Notes on "3+1" contest

[DeucesUp]

I'm probably not the right person to be starting this thread as I'm a sports betting newb and wasn't involved in the contest, but here goes anyway.

I've been interested in the results of the contest as a nice, fairly statistically significant sample of what of a group of mostly educated amateurs could do handicapping the NFL.

I copied the final standings into excel and here's some numbers, I hope I didn't make any additional errors:

<font class="small">Code:</font><hr /><pre>
Group W L T win% Ave$seas Ave$/wk
------------------------------------------------------------
Active 1251 1155 47 52.0% $593 $41
Dropouts 588 623 17 48.6% -$135 -$30
ALL 1839 1778 64 50.8% $141 $17
</pre><hr />

Obviously this is not a group of professional handicappers, but I think this gives a pretty good indication that beating the spread in the NFL is not trivial in the long run. The still active players beat the spread by a little but I think is important to include the dropouts as well to avoid biasing the results. The group as a whole barely beat spread and even the win rate posted by the still active players is not enough to overcome a standard 11-10 (+110) vig. Net results for the season would be a significant loss after juice (with the still active players almost breaking even).

Results might be improved a bit if players were allowed to shop around for the best lines and/or lower than normal vig. On the other hand it seems the way this contest is set up (if I understand the rules), the players may benefit from a "static" line. That is, the line is set early in the week and doesn't change. If major news on, say, an injury comes out, participants in the contest might be able to take advantage of a line which is no longer available at the sports books (because their handicappers have also incorporated the news into their line).

I'm not sure how big either of these factors are, but I'd be interested in some opinions.


-------------------------------



"Best games of the week"

I wasn't sure of the best way to look at how well people chose which game to put the most money on since I wasn't about to go through all of the picks, but I made a simple estimate:

<font class="small">Code:</font><hr /><pre>
Group total$ expect$
--------------------------------------
Active $24900 $24000
Dropouts -$9300 -$8750
All $15600 $15250
</pre><hr />

Where total$ is the actual amount won and expect$ is the amount which would have been won if $250 were bet on each game instead of $400, $300, $200, $100. Note that if there were a bunch of cases in the contest where people chose not to place all 4 bets in a week, these numbers could be flawed.

Not too surprizingly, it appears that prioritizing among picks is also difficult. The results for the group would have been essentially the same if a constant wager were used.


Honestly, I'm not terribly surprized at the results, but it is a little discouraging. Perhaps there are some experts in the group who can make money betting the NFL but their results are diluted by folks who are just doing this for fun. But this pool is probably a decent cross-section of readers of the sports betting forum, so it would appear that one would have be SIGNIFICANTLY more knowledgable than the average reader of this forum to win at betting the NFL.