Chance of predicting exact basketball scores?

[lorinda]

I know the two of you are interested in this, I simply don't have the time to do the math, but this is what I'd do.

I'd go to goldsheet.com's historical data page and for each game I'd look at the under over and the point spread.
I would assume that this is the most likely outcome, so if the u/o is 200 and the home team are 6 point faves, I'd guess the score at 103-97.

I'd then see how many teams hit the exact score.

Then you should be able to work out the odds of hitting both.
I'd do this for as many seasons as I could before deciding to rip my hair out and never look at those numbers again (been there, done that, gone mad)

Oh, and if this helped, I'd be interested in knowing the result.

Lori

[asdfasdf32]

Very good idea. I guess excel is going to get a workout.

[ImBen]

6689 NBA games since 1990 with integer spread and total.

3 of these have been graded push/push.

20030408 TOR MIA -6 172 83 89
20051109 SAS CHA 8 180 94 86
20001104 LAL MEM 9 187 98 89

[knicknut]

wow

[Lori]

[ QUOTE ]
6689 NBA games since 1990 with integer spread and total.

3 of these have been graded push/push.

20030408 TOR MIA -6 172 83 89
20051109 SAS CHA 8 180 94 86
20001104 LAL MEM 9 187 98 89

[/ QUOTE ]

Thanks.

I think if anyone has the time, my method would be a way of getting a lot more accurate results because of the inherent unlikelihood of the event, but this is obviously incredibly helpful too.

Lori

[bigalt]

[ QUOTE ]
[ QUOTE ]
6689 NBA games since 1990 with integer spread and total.

3 of these have been graded push/push.

20030408 TOR MIA -6 172 83 89
20051109 SAS CHA 8 180 94 86
20001104 LAL MEM 9 187 98 89

[/ QUOTE ]

Thanks.

I think if anyone has the time, my method would be a way of getting a lot more accurate results because of the inherent unlikelihood of the event, but this is obviously incredibly helpful too.

Lori

[/ QUOTE ]

isn't that what he did?

[Lori]

[ QUOTE ]
[ QUOTE ]
[ QUOTE ]
6689 NBA games since 1990 with integer spread and total.

3 of these have been graded push/push.

20030408 TOR MIA -6 172 83 89
20051109 SAS CHA 8 180 94 86
20001104 LAL MEM 9 187 98 89

[/ QUOTE ]

Thanks.

I think if anyone has the time, my method would be a way of getting a lot more accurate results because of the inherent unlikelihood of the event, but this is obviously incredibly helpful too.

Lori

[/ QUOTE ]

isn't that what he did?

[/ QUOTE ]

I would count the times that each team hit, not the times that both hit. Just to increase sample size A LOT.

having a 12k sample of 40-1 shots is a lot better than a 6k sample of 1500-1 shots, even though it's actually just an estimation.

Lori

[bigalt]

ahh, wise

[ImBen]

[ QUOTE ]
6689 NBA games since 1990 with integer spread and total.

3 of these have been graded push/push.

20030408 TOR MIA -6 172 83 89
20051109 SAS CHA 8 180 94 86
20001104 LAL MEM 9 187 98 89

[/ QUOTE ]

For those keeping score at home, the sample size is 6481, not 6689.

[ImBen]

6481 NBA games since 1990 with non-zero integer spread and total.

There were 254 pushes against spread [3.919%].

There were 100 pushes against total [1.543%].

Assuming these are independent (questionable), the likelihood of both occuring is 0.060%, slightly higher than the actual historical probability of 0.046%.