>>>Looks like Swami Jr. has a challenge from "Jake the Shake" (as in "Shake" from Aqua Teen Hunger Force....not the other one, Omar Abdel-Rahman)
>>>
>>>Swami Jr and Jake the Shake are currently tied at 21 wins....
>>>
>>>Week 3 picks from Swami and Shake
>>>
>>>Eagles vs Chiefs: Swami: KC Shake: Eagles
>>>Ravens vs Texans Swami: Texans Shake: Ravens
>>>Chargers vs Titans Swami: Titans Shake: SD
>>>Cowboys vs Rams Swami: Cowboys Shake: Cowboys
>>>Patriots vs Bucs Swami: Bucs Shake: NE
>>>Vikings vs Browns Swami: Vikings Shake: Vikings
>>>Saints vs Cards Swami: Saints Shake: Saints
>>>Redskins vs Lions Swami: Lions Shake: Redskins
>>>Packers vs Bengals Swami: Packers Shake: Packers
>>>Panthers vs Giants Swami: Giants Shake: Panthers
>>>Dolphins vs Falcons Swami: Falcons Shake: Miami
>>>Seahawks vs Jags Swami: Seahawks Shake: Seahawks
>>>49ers vs Colts Swami: 49ers Shake: 49ers
>>>Jets vs Bills Swami: Bills Shake: Jets
>>>Bears vs Steelers Swami: Bears Shake: Bears
>>>Broncos vs Raiders Swami: Broncos Shake: Broncos
>>
>>What are your records on beating the spread?
>
>Thus far I've been godawful. 12-20. My worst start in years. Note : My picks against the spread are not the same as straight-up.
>
>For reference :
>2012 : 146-110
>2011 : 120-136
>2010 : 142-114
>2009 : 124-132
>2008 : 135-121
>2007 : 123-133
>2006 : 128-128
>2005 : 138-118

Just for fun, suppose we chose randomly without any knowledge of the teams. This would resemble coin-tossing or a so-called "Bernouilli trial". The standard deviation of such a trial is

SQRT( #Games * (probability of win per game) * (probability of loss per game) )

So for 256 games it would be

SQRT ( 256 * 0.5 * 0.5 ) = 8

So we would "normally expect" the results of random game picking to be 128 +/- 8 wins, or somewhere between 120 and 136. By "normally expect" I mean within 1 standard deviation i.e. ~68%

You've never been worse than "expected" and you were significantly better than "expected" in '05, '10 and '12. Although, what's up with your worst showing being between your two best? ;)