/* ]]> */

Why the Demographics of Your Pool Matter

 neglected teams, Office pool, Office pool strategy, Weekly Payout, WPM  Comments Off on Why the Demographics of Your Pool Matter
Nov 112009
 

The only way the demographics of your pool would not matter would be if all other participants picked teams at random. If that were the case, you could just pick all the favorites all the time. You’d win all the season-ending payouts and more weekly payouts than anyone else.

But they don’t pick at random, do they? They mostly pick favorites, sprinkling in a few upsets here and there. That creates an opportunity for you, at least in terms of winning more weekly payouts than anyone else.

By picking mostly favorites most of the time, the other participants are behaving rationally as individuals but irrationally as a group. Take a look at some of the weaker favorites — those whose Win Probabilities are less than 70% — and you’ll notice that frequently the percentage of participants who picked them was still quite high, maybe over 90%. If the Win Probabilities are accurate, those favorites might lose between one third and half the time but when they do almost nobody in your pool will reap the benefit. So why don’t you reap the benefit?

The trick is threefold: 1) Forecast the number of participants in your pool who will pick each team, 2) identify underbet teams relative to their win probabilities, and 3) simulate game/pool outcomes thousands of times to determine the optimal number of underbet teams to pick in your pool. Those form the basis of Weekly Payout Maximizer recommendations.

The right number changes from week to week, according to the win probabilities and fan pick forecasts. For Regular scoring pools it’s also different from the right number for Confidence pools.

 Posted by on November 11, 2009 at 10:05 am

NFL Contrarian: Weekly Payout Maximizing Picks for Week 9

 neglected teams, NFL, Office pool, Office pool strategy, Weekly Payout, Win probability, WPM  Comments Off on NFL Contrarian: Weekly Payout Maximizing Picks for Week 9
Nov 082009
 

For Confidence Pool participants with fewer than 22 participants, Tennessee is likely to be underbet relative to its WinProbability. You’re more likely to win this week’s payout with the picks below. If you’re participating in a pool with several fans of Tennessee, since Tennessee might not be underbet in your pool, you’ll find alternative picks are below.

Points Pick WinProb
13 Tennessee 39%
12 New Orleans 79%
11 Atlanta 76%
10 New England 75%
9 Seattle 75%
8 Green Bay 75%
7 Indianapolis 72%
6 Jacksonville 67%
5 New York G 62%
4 Pittsburgh 58%
3 Baltimore 58%
2 Chicago 58%
1 Philadelphia 58%

For pools with Tennessee fan bias and fewer than 22 participants:
Points Pick WinProb
13 Kansas City 33%
12 New Orleans 79%
11 Atlanta 76%
10 New England 75%
9 Seattle 75%
8 Green Bay 75%
7 Indianapolis 72%
6 New York G 62%
5 San Francisco 61%
4 Pittsburgh 58%
3 Baltimore 58%
2 Chicago 58%
1 Philadelphia 58%

For pools with 22 or more participants, you’re advised to pick 2 underbet teams.

Points Pick WinProb
13 Tennessee 39%
12 Kansas City 33%
11 New Orleans 79%
10 Atlanta 76%
9 New England 75%
8 Seattle 75%
7 Green Bay 75%
6 Indianapolis 72%
5 New York G 62%
4 Pittsburgh 58%
3 Baltimore 58%
2 Chicago 58%
1 Philadelphia 58%

For pools with 22 or more participants and several fans of Tennessee, use these picks:

Points Pick WinProb
13 Kansas City 33%
12 Dallas 42%
11 New Orleans 79%
10 Atlanta 76%
9 New England 75%
8 Seattle 75%
7 Green Bay 75%
6 Indianapolis 72%
5 New York G 62%
4 San Francisco 61%
3 Pittsburgh 58%
2 Baltimore 58%
1 Chicago 58%

For pools with 22 or more participants and several fans of Kansas City, use these picks:

Points Pick WinProb
13 Tennessee 39%
12 Dallas 42%
11 New Orleans 79%
10 Atlanta 76%
9 New England 75%
8 Seattle 75%
7 Green Bay 75%
6 Indianapolis 72%
5 Jacksonville 67%
4 New York G 62%
3 Pittsburgh 58%
2 Baltimore 58%
1 Chicago 58%

 Posted by on November 8, 2009 at 6:37 am

Helpful Information Hidden in Plain Sight: Other Participants’ Picks

 neglected teams, Office pool, Office pool strategy, Weekly Payout, Win probability, WPM  Comments Off on Helpful Information Hidden in Plain Sight: Other Participants’ Picks
Oct 312009
 

If you play office pools using some of the largest online pool websites — ESPN, Yahoo, UPICKEM, etc. — you can see the percentages of other users’ picks for every team. If you take a look at them, you’ll notice they’re remarkably similar. But how do you use this information? Here’s how: it is a decent initial approximation for the percentages of picks in your own local pool.

Let’s say 10% of all users on ESPN picked Cleveland, and you play in a pool with 15 participants. Using the ESPN distribution estimate, that means you can expect 1.5 of your participants picks Cleveland.

But it can’t be 1.5. It could be 1 or 2. It could also be 0, 3, 4, or 5, or more if you’re playing in a pool of Cleveland fans. Given the ESPN probability p, where p=10%, the formula for estimating the probability in an N-participant pool that a certain number i picked a team is:

Combin(N,i)*p^i*(1-p)^(Ni)

Enter that formula in Excel, substitute the is and you can see the probabilities that the number of your friends who picked Cleveland are:

0 21%
1 34%
2 27%
3 13%
4 4%
5 1%

If you’re not considering this for every game every week, you’re short-changing your office pool picks.

 Posted by on October 31, 2009 at 11:25 am

Office pool strategy: It depends on the payout frequency

 Office pool, Weekly Payout, Win probability, WPM  Comments Off on Office pool strategy: It depends on the payout frequency
Oct 132009
 

Office football pools have different payout frequencies, including:

Single (at season-end)
Weekly
Mixed (both weekly and season-end)

You should apply different strategies for each, as I explain below.

1.  Single payout:  Pick favorites consistently

A season payout pool is not a game of prediction, it is a game of chicken.  The first participant to flinch and try to predict an upset loses most of the time.  If you start guessing upsets you’ll likely fall behind and might never climb out of your deficit.  Worse yet, should you get lucky and guess a few upsets correctly, you might get overconfident and wrongly conclude that you really can pick upsets.  Don’t let this happen to you.  Let the others make the mistake.  Nobody can predict upsets consistently so don’t try.  You cannot afford a mistake.  In a season payout, until the final few weeks of the season there is no such thing as picking too many favorites!

2.  Weekly payout:   Pick neglected teams

Instead of rushing to predict upsets, pause for a moment and consider the other pool participants’ self interests. Most of them understand the dominance of picking favorites for the Single Payout game. But they don’t stop to consider whether that strategy is a good one for winning the most weekly payouts. (Hint: it could be if so many of them didn’t already use it.)

Everyone obsesses so much about predicting which teams will win, but winning the highest number of weekly payouts depends as much on selecting neglected teams, teams nobody (or few others) in your pool also pick.  Being neglected doesn’t improve a team’s chances of winning but it does improve your chance of coming out on top if they do.

 Weekly Payout Maximization (WPM) is a strategy I designed to exploit the advantage of selecting neglected teams in your pool.  It considers all the teams’ Win Probabilities, the number of participants in your pool, and the likely distribution of picks on each team.  Once I model the distribution of participant picks I run thousands of Monte Carlo simulations to identify how many underbets win the most weekly simulations.  In some scenarios you could win twice as often as the Average Joe.  WPM typically suggests between 1 and 4 upsets each week, depending on the number of participants in your pool and which teams are most neglected.

3.  Combination payout:  Start with the Single Payout strategy, but if you fall significantly behind shift to Weekly Payout

First, do a few calculations on your pool’s payouts and decide which is more significant – weekly payouts or year-end.  If the year-end pot is considerably less than 3x the size of a weekly payout, you should probably just focus on the weekly payouts and follow the Weekly Payout strategy.

Otherwise, start out with the Single Payout strategy.  If by 2/3 through the season  (around Thanksgiving) you conclude you’re hopelessly behind — maybe you missed a week or you didn’t begin Single Payout strategy until well into the season — start playing the Weekly Payout strategy.  (In the future I plan to add tools to identify when you’re behind enough to shift from picking only favorites, and when you’re hopelessly behind).

Your chance of winning the year-end payout might be slightly better than 1/N, where N is the number of players in the pool, but probably no better than 2/N.  Meanwhile, you have 17 chances to win the weekly pool in an NFL season.

 Posted by on October 13, 2009 at 6:05 pm
Oct 112009
 

Office football pools tend to payout in any of these schemes: 1) at season end, 2) weekly, and 3) mixed, a combination of season end and weekly payouts. Did you know the best strategy for each payout scheme is different? Below I discuss the best strategies for season end payouts and weekly payouts. I will discuss strategies for mixed payouts in a future post.

Strategies may vary by whether your picks are confidence weighted.

How to win the season ending prize: PointMax
Your best chance at winning the season ending prize is the Point Maximization (PointMax) strategy. It works against overconfident participants because they’ll ruin their season total by picking too many upsets. It works against the weak because they give up on PointMax if it’s not in first place after Week N. You just pick the favorites each week and never flinch.

Confidence scoring
In a pool with confidence scoring, sort the favorites by magnitude of Win Probability in descending order. I don’t charge for WinProbabilities or PointMax picks because they’re an open secret. I offer them as a convenience, but if for some reason you don’t like WinProbability, substitute Vegas point spreads, Sagarin Pure Points ratings (on USA Today), AccuScore predictions (Wall Street Journal), or Advanced NFL Stats Win Chance (on NYT’s Fifth Down column). The sorts won’t be too different most of the time. All of these are widely available free of charge.

Regular scoring
If your pool doesn’t have confidence scoring, just pick the favorites. Look at this site or USA Today to see their gambling lines page in the sports section. Now both the Wall Street Journal (AccuScore) and New York Times (Fifth Down, on Fridays) also publish such predictions.

How to win the most weekly payouts: Weekly Payout Maximizer
As good as it is over the season as a whole, PointMax is the worst strategy for winning weekly payouts! It’s too conservative. Weekly Payout Maximizer (WPM) will greatly increase your chances of winning weekly payouts.

Here’s the free advice that in the future will cost money:
For pools without confidence scoring, the list below is organized according to the number of participants in your pool. (Confidence pool players will have to wait for WPM to be tuned to allow for all those permutations).

In a regular scoring (no confidence weighting) pool, pick all the favorites except:

If your pool has 10 participants: pick against Indianapolis
If your pool has 11 to 30 participants: pick against Indianapolis, New York Jets
If your pool has 31 or more participants: pick against Indianapolis, New York Jets, Dallas, and Jacksonville

Below I discuss the philosophy and some simulation results behind WPM. My wife strongly encourages you to stop reading right here. However if you’re interested understanding details on the strategy behind WPM, read on.

Pool Entrants are Biased toward Favorites
Football pool contestants have learned not to pick too many upsets. As a result, each week they overbet some favorite teams. You have an excellent chance at winning more weekly payouts if you recognize such opportunities and make calculated picks when favorites are overbet.

Weekly Payout Maximizer: Identifying When to Bet Against Overbet Favorites
Given each week’s Win Probabilities, WPM identifies which, if any, underdogs to pick. WPM doesn’t predict which teams are likely to upset, it identifies whether any favorites are overbet.

The number of upsets it recommends will vary from week to week, and it will vary according to the number of participants in your pool. The number of upsets depends on the distribution of Win Probabilities, the distribution of office pool selections, and the number of participants in your office pool. If your pool has only 10 participants, for 2009 Week 5 it recommends picking only one upset – Indianapolis – but if your pool has 40 or more participants this week it recommends picking four: Indianapolis, NY Jets, Dallas, and Jacksonville.

This is not some tout’s “upset special;” it is an actuarially determined strategy for winning more weekly payouts than the other participants in your pool. The strategy depends not on expecting the Colts to lose, but on recognizing that if they do lose (they have roughly a 40% chance of losing this week) then your pool participants’ picks will fare very poorly. And if they win, so many of your pool participants will have picked them it won’t help anyone who picked them very much.

WPM is a high risk strategy. In order to reap the most weekly wins you need to risk having some bad weeks.

Do not use WPM if you hope to win the season total prize!

Monte Carlo Simulation Analysis: WPM Tailors Recommendations according to #Participants
16,000 Simulations Each for Various Pool Sizes, (Nonconfidence scoring)
Taking NFL 2009 Week 5 as an example of how this works, here are the WPM results from thousands of simulations using the Win Probabilities and the distributions of online office pool selections. The optimal number of upsets gets smaller depending on the number of contestants in your pool.

Scenarios: 10, 20, 30, 40, 50, 75 Participants
In each scenario there are 1 WPM, 1 Point Maximization Participants, and the rest are Average Joes. To avoid depending on unrealistic assumptions, I randomly apply the actual pick distributions of online office pool Participants to the Average Joes.

Percentage of Weekly Payouts Won by WPM vs. Average Joe and PointMax
10 Players
WPM recommendation: Pick Indianapolis to be upset
Percentage of weekly payouts won by
PointMax/Average Joe/WPM
9.3 / 9.4 / 16.2

Percentage of Weekly Payouts Won by WPM vs. Average Joe and PointMax
20 Participants
WPM recommendation: Pick Indianapolis and NYJ to be upset
Percentage of weekly payouts won by
PointMax/Average Joe/WPM
3.3 / 4.9 / 8.7

Percentage of Weekly Payouts Won by WPM vs. Average Joe and PointMax
30 Participants
WPM recommendation: Pick Indianapolis and NYJ to be upset
Percentage of weekly payouts won by
PointMax/Average Joe/WPM
1.6 / 3.3 / 6.7

Percentage of Weekly Payouts Won by WPM vs. Average Joe and PointMax
40 Participants
WPM recommendation: Pick Indianapolis, NYJ, Dallas and Jacksonville to be upset
Percentage of weekly payouts won by
PointMax/Average Joe/WPM
1.0 / 2.5 / 5.2

Percentage of Weekly Payouts Won by WPM vs. Average Joe and PointMax
50 Participants
WPM recommendation: Pick Indianapolis, NYJ, Dallas and Jacksonville to be upset
Percentage of weekly payouts won by
PointMax/Average Joe/WPM
0.7 / 2.0 / 4.4

Percentage of Weekly Payouts Won by WPM vs. Average Joe and PointMax
75 Participants
WPM recommendation: Pick Indianapolis, NYJ, Dallas and Jacksonville to be upset
Percentage of weekly payouts won by
PointMax/Average Joe/WPM
0.4 / 1.3 / 3.5

This means that in a very small pool – one with 10 participants, if every week is like Week 5 the participant applying WPM will win 16.2% of the weekly payouts. In a 20 participant pool, he’ll pick two upsets and win 8.7% of the weeks. In a 75 participant pool, he will pick 4 upsets and will still win 3.5% of the weeks. That compares to the Average Joe who wins 1/75 of the time. PWM will probably not win the Season Total, but he’ll win the most money over the course of the season.

Notice a few aspects about these simulation results:
1) The Average Joe’s chance remains consistently 1/#Players
2) PointMax’s chance is better than Average Joe only in very small pools
3) The more participants in the pool, the better the improvement for WPM over Average

PointMax’s consistency is your opportunity
Although it does a lousy job at winning any single week, PointMax still wins the most points over the 16,000 simulations for each scenario. But its consistency give risk to opportunity for you to win more weekly payouts by using WPM.

 Posted by on October 11, 2009 at 11:12 am