... is "probably not as much as" Δ0.10 logE.
We can't lock down how much. One reason it can't be locked down: "Grade 2" paper of different manufacturers may have different characteristics, so the y-axis Δdensity is not locked down in Zone System standards. Sure we can give some examples of what some people get for actual difference on logE scale. But no matter what, this is going to be one of those "approximate" numbers.
Here is a new intrigue: We have a numeric calculation result in a lower EI. But the "real" influence of this factor is effectively "nothing". We changed a number we didn't have to. Later on this MUST come back to haunt us. We must have to make this up or cancel this out in practice!
So theoretically if we keep the ISO speed, add in flare we could be alright.
If I look at the latest graph from Stephen we need to decrease exposure using the ZS model, but we would need to increase development to compensate the higher zones.