[QUOTE=bkeiller;38722]I would have to think that with a tighter line the incidence would go up slightly. Any backup on this??[/QUOTE]
There's certainly evidence of this, but not as much as you might think. Here's the breakdown for the last 10 years, including playoff games. There are few good ways to format tabular data with this forum software, so my apologies for that.
[code]
line range: ot - non-ot: ot%
all 183 - 2827 0.0607973
pk to 2.5 30 - 406 0.0688073
3 45 - 601 0.0696594
3.5 to 6 45 - 730 0.0580645
6.5 to 8 32 - 442 0.0675105
8.5 to 12 22 - 369 0.056266
12.5 and up 9 - 279 0.03125
[/code]
So, there's a big drop-off once you get to games lined more than 12, but that's a statistically small subset, and below 8.5 it's difficult to infer a really strong pattern.
The bottom line is that in sports betting we're almost always dealing with statistically small sample sizes, so we shouldn't be surprised to see very noisy results when we break the data we have into subsets.
