Originally Posted by
tonywayne
Question about evaluating the "strength" of a trend...
Do you guys ever look at the "avg line, avg total" and compare that to the stat breakdowns of "Team, Opp"? So, for instance, in the trend above:
p:AL and A and season=2013 and line<=8 and division!=o:division and p:line<2
We see:
ATS: |
42-22-0 (0.37, 65.6%) |
avg line: -0.2 |
And then we see:
|
FG |
Pct |
FT |
Pct |
3s |
Pct |
BLKS |
O-RBND |
RBND |
Fouls |
AST |
TOvers |
Q1 |
Q2 |
Q3 |
Q4 |
Final |
Team |
37.03 |
44.9 |
19.52 |
77.0 |
7.91 |
36.1 |
4.16 |
11.12 |
42.98 |
21.41 |
21.45 |
13.86 |
25.2 |
25.6 |
24.7 |
24.8 |
101.5 |
Opp |
37.25 |
44.8 |
18.67 |
76.6 |
7.70 |
35.3 |
4.06 |
10.91 |
42.92 |
21.48 |
21.81 |
14.48 |
25.8 |
25.1 |
24.8 |
24.0 |
100.9 |
What I'm thinking is that for the game tonight with a 4.5-pt spread, the average results don't line up with the percentages. If the games are averaging a 1-pt finish, wouldn't that suggest an underdog cover, rather than the other way around? Should we be filtering favorites & underdogs into separate trends? I'm wondering how the positive & negative numbers might be skewing results - we're looking at favorites/underdogs, not actual integers.