Survivor: Backing the Cowboys

As for last week's near-fiasco, I can't describe it any more clearly than I did in East Coast Offense:

My near-death experience backing the Bears in Week 1 brings to mind something I read (in a paper by Jennifer Jay) about the life of another artist (of similar caliber to me), Fyodor Mikhailovich Dostoevsky:

[He was] led onto a square and lined up on a gallows. The men were sentenced to be shot; they were given a cross to kiss, the chance to confess to a priest, and then were dressed in peasant shirts and hoods for the execution. The first three men in line were led to some stakes and tied; the soldiers took aim, and held their positions. Soon Dostoevsky heard a drum roll and realized that he, the sixth in line, and his fellow prisoners, were to be saved...

This scene on the square, the staged execution, and the last minute pardon were to have a lasting effect on many from the group. Two even went permanently insane from the psychological trauma experienced as they faced certain death. While the experience was definitely traumatic for Dostoevsky, he internalized the situation much differently. This scene was to serve as a reaffirmation and strengthening agent of his religious beliefs. This close scrape with death gave Dostoevsky a new appreciation of life.

The major difference between Dostevsky's situation and mine is that his was less traumatic. While I too was pardoned at the last minute by the inexplicable ruling that Calvin Johnson's clear touchdown was an incomplete pass, at that point, I had nearly made peace with death. The more torturous part was the Bears, down one, in the fourth quarter getting stuffed four times from the one yard-line. It was the Bears scoring the TD and missing the two-point conversion, then inexplicably playing prevent defense against Shaun Hill after not allowing Hill to get even a first down for the five previous series. It was the turnovers, the penalties, seeing Cutler run for his life on every play behind that abysmal offensive line.

It's a terrible thing for the average survivor player to lose his entry in Week 1, but quite another for me to lose five of seven, and more importantly, take down all those who trusted in my advice on the site. At least for the regular player, you curse fate and are done with it, but I still have to muster the enthusiasm for the column for 16 more weeks.

Surviving Week 2

For this week, I already detailed why I'm going to avoid the Packers even though I think they easily have the best chance to win this week:

Keep in mind that 58 percent of those playing the Yahoo! game are on them. Let's call it 60 percent to keep the math simple. Meanwhile, only seven percent of Yahoo! players are on the Cowboys at home against the Bears. (Let's call it 10 percent for the sake of simplicity). Let's assume you were in a 10-person survivor pool for $100 each. That would mean you started off with $100 in "pot equity." Let's say six people took the Packers, three took other teams and you were the sole backer of the Cowboys (as per the Yahoo percentages.)

If the Cowboys win, and the Packers lose, you'd be one of the final four. (Probably final three as we'll stipulate that one of the other three players loses, too). So your equity would be $333. But if you took the Packers, and they won, but the Cowboys lost, then you'd be one of eight people left — (the Cowboys owner and again, we'll stipulate that one of the other three owners loses, too). One eighth of $1000 is $125. So you can see that the payout for taking the Cowboys is far better than the one for taking the Packers. But is it better enough to justify the higher risk?

Well, let's look at what Vegas (or virtual "Vegas," probably Costa Rica) has to say:

The Packers/Bills are -$675/+$475, and the Bears/Cowboys are -$380/+$310. That means you'd have to risk $675 to win $100 on the Packers straight up, and if you bet $100 on the Bills straight up, you'd win $475. If we stipulate that the "true" line is halfway between, i.e., $575 to $100 in the Packers/Bills and $345 to $100 in the Cowboys/Bears, we can figure out the odds. 5.75 to 1 is the same as 5.75 out of 6.75, which is 85 percent. For the Cowboys, it's 3.45 out of 4.45 or 77.5 percent. So is that difference small enough to take Dallas and go for the bigger payout?

There are four possible outcomes here:

So the task is to add up how much pot equity one can expect under the various scenarios by taking each team. If you take the Packers, we already showed you have $125 in equity if they win and Dallas loses (Scenario 1 in this example) because eight people would be left. That happens 19 percent of the time, so you'd have $125 * 19/100 = $24. Under the second scenario, let's say there are nine people left, so that's $111 of equity, 66 percent of the time = $73. Under the third and fourth scenarios, you'd have zero, because Green Bay loses. So the total expected equity is $24 plus $73 = $97. In other words, picking Green Bay if you knew six of 10 people took them would give you a negative expected return! Now, let's look at Dallas. Under the first scenario, you get zero because Dallas loses. Under the second scenario, you'd get $111 66 percent of the time for $73 — it's the same as picking the Packers. Under scenario 3, you get zero, obviously. But under scenario 4, six Packers entrants are gone, and so is one other, so you're down to the final three. Your equity is thus $333 12 percent of the time or $40. And $40 plus $73 = $113! Your expected equity goes up by $13 by being the sole backer of the Cowboys. That's why it's very important to discount the team that others are picking!

There are two potential problems with the above reasoning: (1) That Vegas' lines actually are designed to split the public betting and not necessarily a true barometer of a team's chances to win; and (2) That just because 60 percent of the Yahoo! pool is picking the Packers doesn't mean your pool will have the same distribution. As for (1), that's true, but if Vegas is way off, and sharps have a much better idea of the real odds, it will lose vast sums of money. So Vegas has as big incentive to be pretty good at setting the lines initially at least — certainly better than most people just eyeballing it and guessing. And (2) if there are enough people playing the Yahoo! game (and I imagine there are at least 10s if not 100s of thousands), you can be pretty sure within some small margin of error that it's a fairly representive sample of the whole. Whether that applies in your 10 person pool based in Minneapolis where everyone hates the Packers, is hard to say, but it's useful information generally, and it could even turn out that a higher percentage take the Packers.

So I'm going to have to go with Dallas. I agree the Packers are the better bet, but I can't turn down the payout. It doesn't matter whether you lose in Week 2 or Week 7 — you need to play to win the whole thing, and the way to do that is to steer clear of the herd.

Here are my full slate of picks (with Yahoo! percentages in parentheses):

1. Dallas Cowboys (7%)

I trust Wade Phillips as far as I can throw him, and unless I double-up on the current roid regimen I'm on, that Rex Ryan-esque carcass isn't going far. Still, Dallas is coming off a tough loss, at home and facing a Bears team with an average defense and a terrible offensive line. I give the Cowboys an 80 percent chance to win this game. (I know Vegas says 77.5, and I used it in my math to be conservative in making my case, but I think it's a bit higher).

2. San Diego Chargers (3%)

I trust Norv Turner about as much as I do Phillips, but the Chargers have a home game against the Jaguars, a team that got lit up by Kyle Orton last week. San Diego has struggled to get in sync early the last few seasons, but Philip Rivers was in top form last week, despite a hostile crowd and bad conditions. I give the Chargers a 77 percent chance to win this game.

3. Green Bay Packers (58%)

The Packers are easily the best game on the board as Buffalo simply lacks the firepower to keep up, especially on the road. The one thing the Bills have going for them is their strong pass defense - maybe that allows them to hang around for a while. But given how many people are on Green Bay, they're a bad choice here. I give the Packers an 87 percent chance to win this game.

4. Atlanta Falcons (5%)

The Falcons should be hungry off an opening week loss in Pittsburgh, and I expect them to move the ball effectively against Arizona at home. Derek Anderson would have to play like he did in 2007 to keep, something that's possible, but not that likely. I give the Falcons a 74 percent chance to win this game.

Notable Omissions

New Orleans Saints - This is a bad setup for the defending champs - Monday night road game off a 10-day layoff against a tough defense coming off an embarrassing upset. Stay away.

Minnesota Vikings - The Vikings were one of the best teams in the league last year, but I'm not sure this is the same squad in 2010. Favre's another year older and coming off a recent surgery, and his top receiver Sidney Rice is out. Maybe they'll revert to their previous form, but I'm not going to chance it against a non-doormat like the Dolphins.

Indianapolis Colts - One just assumes the Colts can't start out 0-2 - after all, they always seem to be 7-0 at some point in October, but every year is different, and the Giants are a dangerous and balanced team and one that could upset them even in Indy.

Philadelphia Eagles - I don't like backing a road team with its backup quarterback, even if Michael Vick did play exceptionally well for a half. The Lions defense looked better last week, and Shaun Hill should be better off a full week of practice.

Tennessee Titans - The Steelers are very tough against the run, and that means this will likely be a violent, close-to-the-vest, smash-mouth game that either team can win.