Oregon State travels to Provo this Saturday to take on the BYU Cougars. Our system favors BYU, who is two heart-breaking road losses from an undefeated record, and Vegas has them favored by 6. Despite the point spread and a key personnel loss for the Beavers, an overwhelming majority of the public has picked against the Cougars at home. Rolling with BYU this week is just one of several excellent chances to pick against the public in week 7.
West Virginia has put up some pretty impressive offensive numbers this season. That success could be misleading, though, since we rank the Mountaineers’ strength of schedule as just the 70th most difficult in the nation. In fact, their run should end this week at Texas, which currently ranks #4 nationally in our Overall Power Ratings.
The general public disagrees, which makes this one of several games this week where there’s an opportunity to both pick the favorite, and gain ground on the majority of your opponents.
Week 5 looks like a great opportunity to both play it safe and pick against the public, the holy grail of pick’em strategy. Before we get into the analysis, though, let’s see how our pick sets are doing so far.
Where We Stand After Week 4
Week 4 was sort of ho-hum for our pick sets. There were no huge wins to celebrate overall, but we also did well enough to keep our entries at or near the top of the charts nationally. Overall, our pick sets slipped a bit in Week 4, but all six pick sets currently rank in the top 10% of the country, and three of six are in the top 5%.
Here is where our Game Winner pick’em strategies currently stand on ESPN (which uses confidence points) heading into Week 5. The Conservative pick set should have you in the top 3 of a 100 person pick’em pool:
- Conservative: 98.6th percentile (-0.6 from last week)
- Aggressive: 94.1st percentile (-1.1)
- Very Aggressive: 91.8th percentile (+0.6)
Here is where we stand on Yahoo! with our Point Spread pick’em strategies. Some of our more aggressive pick risks didn’t pay off last week, but all strategies are still doing well, again highlighted by Conservative:
After a strong Week 2, we followed up with another solid performance in Week 3. On the game winner side, all three of our office pool pick sets are top 10% on ESPN. Our Conservative strategy is in the 99th percentile; Agressive picks are at the 95th percentile. After making some big bets on UNC and Arizona State that didn’t play out, our Very Agressive pick set dropped back a bit to the 91st percentile.
On the point spread pick’em side (where we use Yahoo!, since ESPN doesn’t report percentiles there), we are simply killing it so far. All three of our pick sets are in the top 1% of Yahoo!. Week 3 was just average for our Aggressive and Very Aggressive plays, which both hit around 50% of games, but that was enough to keep both within the top 250 entries in the nation.
However, last week was our Conservative strategy’s time to shine on the ATS side. That pick set went 13-6-1 on Yahoo!, and is now one of the top 50 entries in the nation through three weeks. Let’s hope we keep it up and get to the Week 4 advice…
Let the good times roll. After a decent Week 1, our college football pick’em strategies had a smoking Week 2, thanks to several calculated risks paying off.
The schedule this week gives us plenty of chances to make small gains in confidence point pools, but there are not as many smart opportunities for big risks in non-confidence pools. In other words, it’s a good week to chip away at your opponents, and play it relatively safe.
We got off to a great start in Week 1 of college football pick’ems. Two of our three highest-value game winner upset picks came through with wins (Ohio over Penn State, and Colorado State over Colorado), and our spread-based pick’em strategy would have led you to place near the very top of Yahoo!
Our Conservative ATS strategy — which takes into account stale lines and public pick biases as well as our projected cover odds — went 13-7 in Yahoo!’s Week 1 games, which snagged us a top-300 rank and means we’re beating 99% of all contestants. It is, of course, super early, but we’d rather be off to a great start than a poor one! Our more aggressive strategies — which, remember, are designed to be more “boom and bust” — are sitting in the 93rd percentile and the 97th percentile.
As if you didn’t know, college football is back, and the first regular season game is Thursday evening. That means it’s time to email your Week 1 pick’em picks to the guy in accounting who runs your office pool. You should probably send him that PayPal payment he’s been pestering you for, as well.
To help you get back in the swing of things, here’s a little assistance on those picks. We’ll be doing this every week, all season, just like we did last year. Our conservative picks ended up finishing in the 94th percentile in ESPN’s game winner based college pick’em last year, after finishing in the 97th percentile in 2010, so it’s a good bet that following this column won’t be a waste of your time.
This is the second chapter of our Pick’em Strategy Playbook, a series of blog posts on football pick ‘em contest strategy. In general, we are compiling this Playbook from past related posts on TeamRankings.com, as well as recent insights and studies we’ve done. We also post advice on NFL survivor pool strategy, as well as weekly contest analysis during football season.
Our first chapter of the Pick’em Strategy Playbook explored how skilled players can generate a substantial edge in pick’em style contests, and why pick’ems represent an attractive ROI opportunity. In this second chapter, we’ll review how we define success in pick’em contests, and illustrate a few of the more common traps that tend to snare a lot of misinformed pick’em players.
This is the first post in a preseason series on football pick ‘em contest strategy, which we are updating from original posts published last year. We also post advice on NFL survivor pool strategy, as well as weekly pick’em and survivor contest analysis during the NFL and college football seasons.
Case Study #1a: The Cumulative Impact Of Randomness
OK, here’s the situation. 49 monkeys and 1 IBM Watson supercomputer are competing in an NFL pick’em contest this year. They are picking against the spread.
Watson makes picks by analyzing 20 years of historical NFL data, current Vegas betting lines, and every bit of news that has ever been published about teams, players and coaches. As a result of all this awesome data and computing power, the ultra-nerds at IBM have gotten Watson to achieve a long term win rate of 55% against the spread, picking every single NFL game. That’s really impressive. In fact, IBM is just about to send Watson to Vegas to open an account at Cantor Gaming and go to town.
The monkeys, on the other hand, pick their winners by throwing banana peels at one of two targets, each representing an opposing team. But first, they each get to drink four quarts of Michelob Dark. So it’s a safe assumption that these inebriated primates will be picking their winners at random.
You are offered a side bet on this competition. At even odds, you can bet $100 that Watson wins the pick’em contest, or that the field (i.e. one of the 49 monkeys) does. Which side of that bet do you take?
Now that the 2011-2012 college bowl season has come to a close, we wanted to recap how we did with our picks and advice for college football bowl pick’em contests.
Overall, we did very well relative to the public this year, although not quite as good as last year’s super performance. Still, we’ve received several emails and comments from customers who either won or scored in the money in their bowl pick’em contests, which is always good anecdotal evidence.
Based on final pre-kickoff predictions, our algorithmic game winner picks went 24-11 (68.6%) overall, their third best performance over the past seven bowl seasons.
Straight Up Winners With Confidence Points
Most bowl pick’em pools are based on picking game winners while also ordering your picks with a “confidence rating” from 35 (highest) to 1 (lowest). Get a pick right, and you get the associated confidence number as points, and the person with the most total points once all the games are done is the winner.
Note that I put “confidence rating” in quotes. In fact, determining optimal strategy for most confidence point based pools is a lot more complex than just actually ordering your picks based on the level of certainty you have in them winning. That’s where we come in, with data-driven strategies to help our users game the system, in effect, to increase their overall shot at winning a pool.