How accurate is the 50% battery discharge rule of thumb?

I've been looking at this graph from US Battery... (see attached)

[attachimg=#]

Perhaps I'm missing something here but the slope of the curve

between 40% discharge and 90% discharges makes me question

the 50% rule for battery discharge. (I know it's a semi-log graph)

The graph shows that if you run your batteries down 40%

(60% left) you should expect 1475 cycles in it's life.

However if you run it down twice that, using 80% of the battery

capacity, you should only expect 675 cycles.

But here's the catch that I'm wondering about...

Since 80% DoD is twice amps the 40% DoD one would

only have to cycle half as often.

So doing the math the life cycles for a similar total

energy usage would be: 1475 vs. 1350 (2 X 675)

That's less than 10% difference... not bad considering

how much better the acceptance is on a 80% discharged

battery and hence less overall time running the engine.

And not bad since you would only have to by half the

battery bank to get the same number of amps between

charges.

Surely I'm missing something considering how many times

I've heard the 50% rule but it sure looks like if you

really want to get life out of your batteries don't drain

them much below 20% ... and once you get get below

30 or 40% (and the curve straightens out) taking them down

to 80% might not be as harmful as you may think, especially

if it's only for that week long cruise now and again.

Here's a link to the original chart:

http://www.usbattery.com/usb_images/cycle_life.xls.pdfThen again perhaps this chart is not be believed as I've

also read that number of cycles to 80% will be less than

half the number of cycles of one cycled to 50%.

One would think US Battery didn't just make those

numbers up.