## Wednesday, January 30, 2013

### Super Bowl Squares - Every Score Pays Out

2/3/2013 Update: A more mobile friendly version of the chart: Squares Payouts.

My most recent two posts covered the concept of in-game Super Bowl squares probabilities, based on score, time, field position, and down.  I still hope to complete that project in time for the Super Bowl this Sunday, but there are only a few days left, and I saved the hard part (the 4th quarter score) for last.

As I indicated in my in-game squares post, there are plenty of sites out there which will give you the expected payout for each square, and many will give you the payouts for each quarter as well.  However, I haven't seen any site that calculates the probabilities for a somewhat less popular version of the squares pool: The "Every Score Pays" pool.

## Monday, January 28, 2013

### Superbowl Squares Probability - 2nd Quarter

 I can't believe this movie is 20 years old.
Update 2/3/2013: See this post for an expected payout chart by quarter, and for expected payouts for the "Every Score Pays" option.

This is a continuation of Saturday's post on in-game squares probability.  The original post covered in-game probability for the first quarter square, using last year's Giants-Patriots Superbowl as an illustrative example.

This post covers in-game probability for the second quarter square, where again I use last year's Superbowl as illustration, and validation, of the probabilities.  The model I have built calculates probability of each square paying out as a function of current score, possession, time remaining, yardline, and down.  It is built from actual game results and a lot of smoothing.  My original post (linked above) has additional details on the methodology.

## Saturday, January 26, 2013

### In Game Squares Probabilities

The purpose of this post is to lay out my initial attempt at building an in-game probability model for Superbowl Squares.  This post focuses on probabilities for the 1st quarter square, with the hope that I can extend the methodology to all four quarters.

### In Game Cover Probability

In November of last year, I sketched out a rough approach for building an in-game cover probability model for the NFL.  The basic idea was to extend the Advanced NFL Stats win probability model to calculate the in-game probability of a give team covering the point spread, as opposed to just winning outright.

I've made some progress, but I'm still far from the finish line.  Along the way, I've reported out on some of the baby steps I've made in working towards my final goal; examining how scoring margin distributions, expected points, and 4th and 1 conversion rates all vary by the point spread of the game.  My hope when beginning this project was to have a final product ready for the Superbowl, but with the game just a week away, that doesn't look too likely.

## Monday, January 21, 2013

### College Basketball Rankings - Updated Daily

 Are you suggesting that I, the president of Huxley College,go into a speakeasy without even giving me the address?
Betting market rankings for college basketball are now up and running.  As with my rankings for other sports, these are updated daily with the latest lines and game results.

My methodology page has more background, but the point of these rankings is to figure out what (and how) Vegas "thinks".  I use the point spreads and over/unders to reverse engineer an implied ranking, called Generic Points Favored (or GPF).  It's what you would expect a team to be favored by against an average team on a neutral court.  By combining the point spread with the total, I can decompose team strength into its offensive and defensive components (oGPF and dGPF).

I first published these rankings for the 2011-2012 college basketball season, not long after launching this blog.  At the time, I employed an Elo-style ranking system (see my original post for the details).  Since then, I've hit upon a better way to create these rankings.  As mentioned in this post on my MLB rankings, the new methodology weights recent games more favorably in a way that assumes that a team's ranking jiggles up or down randomly over time (a similar assumption is often used to model stock price movements).  Bottom line, it's less kludge-y, less volatile, and, most importantly, more accurate in predicting point spreads for upcoming games.

## Sunday, January 20, 2013

### Superbowl Line Predictions

As I did with the BCS Title Game, here are predictions for the point spread and over/under for each of the four possible Super Bowl matchups.  These are taken from my daily rankings page.  The whole point of my team ranking system is to predict point spreads for future matchups (i.e. what and how Vegas "thinks").  So this is just a high profile test of the rankings' accuracy.

As I noted earlier in the week, you really couldn't pick a better matchup from my rankings than Patriots-Niners.  It's the top two teams, they're evenly matched, and it's the number one offense facing off against the number one defense.

See below for the predicted totals and point spreads.  I'll check back over the next two weeks to see how the predictions fared.

Superbowl Matchups
Negative Point Spread Means NFC is Favored
MatchupProbabilityOddsLineTotal
NE vs. SF  47.2% +112 0.0 52.0
NE vs. ATL  28.6% +249 4.5 55.0
BAL vs. SF  15.1% +564 -6.0 44.5
BAL vs. ATL  9.1% +995 -1.5 47.5

## Monday, January 14, 2013

### Niners - Patriots Superbowl

Based on today's rankings (January 14, 2013), a Niners-Patriots Superbowl would be a toss up in the point spread.  And it would be the number 1 offense against the number 1 defense.

It's the Alabama-Oregon matchup we were denied.

## Saturday, January 12, 2013

### 4th and 1 Conversion Rate and Team Strength

This week at Advanced NFL Stats, Jack Moore published a post on passive 4th down decisions in last week's Wild Card games.  According to Jack (and his slick visualizations), coaches forfeited a total of 0.24 in win probability as a result of choosing field goals and punts on fourth down instead of an attempted conversion.

A common objection I see from fourth down skeptics has to do with the probability of a team converting on fourth down.  The skeptic will argue that the probability of conversion is based on league averages, but a true expected conversion rate should vary by team strength (good teams should have a higher probability of converting than poor teams).

While I don't doubt this is directionally true, my assumption has been that the conversion rate simply doesn't vary that much by team strength, and so most fourth down analyses are still valid when using league average rates.

I decided to see if my assumption was correct by looking at how converting a 4th and 1 varies by the point spread of the particular game; where the point spread is being used as an a priori proxy for relative team strength  If you know of a better one, please let me know (I would like to make some money).

## Saturday, January 5, 2013

### NFL Playoff Probabilities - Updated Daily

Earlier this week, I finalized my NFL regular season betting market rankings.  I have now created a new ranking table, with a focus on playoff probabilities (click here for the new table).

I will continue to track the Generic Points Favored metrics (GPF, oGPF, dGPF).  There are also four new columns, corresponding to each team's probability of advancing to the next playoff round.  The last column is just the probability of a Superbowl victory converted into Vegas-style odds (it's the "fair" payoff for a \$100 bet).

The probabilities are actually calculated explicitly (there are only 32 ways each conference can play out, so it's not necessary to do a simulation).

## Tuesday, January 1, 2013

### 2012 NFL Final Regular Season Rankings

Since September 29 of this year, I have posted my betting market rankings for the NFL on a daily basis.  The purpose of these rankings is to determine who the betting market thinks are the best and worst teams in the league (see the Methodology page for more background).  Publishing these rankings also afforded me the opportunity to indulge in one of my current interests: data visualization.  My goal was to produce as "data-rich" a ranking table as possible that was still relatively simple to understand.

Now that the regular season is over, I am finalizing these rankings, with plans to now focus on the playoffs.  But before I move on to simulations and probabilities of advancing to each round, I thought I'd take a look at how each team performed from a win-loss perspective, compared to the implied betting market expectations.