Are computers better at guessing random numbers than humans? An integer between 1 and 100 is selected at random. Statistically, would a computer guess that number correctly more often than a human?
The assumption is that the computer has no bias at all, while the human has some, meaning that the person does guess all the numbers over time, just some more often.
 A: If the random integer is uniformly distributed across the range $1$ to $100$ then the probability of guessing the integer is always $1$ in $100$ regardless of any bias in the guess.
So, yes, a random guess will succeed in one trial in $100$ on average, but so will guessing $39$ every time, or any other strategy.
A random lottery ticket is no more or less likely to win than a lottery ticket based on your favourite numbers or the birthdays of family members etc. However, with a random ticket you are less likely to have to share any large prize that you win, because it is less probable that anyone else has chosen exactly the same numbers.
A: A randomly picked number is just as likely to match another randomly picked number as does a human-picked number.
Thinking that a computer-picked number has a better chance of winning than does a human-picked number because of the fact that $70$% of all wins are scored by a computer does not mean that computers are better at this, and commits the base rate neglect fallacy. Most likely, computers winning $70$% of the time is the simple result of $70$% of all tickets being picked by computers. That is, $70$% of all people chose the option of having their numbers picked by the computer random algorithm.
Simply put: the random tickets won more than hand-picked tickets did simply because there were more random tickets in the first place.
Your question would be a lot more interesting if we are trying to match a set of numbers picked by a human. For example, if I know a human generated a number between $1$ and $100$, and I have to match that number, I make sure to pick something like $37$, because psychologically, $37$ feels more 'random' to people than something like $40$.  
In fact, I was just watching the Numberphile video on the Monty Hall problem last night, and at some point they discuss the scenario of having $100$ doors instead of just $3$. So, you know that Monty is going to open up every door but one, and I was thinking to myself: "Hmm, I bet Monty is going to leave door $37$ open". Well .... just watch the video yourself to see if my 'randomly' picked number matched the number 'randomly' picked by the person in the video. (of course, this very story itself is anecdotal evidence ... but i think you see the point)
A: There is no method for rating your ability to guess a random number.
It is not possible to have a method that is correct more often.
Humans are bad at picking random numbers. I think the article is hinting at that. We have emotional attachment to numbers and proportion. To improve your odds, pick uncommon numbers because you are less likely to split the winnings and therefore your return per-ticket is higher (although still negative).
