That's cute. I'm amused that when I stopped to make my five-second guess, I thought to myself, "it looks like 250-300, but I know it must be more, so let's say 400." Then I read on, saw the author had done the exact same thing, and thought, "oh, must be even bigger then." Yup.
For credit cards or shopping on a budget, I make a point of rounding up to the next dollar as I go, so as to consciously overestimate the total cost so far.
I suspect that this phenomenon arises because humans file 100 as "a really big number," and 20 as "a reasonable/moderate number," especially when thinking about the cost of purchases. It doesn't "feel" quite right that 100 is only five 20s!
Or, an example that still boggles me: $100 is only two new big-name video games on any of the most recent generation of consoles.
$100 is only two new big-name video games on any of the most recent generation of consoles.
In Australia it's only one, sometimes not even that.
I've noticed this when looking at cricket scores. At first glance, the total innings score often seems higher than it "should" given the individual batsmens' scores.
Note that this happens with multiplication also. There's a classic study of anchoring bias where people are asked to estimate about how large 1*2*3*4*5*6*7*8 is or are asked to estimate how large 8*7*6*5*4*3*2*1. People asked either underestimate on average and underestimate the first much more than the second.
This makes me suspect that there's a general tendency to underestimate independent of what operations are being performed.
ETA: Some of this is discussed (with citations) in a previous post by Eliezer here.
"Outsmart your brain by knowing when you are wrong":
http://troysimpson.co/outsmart-your-brain-by-knowing-when-you-are-w