ggplot2简单使用_用ggplot2预测足球比分

ggplot2简单使用

胡闹 (Horsing Around)

In one of my earlier posts, I mentioned that the scores in a football match can be approximated somewhat using the Poisson distribution, but I didn’t go too much into the topic. Well, you’re in luck today… we’re going to have a look at the subject, and by the end of this post we’ll have Ggplot2 visual illustrating the likely scores of a football match, which really is excellent.

在我以前的一篇文章中，我提到足球比赛的得分可以使用泊松分布 ( Poisson distribution)进行某种程度的近似，但是我并没有过多地讨论这个话题。 好吧，今天您很幸运……我们将对这个话题进行一番探讨，在本博文结尾处，我们将使用Ggplot2视觉效果来说明足球比赛的可能得分，这确实很棒。

First off, a little bit on the roots of the Poisson distribution. The discovery of this probability distribution is attributed to Siméon Denis Poisson, a French mathematician from the nineteenth century. Like many of his contemporaries at the time, he dealt mainly with its theories. It wasn’t until Russian statistician Ladislaus Bortkiewicz published his book The Law of Small Numbers in 1898 that the Poisson distribution became widely used in practice.

首先，介绍一下泊松分布的根源。 这种概率分布的发现归因于19世纪法国数学家SiméonDenis Poisson 。 像当时的许多同时代人一样，他主要研究其理论。 直到俄国统计学家Ladislaus Bortkiewicz于1898年出版他的《 小数 定律》时，泊松分布才在实践中得到广泛使用。

Bortkiewicz famously had access to a data set of the Prussian army over a period of 20 years which recorded the number of soldiers that were killed by their horses’ kicks. Grim.

著名的Bortkiewicz可以使用20年来的普鲁士军队数据集，该数据集记录了被他们的马踢杀死的士兵人数。 严峻。

Anyway, he showed that these deaths could be approximated by the Poisson distribution, and the rest, as they say, is history.

无论如何，他证明了这些死亡可以通过泊松分布来估计，其余的，正如他们所说的，都是历史。

Now… how does one go from Prussian cavalry to football scores?

现在……从普鲁士骑兵到足球比分如何？

做我和你的屁股 (Making an Ass of U and Me)

Let’s first have a look at the basic assumptions that satisfy a Poisson process:

首先让我们看一下满足泊松过程的基本假设：

1. The probability of an event occurring in a given time interval does not vary with time

   在给定的时间间隔内发生事件的概率不会随时间变化

2. The events occur at random

   这些事件是随机发生的

3. The events occur independently

   事件独立发生

Now think of goals as an event in a football match. If we had a metric that could express a team’s scoring rate/intensity over 90 minutes, we can say that the first assumption is satisfied. This would’ve been a pain in the arse to define, but thankfully the sports nerds (people a hundred times more intelligent than me, I should add) have come up with expected goals (xG) for this. Thanks for today!

现在，将目标视为足球比赛中的一项活动。 如果我们有一个指标可以表示一支球队在90分钟内的得分率/强度&#