There is something terribly wrong with their math. Solar energy reaching the earth's surface averages about 150 watts per square meter for a panel always mounted perpendicular to the sun (when it's up), averaged over a year*. Assuming their panel is a meter wide, and always pointed at the sun, it makes sense that the power INPUT to the panel is 'about 210 watts' per day, not its output. At 15 percent efficiency, that would give them 32 watts per day per panel. Twenty panels would give them 640 watts per day, on average. That's about a dime's worth of electricity per day. Not very many people could heat and electrify their homes for three bucks a month worth of electricity.
* Wikipedia - "Sunlight" - "The World Meteorological Organization uses the term "sunshine duration" to mean the cumulative time during which an area receives direct irradiance from the Sun of at least 120 watts per square meter.[1] Other sources indicate an "Average over the entire earth" of "164 Watts per square meter over a 24 hour day"
Watt is a unit of instantaneous _power_. In this case 210 watts is likely the power produced at maximum surface solar radiation of about 1000 watts per square meter because that is how solar panels are rated. Your 150 W/m^2 figure of solar radiation is the average over the whole year. To figure the average daily _energy_ hitting earth you have to multiply by 24 to get 3600 watt-hours per square meter. Multiply by 365 to get the total per year: 1,314,000 million watt-hours per square meter. Multiply by the given 15% efficiency and convert to more common units and you get 197 kilowatt-hours per square meter or about $20 of energy.
I wish they had given the surface area of the panel instead of just saying it was 1.65 meters long. The conversion from solar radiance to the 210 watt rated output needs area to see their assumed efficiency.
Also my gut feeling is that the 150 watts per square meter average solar radiation over the entire year is high. At 1000 watt per square meter peak you get a 250 watt average above the clouds (total surface area is 4 times the cross sectional area). Since an overcast day might only be 1% as bright as a sunny day (or approximately zero) the 150 W figure implies that days are 60% sunny.
The difference between energy and power is similar to distance versus speed. We pay for electricity based on energy (analogous to total distance) but solar panel ratings are in power (analogous to speed). You have to multiply by time to see how much total you get.