Let's say that row 3 is your then-current stasche size.
I'd sum up all your expense cashflows (into, say row 6), then compute a row 4 which is "annual expenses divided by stasche size" (aka "C4's formula is =C6/C3, formatted as a percent with 1 decimal place). Then, look across row 4 and see what the percentages look like.
If you want to model different growth rates, make a cell (say B1) of expected growth rate of stasche, and then make D3 = (C3 - C6) * (1+$B$1) and then run out row 3, 4, and 6 as many years as you need.
This is fairly rough and simple-to-understand method, but not particularly directly comparable to the Trinity study 4% SWR conclusion. It just gives you an idea of what is likely to happen (hence, looking at more than 2 significant figures is misleading at best). As for how to get back to something you can directly compare to the 4% SWR, any method will be tenuous.
I could imagine making a row 5, which is, for each year, the average of cash flows out from year 1 through the current year. Then, look at how that changes over time. (To make this easier, do all calculations in real dollars [meaning already adjusted for inflation] and take cell B1 down to "gains above inflation" rather than nominal gains.)
That would make C5 = average($c6:c6) and then fill that right, so D5 = average($c6:d6), etc. If you then look across row 5, take the highest average, divide by 0.04 (which is multiplying by 25), you'll get your "kind of worst case year" stasche required to maintain 4% SWR.
This is all hand-wavy, of course, but I hope it helps. Realize that in any modeling like this, it's possible to be off by 50%.