In light of crushing and faulty human bias, I turn my college picks this week over to pure science. Saturday’s unmitigated bloodbath proves beyond reasonable doubt that I cannot be trusted to pick against the spread with any realistic expectation of accuracy. Therefore, I shall refrain from further embarrassment by relying on the sage-like wisdom of the Excel Spreadsheet. I will couple this leap of analytical faith by also upping the ante against Hubris- I’m going double or nothing, baby, and posting all my picks in The Pool in which I compete against the RedHookRaider, and where I have been getting my ass handed to me in spectacular fashion over the past 5 weeks. And because any use of a “jump” or “break” within posts turns all the previous posts cursive on this wonderful blog of ours, I am forced to subject you, dear reader, to the cold and brutal logic of the entire process.

(Please note that all numbers cited  to below refer only to Yahoo’s College Pick ‘Em Pool, and do not include any outside statistics and/or trends. As lines change over the next couple days, I may update some picks, but will note where and when I do so.)

In years past, I have made picks based on blind faith, gut reaction, and dumb luck, and this process has provided middling results. This year I was determined to data-mine results and ride tidal-wave trends and rational investigation to huge wins. But I found myself unable to trust the numbers I was given, continuing my proclivity to make the same ludicrous mistakes I always had. For instance, outside of Week 1, no less than 7 underdogs have beaten the spread each week, out of an average of 19 total games on the slate. In fact, 10 underdogs have beaten the spread 4 out of 10 weeks, going better than 50% in 3 of those weeks. Yet I stubbornly chose an average of 5 underdogs per week, and was justly slaughtered. The graph below tracks the number of underdogs who beat the spread each week, as well as the number of ‘dogs, picked and correct, by myself and RedHookRaider. As you can see, our stubborn insistence on going with favorites hasn’t helped us out one bit:

In fact, our respective winning percentages have been beaten by the underdogs alone 3 weeks: 

But once the decision is made to start picking more ‘dogs, the problem then becomes, which dogs to pick? In only 4 weeks out of 10 have more than 50% of the underdogs beaten the spread, so if you’re picking 8-10 ‘dogs, they better be the right breed, or you’ll get your face mauled off. A couple weeks back I started looking at the size of the line, and it’s relation to teams covering, but found no real correlation- teams covered 30+ point spreads just much, and even more frequently than spreads between 15-20 points. Plus, the lines are steadily decreasing as less teams play out-of-conference cupcakes. So I started to take a look at conferences, and how their teams fared against the spread. And some numbers began to stick out. As the chart below demonstrates, the least reliable favorites hail from the SEC, Big 12, and Pac 10 & ACC. And, since most teams play games in-conference, your most reliable underdogs are from the same 4 conferences:

So with the cogent knowledge that I need to pick 10 underdogs, and that they need to be in games involving the SEC and Big 12, I went over the slate for this Saturday. In fact, I picked every dog from those conferences, which gave me 6 dogs already. I then added 2 ACC teams and a Pac-10 team, as well as Wisconsin with no line, and there’s my 10: 

Some extra clarification points: With no line on the Wisconsin game, I simply picked the winner. Also, logic  tells me to go against both Miami and FSU (which would keep the same numbers of ‘dogs), but I can’t pick against the U in this troubled time (NOT SARCASTIC) and I could never forgive myself for picking against my ‘Noles, who historically are one of the worst plays against the spread. Also, these picks attained a second goal of mine for the week: don’t pick against rooting interest- it makes for a shitty viewing experience. I truly dislike Tennessee and have nothing against LSU, but was forced to cheer for the Vols last week because I had picked them to win. No longer! Even though I should’ve picked Tennessee as an SEC ‘dog, do I really want to sit around Saturday backing a gargantuan Phil Fulmer? This is all about having fun, people, and I’m getting back to fun.