A little basic math can help you not just survive, but win.
By Chris Liss
This article is part of our fantasy football help series.
Survivor is arguably the most compelling NFL-derived game of all – even better than picking games against the spread, and possibly fantasy football itself. The rules are simple:
Pick one team to win outright each week. If you are correct, you move on to the next week. If you're wrong – and the team loses – you're out. The last man standing is the winner. But there's one catch: You can't use the same team twice.
The beauty of survivor is that unlike with handicapping games against the spread or playing fantasy football, your interests are 100 percent aligned with the teams' (except maybe in Week 17, but that's a problem you hope to have because it means you made it that far.) There's no getting screwed when the backup running back scores the touchdown, or when the team with the big lead takes its foot off the gas in the second half and allows a backdoor cover. You just need them to win, and you don't care how they do it or by what margin.
The other compelling aspect of survivor is that, depending on the size of the pool, you can win quite a bit of money. In a $50 buy-in pool with 1,000 people, you're talking about $50,000 to the winner. Even in a modest $20 buy-in with 100 people, there's a $2,000 pot. And it requires little work – just clicking on one team each week.
But the ultra-simple setup belies a surprising strategic complexity. At first, it seems the best way to win is simply picking the biggest favorite each week. For Week 1, that would be the Seahawks at home against the Dolphins. It's actually not a bad strategy, but by doing that you will use up the league's elite teams (usually the biggest favorites) quickly.
Realizing this, some will try to preserve the best teams by looking at the NFL schedule and working backward from Week 17. The idea is to find reasonably big favorites each week and plan it out ahead of time. If the only good game on paper in Week 11 is the Seahawks over the Eagles, then you wouldn't use the Seahawks in Week 1 because you'd want to save them for that difficult week.
This is a terrible strategy.
We don't know how the NFL landscape will look 10 weeks hence. Often, a team picked to finish last will contend for a division title, and a team picked by many to win will be a doormat. What if Russell Wilson gets hurt? What if the Eagles are much better than we think? Moreover, even if Wilson is fine, and the Seahawks are 87/13 favorites over the Eagles, maybe there will be another lopsided matchup due to injuries or a team emergence about which we have no idea. There's so much uncertainty even for the coming week in the NFL that looking multiple weeks ahead is a mistake.
So, picking the biggest favorite each week is the foundation of sound survivor strategy, but it's incomplete. That's because it leaves out a key variable: ownership percentage.
Ownership Percentage is Key
While the short-term goal in survivor is to survive, the only way to actually win is by surviving while everyone else perishes. It is not enough to make it through Week 12 if 80 percent of your pool is standing right there with you. Your $10 entry in a 100-person pool will have an "equity value" of only $12.50 ($1,000/80) if 79 others are still alive, too.
So, while the Vegas odds can tell you how likely it is the team you're considering will win (the Seahawks are roughly 78.5 percent favorites), they cannot tell you how much that win is worth to you.
Wait, aren't all wins worth the same because they mean you survived to the subsequent week? No, they're not, because making it to Week 2 with 50 people left is worth a lot more than making it to Week 2 with 80 people left.
Taking the $10 buy-in, 100-person pool as our example (entry fees and number of entrants vary widely), if half were to lose in Week 1, then your surviving entry would be worth $20 in Week 2, i.e., you'd be one of 50 people still vying for the $1,000 prize, and $1,000/50 = $20. That's much better than the 80-person scenario worth only $12.50. Now, you don't get to cash out with 50 people left (unless you all agreed to it, which would never happen), so that $20 is not liquid, but it is the amount of equity you have in the pool.
If the Vegas odds determine the likelihood of winning, the ownership percentages determine what our payout is. For example, if you knew 50 percent of people would be on the Seahawks to beat the Dolphins, but only 20 percent on the Chiefs (77.8 percent likely to win vs. the Chargers, let's say), you also know that by taking the Chiefs, you'd be in a position to survive with only 50 people left should the Seahawks lose**. By taking the Seahawks, your upside would be surviving with 79 other people if the Chiefs lost**.
But how could you know in advance half the people in your pool would take the Seahawks? What if you guess wrong, take the Chiefs, and actually more people are foolishly saving the Seahawks for Week 11, and in fact 50 percent are on the Chiefs? There is no way to know for sure, but we can get a pretty good idea using "polling data."
I use Officefootballpools.com (you could also get numbers from ESPN or Yahoo, but they have more free leagues, which provide noisy data) and look at ownership percentages across all its pools (it has tens or hundreds of thousands, so the sample is quite large). If I see the Seahawks are 49 percent used and the Chiefs 21 percent, I'll have a pretty good idea most people in my pool are leaning that way too. It's not perfect, but especially early in the year when everyone has all teams available, it's an excellent way to estimate ownership percentages.
Why Skeptics of Ownership Percentage are Wrong
Some are skeptical of this reasoning: "You're really going to take worse Team X over better Team Y in the hopes of more equity the following week? No way. Name of the game is to survive, and I'm not taking that risk."
The objection is absolutely correct if your goal were simply to survive as long as possible. But, as mentioned, surviving is not the way to win; you must survive while everyone else does not. And if my Seahawks/Chiefs example was unpersuasive, let me use a more extreme one that makes it obvious:
Let's assume in your 100-person pool, you knew 99 were taking the Seahawks. And let's (for the sake of the example) assume for whatever reason you only could take them or the biggest underdog on the board, the Dolphins, who play the Seahawks. (You would never actually do this, but it'll drive home the point.) If you take Seattle you have a 100 percent chance to make it to Week 2 (win or lose, everyone else is on them too). If you take Miami you have a 21.5 percent chance. Which would you take?
The answer should be obvious: the Dolphins, of course! Because with the Seahawks you're alive, but you still have the same amount of equity you started with, $10. But if the Dolphins win, you win the whole pool then and there, i.e., your equity is $1,000! By taking the Dolphins you have a 21.5 percent chance to win $1,000 and a 78.5 percent chance to lose $10. By taking the Seahawks you have a one percent chance to win $1,000 (you're still one of 100 people trying to get there) and a 99 percent chance to lose $10 eventually. The Dolphins – the biggest underdog on the board – are the better choice by a factor of 21.5! Even so, it remains the case that if your goal were merely surviving as long as possible (rather than winning the pool), the Seahawks give you a 100 percent chance to see Week 2, and the Dolphins merely 21.5. But you can see how worthless that is.
So the ownership percentage variable is a significant one, and it must be incorporated into our Survivor strategy.
Simple Math = Significant Advantage
The math involved in choosing between two teams with different ownership percentages and odds to win gets a little more complex when it's not a simple 99:1 scenario outlined above – but it's still not multivariable calculus. When comparing two teams, calculate the likelihood Team X wins while Team Y loses vs. Team Y wins/Team X loses. (If they both win or both lose, it doesn't matter which one you took.) Then compare the ratio of the likelihood of the former scenario to the latter. That's your "risk ratio."
Then you use ownership percentages to figure out how many people would remain in your pool in each scenario, and use that to calculate the prospective equity in each case and compare them to generate your "reward ratio."
Let's use the Seahawks and Chiefs as our example:
SEA Win % - 78.5
KC Win % - 77.8
SEA W & KC L % (.785 x .222) - 17.427
KC W & SEA L % (.778 x .215) - 16.727
Risk Ratio (17.427/16.727) - 1.04
SEA W & KC L is 4 percent more likely
SEA Own* % - 50
KC Own* % - 20
SEA W & KC L Equity** ($1,000/70) - $14.29
KC W & SEA L Equity** ($1,000/40) - $25
Reward Ratio (25/14.29) - 1.75
KC W & SEA L is a 75 percent better payout
* These are not real ownership numbers (not available). They are made up for the purposes of the example.
** The Equity in the event of a Seahawks win/Chiefs loss is calculated as follows: If 20 people are on the Chiefs, and they lose, that means 80 people will remain. But remember, if 50 are on the Seahawks and 20 on the Chiefs, that means there are 30 more people on other teams, some of which will also lose. (You can calculate this number more precisely when the polling data comes in, and you match the percentage owned for those other teams with their Vegas odds. For the sake of simplicity, I assumed 10 of those 30 will lose.) If we add 10 more losers on other teams, that means 70 remain. So in the Seahawks win/Chiefs lose scenario, those who took the Seahawks have $1,000 (total money in the pool) divided by 70, which equals $14.29. Doing the same thing for the Chiefs win/Seahawks lose scenario, there are 40 people left, and $1,000/40 equals $25.
As you can see from the chart, the Seahawks win/Chiefs lose scenario is 4 percent more likely to obtain than the reverse. But the payout is 75 percent better should Chiefs win/Seahawks lose scenario come to pass. Under this example, it's clear – assuming you agree with Vegas' assessments of the teams' chances – the Chiefs would be the better Survivor pick for Week 1. (Again, this is based on hypothetical ownership numbers and not intended as actual advice for this season.)
The Bottom Line
Survivor is a simple game that requires little maintenance, offers the possibility of a big payout and aligns players with the goals of the team for which they're rooting. And its deceptive strategic complexity allows those who grasp it to have a significant edge.
That said, I don't want to undersell its unique capacity to inflict misery – only a small number of people win every year (if the pool is large, the last few survivors often split the pot), NFL games are often decided on the most arbitrary plays (bad calls, fumbles, the indecipherable catch rule) and being forced to pick among bad choices, if you get far enough, can drive a person insane.
But there hasn't been a year since 1999 when I didn't have at least a few survivor entries.