Why use histogram to illustrated probability distribution Forgive me I am a newbie of random variables. I saw a lot of course which introduce the Discrete Random Variables which always be illustrated with a histogram or bar chart . Like below 

But in my understanding .I think it should be represented by a graph like below, all the probability value for the lowercase x {2,3,4,5} is point. not an area. (forgive me bad drawing). 
My question is why it should be illustrated by a histogram or bar chart instead of point? Please correct me if there is something wrong . Thanks.

 A: It is just a standard way to plot the distributions. 
Theoretically, a point is more adequate to symbolize a single numerical value, e.g. P(x=2) = 2/14. Nonetheless, and despite not having an intrinsic value within themselves, it is visually easier and quicker to represent the values with bars.
A: Showing the normal approximation to $\mathsf{Binom}(16, .5).$ 
x = 0:16;  pdf = dbinom(0:16, 16, .5)
hdr = "PDF of BINOM(16, .5) with Normal Approx"
plot(x, pdf, type="h", lwd=2, col="blue", main=hdr)
curve(dnorm(x, 8, 2), lwd=2, col="red", add=T)
abline(h=0, col="green2")


If $\mathsf{Binom}(16, .5)$ is approximated by data then a histogram on a density scale may be a more appropriate graphical display. A sample of size 10,000 approximates the binomial model
fairly well, but not perfectly. The histogram shows the simulated data data. Centers of open circles show exact binomial probabilities.
set.seed(2019)
X = rbinom(10^4, 16, .5)
table(X)
X
   1    2    3    4    5    6    7    8    9   10   11   12   13   14   15 
   5   29   92  300  675 1265 1743 1936 1757 1195  630  272   75   23    3 

hdr2 = "Histogram of large normal sample"
hist(X, prob=T, br = (-1:16)+.5, col="skyblue2")
 points(0:16, pdf, col="red")


Note: (a) Graphs made using R statistical software. (b) As illustrated in @Glen_b's link, a plot of the CDF of a discrete distribution may sometimes be preferable.
