Broaddus Tweets: Ratliff will be 3, Hatcher the 1

xwalker

Well-Known Member
Messages
57,202
Reaction score
64,708
CowboysZone ULTIMATE Fan
theogt;5083356 said:
I believe the goal is prediction, in which case there isn't a more useful, simple tool than correlation. There's really no need to complicate it just to sound smart. It's not rocket science.

No, it's not in many instances including any running statistics in the NFL vs winning as standalone statistics.

Without determining cause vs effect correlation is not useful.
 

theogt

Surrealist
Messages
45,846
Reaction score
5,912
xwalker;5083896 said:
No, it's not in many instances including any running statistics in the NFL vs winning as standalone statistics.

Without determining cause vs effect correlation is not useful.
If your goal isn't predicting wins, then I can't help you. Mine is, so that's why I use the statistic.

This appears to be a classic case of moving the goal post until you get the desired result. I don't care for such exercises.
 

TheSport78

The Excellence of Execution
Messages
10,396
Reaction score
3,674
This thread is like beating a dead horse. It's going absolutely nowhere.
 

Idgit

Fattening up
Staff member
Messages
58,971
Reaction score
60,826
CowboysZone ULTIMATE Fan
TheSport78;5083918 said:
This thread is like beating a dead horse. It's going absolutely nowhere.

Sometimes the dead horse just has it coming.
 

xwalker

Well-Known Member
Messages
57,202
Reaction score
64,708
CowboysZone ULTIMATE Fan
theogt;5083901 said:
If your goal isn't predicting wins, then I can't help you. Mine is, so that's why I use the statistic.

This appears to be a classic case of moving the goal post until you get the desired result. I don't care for such exercises.

You don't know what you don't know.

If you reviewed single statistical variables in a vacuum, you could conclude that a high number of run attempts in the 4th quarter "correlates" to winning big. That conclusion would be false and can easily be understood just using common sense and without anything complicated; however, common sense takes into account more than the single statistical variable.
 

Verdict

Well-Known Member
Messages
26,230
Reaction score
20,501
xwalker;5083933 said:
You don't know what you don't know.

If you reviewed single statistical variables in a vacuum, you could conclude that a high number of run attempts in the 4th quarter "correlates" to winning big. That conclusion would be false and can easily be understood just using common sense and without anything complicated; however, common sense takes into account more than the single statistical variable.

What he said!
 

Verdict

Well-Known Member
Messages
26,230
Reaction score
20,501
Some statistics are way more valuable than others. Some are highly relevant. Others much less so. None of the statistics referred to above take into the vast disparity in the objective to be obtained on a given play.

If it is first and goal on the one inch line, and the QB sneaks it in for a TD, to win the game, that one inch might mean a lot more than a 40 yard strike and a tackle at midfield to end the half. In one instance that inch gained won the game. In the other, the gain meant nothing other than a meaningless statistic and had no bearing on winning or losing the game.
 
Top