December 14, 2016 - by Tom Federico
Screenshot from our Bowl Pick’em Picks product
Giving yourself the best chance to win a bowl pick’em pool requires a thorough analysis of the risk vs. reward tradeoff of every possible pick you can make.
One key part of this process is identifying a “short list” of potential picks that meet the following two criteria:
Let us briefly explain what we mean, then we’ll get to the games.
1) They aren’t too risky
Most people overestimate the amount of risk you need to take to maximize your odds to win a bowl pick’em pool.
The spirit of making upset picks involves trying to get picks right that most of your opponents get wrong. That’s the only way to win a pool, and it’s a good goal.
However, there is nothing good or noble about making a risky upset pick if you can generate a similar amount of separation from your opponents by picking a less risky team — ideally, a team that’s favored to win.
Although it can be tempting to make a crazy long shot upset pick or two, those decisions often hurt your odds to win a bowl pick’em contest, especially smaller pools.
Which brings us to the next point.
2) Their chance to win is being underrated by your opponents.
Betting odds and good data-driven models can provide reliable estimates of a team’s win odds. For example, 3-point Vegas favorites in college football historically win about 60% of the time.
In bowl pick’em pools, though, a team’s pick popularity often diverges from its actual win odds.
When that happens, it means your opponents may be irrationally biased toward a particular team, or simply misinformed about its chances to win.
If a team’s pick popularity is lower than its actual odds to win, it’s being underrated. Picking that team could be a great opportunity to get an edge in your pool.
Below are three bowl matchups that meet the criteria listed above. None is super risky, and all three are being underrated by the general public.
Making picks like these, and winning them, should have a major positive impact on your odds of winning your bowl pick’em pool.
However, it’s critical to understand that whether you should make some of these picks, or all of them, or even none of them, depends on the characteristics of your pool.
We’ll explain more about that at the end.
(Also, there are other picks this year that offer edge-generating potential comparable to or even better than the three examples below, but we wanted to highlight some of the more well known teams.)
According to the betting markets, USC is one of the more likely teams to win its bowl game this year, with about 70% win odds as 6.5-point favorites over Penn State.
Yet less than 50% of pool entries nationally are picking USC to win.
Penn State beat Ohio State (though it was a close game at home). They won the Big 10 championship. They had a shot at the College Football Playoff. All these events are likely being highly valued in the public conscience.
However, our data-driven predictive ratings and Vegas both agree that USC is the better team.
The Trojans’ season resume includes a number of large-margin wins, including a 13-point victory on the road over Washington.
This is the most value you can currently find on a solid favorite this bowl season.
This matchup features several characteristics that often lead to irrational public bias.
An AP top-15 ranked team with a 10-2 record (West Virginia) is playing an unranked team that’s a “worse” 8-4 (Miami).
Unsurprisingly, West Virginia is getting the big public support, with over 75% of bowl pick’em entries currently taking the Mountaineers.
What the public is likely missing here is that several of West Virginia’s wins this season have come by very close margins: Kansas State by 1, Baylor by 3, BYU by 3, Texas by 4. One more untimely fumble, dropped pass, etc., and any of those games could have gone the other way. WVU also got blown out in losses to Oklahoma and Oklahoma State.
On the other hand, two of Miami’s four losses were very close — Florida State by 1, Notre Dame by 3 — while many of their wins have come by big margins.
In short, the win-loss records tell a misleading story here. If Miami were a slight underdog, in a lot of pools it would be worth the risk of taking the Hurricanes as an upset pick, due to their very low pick popularity.
The kicker is that betting odds currently favor the Hurricanes by 3 points, and our predictive ratings think they’re the better team as well.
Oklahoma is the favorite in this game, by 3.5 points according to recent lines. The Sooners are also playing strongly of late, while Auburn slumped toward the end of the season.
Still, the public appears to be seriously underestimating Auburn’s chances in this game. Less than 15% of bowl pool entries are picking the Tigers.
For that reason, bowl pick’em players looking for a high-leverage pick (and willing to take some risk to get it) should consider this one.
Along with Oklahoma having a better record and AP ranking, recency bias is probably influencing public perception here. Oklahoma enters bowl season on a 9-game winning streak, including an 18-point victory over ranked rival Oklahoma State in their last game.
Auburn, meanwhile, dropped its last two contests against FBS opponents, Alabama and Georgia. Not good, but not exactly horrible either. Both games were on the road, and Auburn actually outperformed Vegas expectations against Alabama, covering the point spread.
All told, our ratings only see Oklahoma as about 2 points better than Auburn.
It’s definitely a risk to pick Auburn. We expect them to lose. But they aren’t a crazy long shot (about 40% odds to win), and the reward if the Tigers pull off an upset would be big.
You’d score points that roughly 85% of your pool opponents didn’t get, and differentiating your score is the only way to win a pool.
There really is no simple answer to that question.
It all depends on the characteristics of your pool — factors such as the number of entries, the scoring system (e.g. whether it uses confidence points or not), and whether local or regional rooting bias in your pool will skew the pick popularity of some teams.
In most pools, it probably makes sense to stick with USC and Miami. But if you’re in a pool full of Miami grads, picking the Hurricanes may not offer nearly as much value as it would in other pools more representative of Miami’s national pick popularity.
Oklahoma vs. Auburn is less clear. Going with the Auburn upset is likely a better fit for larger pools, in which following a higher-risk/higher-reward type of pick strategy usually benefits you in the long run.
Still, it all depends on how you’re picking the other 41 bowl games — i.e., how many calculated gambles you may or may not be making with your other picks.
At the end of the day, with up to 42 matchups to consider, the math involved in optimizing picks for bowl pick’em pools is extremely complicated.
That’s why we developed our Bowl Pick’em Picks product. If you’re serious about winning, please check it out.
We run thousands of computer simulations of 2016-17 bowl pools with dynamics similar to yours, then use objective algorithms to identify a set of picks customized to give you the maximum edge against your opponents.
The whole process only takes a few minutes, and in many cases delivers a result that few humans could ever hope to reach, even after thousands of hours of research — that is, unless you happen to be a math savvy software engineer with a bunch of free time on your hands. 🙂
Most importantly, it works. Last year our customers took first place in bowl pools three times as often as expected.
Printed from TeamRankings.com - © 2005-2018 Team Rankings, LLC. All Rights Reserved.