Is there a way to predict the outcome of any soccer game with 100% accuracy? The honest and simplest answer is…. no. Regardless of what your fantasy football friends say, there is absolutely no way to be 100% certain, but there is a proven, mathematical formula that gets really, really close! And it’s all thanks to baseball.

有沒有辦法以100％的準確性預測任何足球比賽的結果？ 誠實和最簡單的答案是……。 沒有。 無論您的夢幻足球朋友怎么說，都絕對不可能百分百地確定，但是有一個經過驗證的數學公式可以非常緊密地聯系在一起！ 這全都歸功于棒球。

In the early 2000s, Bill James derived a formula that could calculate the percentage of games a baseball team “should” win by the end of the season. It was coined as the Pythagorean Win Expectation and is still used by sports analysts around the world. In order to avoid boring you with the math, all you need to know is that it uses the relationships between Runs Scored and Runs Allowed by a team to predict how many wins they should have by the end of a season.

在2000年代初期，比爾·詹姆斯(Bill James)得出了一個公式，該公式可以計算棒球隊在本賽季末“應該”獲勝的比賽百分比。 它被稱為畢達哥拉斯式的“勝利期望”，至今仍被世界各地的體育分析人士所采用。 為了避免使您感到厭煩，您需要了解的是，它使用團隊得分和跑步允許的跑步次數之間的關系來預測到賽季結束前應該獲得多少勝利。

However, baseball and soccer are 2 completely different sports. For instance, baseball is based on a 162 games season where performance is measured by wins and losses. On the other hand, soccer usually revolves around a 38 game season measured by a point system. So, how can this formula predict outcomes for soccer? Well, it can’t. At least not without tweaking it first.

但是，棒球和足球是兩種完全不同的運動。 例如，棒球以162場比賽賽季為基礎，其中的表現是由勝利和失敗決定的。 另一方面，足球通常圍繞一個得分系統衡量的38個賽季進行。 那么，這個公式如何預測足球的結果呢？ 好吧，那不可能。 至少并非沒有先對其進行調整。

There are thousands of academic papers online explaining all of the intricate math that goes into the process of deriving this formula and there is no way I could ever understand them. Therefore, for the sake of simplicity, we will take the following magic formula as face value:

在線上有成千上萬的學術論文解釋了推導此公式過程中所涉及的所有復雜數學，而且我永遠無法理解它們。 因此，為簡單起見，我們將以下魔術公式作為面值：

Does this really work? Yeah, it does a pretty good job! Not only is the math pretty simple, but with some basic Excel knowledge, one is able to do this for all 20 teams in under 20 minutes! I used this myself to predict the outcomes of the last 10 games of the English Premier League and the results were extremely encouraging! Here is how I did it:

這真的有效嗎？ 是的，做得很好！ 數學不僅非常簡單，而且具有一些基本的Excel知識，一個人就能在20分鐘內為所有20個團隊做到這一點！ 我自己以此來預測英超聯賽最近10場比賽的結果，結果令人鼓舞！ 這是我的做法：

1- Get Your Data ReadySince we are going to try predict the outcome of week 38 of the Premier League, we need to gather all the relevant data up to week 37. That means Standings, Goals Scored, Goals Against, Wins, Losses, Draws, Points, etc. Thankfully, the Premier League has it all readily on their website for use. All we need to do is a simple Control+C and Control+V into our Excel sheet.

1-準備好數據因為我們要嘗試預測英超聯賽第38周的結果，所以我們需要收集到第37周的所有相關數據。這意味著排名，進球數，進球數，勝利，失敗，抽獎，積分等。值得慶幸的是，英超聯賽已經可以在其網站上方便地使用它們。 我們需要做的就是在Excel工作表中添加一個簡單的Control + C和Control + V。

2.- Calculate the Percentage of Points Predicted per TeamHere is where we get to use our magic formula! After you input it into excel on a separate column and apply, you will now have a percentage of points predicted for every team at the end of the season. All we have to do is translate it into actual points. However, the next steps are not as simple and involve some wacky assumptions.

2.-計算每個團隊預測的得分百分比這是我們使用魔術公式的地方！ 在將其輸入到excel的單獨列中并應用后，您現在將在本賽季末為每個團隊預測百分比。 我們要做的就是將其轉換為實際要點。 但是，下一步并不那么簡單，并且涉及一些古怪的假設。

3.- How Many Possible Points Can a Team Get?In theory, if a team today won all 38 games, they would have a total of 114 points (3 points per win). However, due to the way this formula was derived, we need to assume that a win is only worth 2 points (Fun fact: This was the official measurement for wins up to 1994). Therefore, we operate under the assumption that if a team won all 38 games, they would only get 76 points.

3.-一個團隊可以得到多少分？ 從理論上講，如果今天一支球隊贏得了全部38場比賽，他們將總共獲得114分(每次勝利3分)。 但是，由于該公式的推導方式，我們需要假設一場勝利僅值2分(有趣的事實：這是1994年之前勝利的官方衡量標準)。 因此，我們假設一個團隊如果贏得了全部38場比賽，他們只會得到76分。

4.- Calculate the Final Predicted PointsSince we have our percentage of points predicted per team and the total points a team could get, we multiply them in order to get the predicted points for every team at the end of the season. However, you will notice these numbers are extremely low. That is because we are still operating under the 2 point system, which means we need to adjust it back to the 3 point system we are familiar with.

4.-計算最終的預測點數由于我們擁有每個團隊的預測點數百分比和一個團隊可以獲得的總點數，因此我們將它們相乘，以便在賽季結束時獲得每個團隊的預測點數。 但是，您會注意到這些數字非常低。 那是因為我們仍在2點制下運行，這意味著我們需要將其調整回我們熟悉的3點制。

5.- Getting Our Adjusted Predicted PointsThis step is not as complicated as it sounds. All we need to do here is take the current, real world points for every team and convert them into our 2 point system. All you need to do here is take the team’s real world current points and subtract the number of wins that team has. Another more intuitive way to do this is by multiplying the number of wins by 2 and adding the number of draws the team has.

5.-獲得調整后的預測點此步驟并不像聽起來那樣復雜。 我們要做的就是為每個團隊獲取當前的真實積分，并將其轉換為我們的2分制。 您需要做的就是獲取團隊的實際世界點數，然后減去團隊的獲勝次數。 另一種更直觀的方法是將獲勝次數乘以2，再加上團隊的平局次數。

6.- Converting Into the 3 Point SystemOnce again, it is time to trust blindly in the math. While its derivation is not that complicated, this formula allows us to adjust the predicted points from the previous step. Now we get to see what the final expected points for every team at the end of the season and compare them to the points they currently have.

6.-轉換為3點系統再次是時候該盲目地信任數學了。 盡管推導的過程并不復雜，但此公式使我們可以調整上一步的預測點。 現在我們來看看每個球隊在賽季結束時的最終預期得分，并將其與當前得分進行比較。

As you can see, our predictions differ from the actual current points by only 1 or 2 points, and some are even spot on! How you choose to interpret the information is entirely up to you.

正如您所看到的，我們的預測與實際當前點僅相差1或2個點，有些甚至還存在！ 您如何選擇解釋信息完全取決于您。

One of the many things I did with this data was to calculate the difference between actual points and expected points and see if a team was performing better or worse than expected. This way, I could see which teams were underperforming and were expected to either win or tie their next games.

我使用此數據所做的許多事情之一是計算實際得分與預期得分之間的差異，并查看團隊的表現是否好于或低于預期。 這樣，我可以看到哪些球隊表現不佳，并有望贏得或打敗下一局。

By doing this (and using other metrics and some subjectivity), I was able to predict the outcome of 7 out of the last 10 games of the season, including Brighton’s upset and first win after losing 4 in a row, and the 1–1 tie between Tottenham and Crystal Palace!

通過這樣做(并使用其他指標和某種主觀性)，我能夠預測本賽季最后10場比賽中的7場比賽的結果，包括布萊頓的失意和連續輸掉4場后的首場勝利，以及1-1熱刺和水晶宮之間的紐帶！