View New Posts
12
1. St.Louis Blues - Edmonton Oilers 64-16-20

2.  0 1 2 3 4 5 6 7 8 9 10 0 0 8 9 3 0 1 0 0 0 0 0 1 1 23 24 27 17 0 0 0 0 0 0 2 12 30 53 36 17 10 0 0 0 0 0 3 5 49 65 58 23 6 1 0 0 0 0 4 15 42 50 51 27 6 1 0 0 0 0 5 8 37 54 33 12 1 0 0 0 0 0 6 6 17 38 31 8 4 1 0 0 0 0 7 3 12 13 8 11 1 0 0 0 0 0 8 1 0 8 7 4 0 0 0 0 0 0 9 0 2 3 4 1 0 0 0 0 0 0 10 0 0 0 1 0 0 0 0 0 0 0

3. Originally Posted by swordlord
Hello, i have done simulating models for predicting icehockey, football etc based on monte carlo in excel. Its pretty simple procedure. with this you can calculate 1X2, under/over, Spread etc.
Are you still interested? Just let me know.

First you need to calculate mean and standard deviation for both teams.
Then use box-muller method https://www.statisticshowto.datascie...le-definition/ for generate randomized goals. Remove all negative numbers (if occurs) because goals are positive integers. I usually generate 10000 numbers.
Now you can do your calculations in spreadsheet. Transform your numbers first into integers. Positive Goal-difference means home win and negative goal-difference means guest win. you can also delete all goal values from both teams over 7 because they are so rare.
Yes I'm still interested, thanks for posting

4. Always calculate these also: Most important is area between percentile 25% to percentile 75%. You should use those values only when calculating odds for NBA or some other form of sports which has high number of points.

• =AVERAGE
• =STDEV
• =MIN
• =PERCENTILE.EXC(25 %)
• =MEDIAN
• =PERCENTILE.EXC(75 %)
• =MAX
• =COUNT

5. I have no idea what your above posts mean. box-muller method? What is that table you show? Looks like a poisson chart. Sorry for my ignorance and lack of math knowledge. I am most likely light years away from your level. LOL. I do appreciate your sharing and interest, but this is over my head. I'm interested just not quite sure where to begin with this.

6. Instead of generating randomized numbers with box-muller method you can do it just use randbetween (min-max) or rather use numbers between percentile25% to percentile75%

First 12
Top