Snap-to-income scenarios can come from below not just above

I'm still shaking out my software so this is not credible ret-fin analysis, just me playing around.  In this case I took the software for the "WDT utility with random life and random returns 'game' " and ran it like before but here I am just looking at real consumption (not utility) over time under the following circumstances (that are on top of the assumptions already mentioned in "The Wealth Depletion Time EDULC Simulator is live!")  The key things to note here are:

1. There is no reason to do this. I am just horsing around working on the software and curious.

2. Returns here are a more volative and nominal 8% arithmetic and 15% sd. They get geo-ed out in the sim time-averaging process

3. The spending plan is an age-based %-of-portfolio described in the past post

4. The "income" here is made up of two parts: (a) inflation adj SS of 16k starting at 70 and (b) (an initially nominal, but inflation adj thereafter) 10k of annuity income purchased from wealth at 75. Note that much more than 10k in annuity purchase or much higher vol means that there become a non-zero number of scenarios that can't afford to buy an annuity at age 75.  That is a topic for another post. 

5. No expected discounted utility calcs in this post. Just looking at real consumption in a volatile spending world. This is still part of a software shakeout.

6. The point of the visualization (to myself) is to show myself that while I typically think of consumption, in a scenario where wealth runs out, "snapping down" to available income...it can also snap up in a volatile variable spending scenario.  Frankly, though, I'm not convinced this would really unfold in real life.

This is what a snap-up in consumption might look like (again, ignore the hot mess of the graphic...I needed all the lines to see the snap up to a deferred floor). Only the first 200 of the 20000 iterations are charted: