Predicted Score vs. Vegas Line - Week 2

djchad

Registered User
Forum Member
Nov 8, 2001
233
0
0
Melbourne, FL
Greetings all :)

I am back for Year 2 of the Predicted Score vs. Vegas Line thread. Last year, the 3 star games hit for 82%, the 2 stars for 47% and the 1 stars for 54%. Not bad, but plenty of room for improvement. I took off the first week of the season to tweak the numbers a bit. I made a few changes to the way I calculate the star ratings for each game. Details for star rating are listed below. I added a plus rating (+) to games where the majority of the predictions fall on one side of the line - these games deserve special attention. Each week I will post the games that have a star rating, and update as needed.

For the newbies, note that I pull the predictions from http://tbeck.freeshell.org/fb/predncaa.html and the lines from my book. I load all of the predictions into my spreadsheet and calculate the values. I throw out prediction values that are either too far away from the mean or virtually impossible. Then I press a few buttons and ka-pow!



1 star rating (*)
The delta is greater than or equal to 2.85 and less than or equal to 5.85.

Simply put, the average prediction is about 3 to 6 points on one side of the line.

Ex. Line -27.5, Avg Pred -32.72, Delta 5.22



2 star rating (**)
The delta is greater than or equal to 5.85 and the std dev is more than 60% of the delta.

Simply put, the average prediction is at least 6 points on one side of the line, but the predictions are not very reliable.

Ex. Line -16, Avg Pred -22.83, Delta 6.83, Std Dev 5.81



3 star rating (***)
The delta is greater than or equal to 5.85 and the std dev is less than 60% of the delta.

Simply put, the average prediction is at least 6 points on one side of the line, and the predictions are fairly reliable.

Ex. Line 14, Avg Pred 6.50, Delta 8.50, Std Dev 3.22


Plus rating (+)
The standard deviation is less than the delta.

Simply put, at least 68% of the predictions are on one side of the line.

Ex. Line 14, Avg Pred 6.50, Delta 8.50, Std Dev 3.22



Standard Deviation
- http://www.neatideas.com/stdev.htm

Delta
- The difference between the line and the average prediction.

Average Prediction
- The mean of all predictions.



Now, with all that said, here are the numbers with 27 of 39 predictions reporting....

Week 2

Code:
Favors	        Line	Avg PredDelta	Std Dev	Play	
Fresno State	8.50	1.49	7.01	6.26	**      +	
Iowa	      -37.00	-41.25	4.25	9.25	*	
Wake Forest	8.00	3.14	4.86	7.97	*	
Northwestern	8.00	2.30	5.70	6.24	**	
Virginia	-3.50	-8.29	4.79	10.40	*	
Ball State	14.00	8.83	5.17	8.89	*	
Marshall	19.00	10.10	8.90	6.66	**	+
Tulsa	       38.00	32.61	5.39	6.58	*	
Texas Tech	-12.50	-15.57	3.07	5.10	*	
Maryland	15.00	8.88	6.12	6.32	**	
Alabama	         7.00	3.96	3.04	5.69	*	
USC	      -21.50	-27.91	6.41	9.75	**	
Arizona	       12.00	6.68	5.32	6.90	*	
San Jose State	13.50	9.53	3.97	7.08	*	
Utah State	32.00	25.32	6.68	6.55	**	+
Georgia	      -29.00	-35.48	6.47	9.08	**	
Troy State	24.50	19.65	4.85	9.51	*	
North Texas	-15.00	-18.12	3.12	5.51	*	

<br>

Season	W	L	T	%
***	2	0	0	1.000
**	6	1	0	0.857
*	8	6	0	0.571
+	8	4	0	0.667
 
Last edited:

c20916

Slacker
Forum Member
Aug 19, 2000
3,677
6
0
49
St. Charles, IL
djchad, do you know if they update their projections once they make one? b/c Perry is questionable for the Fla st game, just curious if some of those projections took that into account, or once they found out do they adjust?
 

djchad

Registered User
Forum Member
Nov 8, 2001
233
0
0
Melbourne, FL
I don't think they are updated. I've never really paid attention to it, but I believe once the predictions are made, that's it. That is the down side to using predictions.

I'm about to post an update in a few minutes...
 

djchad

Registered User
Forum Member
Nov 8, 2001
233
0
0
Melbourne, FL
Final numbers for Week 2.

33 of 39 reported. No hanging Chad's. :thefinger:

- Northwestern drops from ** to *.
- Virginia from * to **.
- Texas Tech and Alabama dropped - no play.
- Utah State drops the (+)
- West Virginia, Houston, Ohio and Connecticut added.

Some really freakin' high standard deviations this week. And no 3 star games. :mad:

GLTA


Code:
Favors	          Line	AvgPred Delta	Std Dev	Play	
Fresno State	8.50	2.32	6.18	6.03	**	+
Iowa	      -37.00	-42.18	5.18	9.20	*	
Wake Forest	8.00	3.17	4.83	7.39	*	
Northwestern	7.50	3.70	3.80	6.55	*	
Virginia	-3.00	-8.63	5.63	9.69	**	
Ball State	14.00	9.32	4.68	8.15	*	
Marshall	19.00	10.50	8.50	6.23	**	+
Tulsa	       38.00	32.93	5.08	6.25	*	
West Virginia	-9.50	-13.02	3.52	7.84	*	
Maryland	15.00	9.40	5.60	5.96	**	
Connecticut	-16.00	-19.02	3.02	7.71	*	
USC	      -22.00	-28.82	6.82	9.64	**	
Arizona	       12.00	7.23	4.77	6.47	*	
San Jose State	13.50	9.48	4.02	7.67	*	
Utah State	32.00	26.12	5.88	6.38	**
Georgia	      -29.00	-36.85	7.85	9.26	**
Troy State	24.50	19.87	4.63	8.55	*
North Texas	-14.50	-18.03	3.53	5.25	*
Houston	       27.50	24.33	3.17	6.57	*
Ohio	       19.50	15.98	3.52	4.95	*
 
Last edited:

dawgball

Registered User
Forum Member
Feb 12, 2000
10,652
39
48
49
Well, I wish I would have seen this before I placed my wagers. I'm on three of the 2 stars (Virginia, Maryland, Marshall) and against one (Georgia). I'll take 3-1 with my Bulldogs waxing that ass, though!:D
 
Bet on MyBookie
Top