NFL Picks – Week 5

Total (weeks 1-4) –  SU: 39-22-0 (63.93%) ATS: 32-27-2 (54.09% +2.3 Units)  O/U: 36-25-0 (59.02% +8.5 Units)

Week 1 – SU: 10-6-0 ATS: 8-8-0 O/U: 13-3-0

Week 2 – SU: 10-6-0 ATS: 10-6-0 O/U: 10-6-0

Week 3 – SU: 12-4-0 ATS: 9-6-1  O/U: 8-8-0

Week 4 – SU: 7-6-0 ATS: 5-7-1  O/U: 5-8-0

Minnesota at Green Bay

Prediction: Packers 26-20 (67.5%)

Pick: Vikings +9 (57.5%)

Total: Under 48

Chicago at Carolina

Prediction: Panthers 23-22 (54.9%)

Pick: Bears +2.5 (52.2%)

Total: Under 46

Houston at Dallas

Prediction: Cowboys 24-22 (55.0%)

Pick: Texans +6 (61.9%)

Total: Under 47

Arizona at Denver

Prediction: Broncos 26-20 (67.3%)

Pick: Cardinals +7.5 (53.5%)

Total: Under 49

Buffalo at Detroit

Prediction: Lions 24-20 (62.9%)

Pick: Bills +7 (56.7%)

Total: Under 44.5 

Baltimore at Indianapolis

Prediction: Colts 23-22 (52.4%)

 Pick: Ravens +3.5 (57.6%)

Total: Under 48 

Pittsburgh at Jacksonville

Prediction: Steelers 24-19 (64.7%)

Pick: Jaguars +7 (54.9%)

Total: Under 46.5

Tampa Bay at New Orleans

Prediction: Saints 27-20 (68.5%)

Pick: Buccaneers +10 (59.2%)

Total: Under 48.5 

Atlanta at NY Giants

Prediction: Giants 25-23 (54.7%)

Pick: Falcons +4 (56.6%)

Total: Under 50.5

St. Louis at Philadelphia

Prediction: Eagles 25-20 (64.0%)

Pick: Rams +7 (55.7%)

Total: Under 48

NY Jets at San Diego

Prediction: Chargers 23-19 (61.8%)

Pick: Jets +7 (57.9%)

Total: Under 44

Kansas City at San Francisco

Prediction: 49ers 24-17 (68.3%)

Pick: 49ers -6.5 (50.5%)

Total: Under 45

Cleveland at Tennessee 

Prediction: Titans 23-21 (56.8%)

Pick: Browns +1.5 (52.5%)

Total: Under 44

Cincinnati at New England

Prediction: Patriots 26-22 (61.4%)

Pick: Patriots -2.5 (54.4%)

Total: Over 46

Seattle at Washington

Prediction: Seahawks 23-21 (57.7%) 

Pick: Redskins +7.5 (76.7%)

Total: Under 46

 

My NCAA Top 25

2014 NCAA Football Standings

Updated September 29, 2014

 
Teams records
1 AUBURN 4-0
2 ALABAMA 4-0
3 TEXAS A&M 5-0
4 OKLAHOMA 4-0
5 UCLA 4-0
6 OLE MISS 4-0
7 MISS STATE 4-0
8 BYU 4-0
9 NEBRASKA 5-0
10 OREGON 4-0
11 TCU 3-0
12 FLORIDA STATE 4-0
13 NOTRE DAME 4-0
14 LSU 4-1
15 BAYLOR 4-0
16 LOUISVILLE 4-1
17 ARKANSAS 3-2
18 KANSAS STATE 3-1
19 STANFORD 3-1
20 OKLAHOMA STATE 3-1
21 ARIZONA 4-0
22 GEORGIA 3-1
23 UTAH 3-1
24 CALIFORNIA 3-1
25 WASHINGTON 4-1

Full Rankings 

Language Log on “specificity” and “sensitivity”

statsinthewild:

I can finally say that I no longer need to look these two thins up. #progress #experience

Originally posted on God plays dice:

Language Log on “specificity” and “sensitivity” as (poorly chosen words for) properties of medical tests. Mark Liberman asks: why not just call them true positive rate and true negative rate. With the classic “what’s the probability that you got the disease, given that you tested positive?” problem thrown in; you’ve seen this if you ever learned Bayes’ theorem.

View original

NFL Picks – Week 4

Total (weeks 1-3) –  SU: 39-22-0 ATS: 32-27-2  O/U: 36-25-0 

Week 1 – SU: 10-6-0 ATS: 8-8-0 O/U: 13-3-0

Week 2 – SU: 10-6-0 ATS: 10-6-0 O/U: 10-6-0

Week 3 – SU: 12-4-0 ATS: 9-6-1  O/U: 8-8-0

Week 4 – SU: 7-6-0 ATS: 5-7-1  O/U: 5-8-0

NY Giants at Washington

Prediction: Washington Football Team 23-21 (55.4%)

Pick: Giants +3.5 (54.6%)

Total: Under 46.5

Carolina at Baltimore

Prediction: Ravens 23-20 (56.4%)

Pick: Panthers +3 (51.2%)

Total: Over 41

Green Bay at Chicago

Prediction: Bears 23-22 (52.9%)

Pick: Bears +1.5 (57.1%)

Total: Under 50.5

Buffalo at Houston

Prediction: Texans 24-19 (63.3%)

Pick: Texans -3 (55.0%)

Total: Over 41.5

Tennessee at Indianapolis

Prediction: Colts 24-21 (57.7%)

Pick: Titans +7.5 (63.4%)

Total: Under 46

Atlanta at Minnesota

Prediction: Falcons 23-22 (53.3%)

 Pick: Vikings +3 (55.2%)

Total: Under 47

Detroit at NY Jets

Prediction: Lions 22-21 (52.5%)

Pick: Jets +2 (53.15%)

Total: Under 45

Miami at Oakland

Prediction: Dolphins 21-20 (52.6%)

Pick: Raiders +4 (58.8%)

Total: Over 40.5

Tampa Bay at Pittsburgh

Prediction: Steelers 23-18 (52.9%)

Pick: Buccaneers +7.5 (57.2%)

Total: Under 45.5

Jacksonville at San Diego

Prediction: Chargers 26-17 (73.8%)

Pick: Jaguars +13.5 (62.9%)

Total: Under 45

Philadelphia at San Francisco

Prediction: 49ers 25-20 (63.3%)

Pick: Eagles +5 (50.7%) PUSH

Total: Under 51 

New Orleans at Dallas

Prediction: Saints 26-25 (51.5%)

Pick: Cowboys +3 (57.0%)

Total: Under 53.5

New England at Kansas City

Prediction: Patriots 25-21 (60.5%)

Pick: Patriots -3.5 (50.6%)

Total: Over 46

How the 538 model works

Originally posted on God plays dice:

Here’s a explanation by Nate Silver of how his Senate prediction model works. 10,000 words, and denser than the typical FiveThirtyEight post, but it’s food for thought if you’ve been curious about what’s going on under the hood of FiveThirtyEight’s flagship product.

Make sure to click through to the footnotes – lots of links to subsidiary analyses from the past that explicate some of the interesting tidbits Silver and co. have built up over time.

View original

Which sport?

I went the the Cubs game today against the Dodgers.  I got to see Clayton Kershaw pitch (Awesome) and watch the Cubs get destroyed (Expected).  It was 14-3 going into the 9th inning and some people were joking that a touchdown would get the Cubs right back in the game.  (The Cubs “rallied” for two in the 9th to ultimately lose 14-5.)

But, this got me thinking.  14-3 does look like a football score, but 14-5 is clearly not.  On the train back to school, I came up with this question: If all you knew about a game was the score, which sport would you guess generated that score?

To answer this question, I scraped the scores of (almost) all of the baseball and hockey games from 2000-2014, and I found a table of all NFL scores all-time. (I haven’t exhaustively scraped every single NHL and MLB game ever because 1) I don’t think the results will change much and 2) what I did was good enough for Friday night.) I then counted how often a specific score occurred each of the three sports (I did consider basketball, but it’s not really that interesting here) and converted that count to a percentage.  Then for each score, I assigned the sport with the highest percentage of that score.  The graphic below shows the results of this.  If you click on the image, you can zoom in on it for more detail.  whichSportScores

 

Some interesting things:

  • If the winner scores 4 or fewer goals, those games belong to hockey.
  • Additionally hockey gets 5-1, 5-2, 5-3, and 5-5.  5-0 and 5-4 are baseball scores though.
  • If the winner scores between 6 and 9 runs, those games all belong to baseball.  (With 6-6 and 7-7 ties belonging to football.
  • Every shut out where the winner scores 12 or more (12-0, 13-0, etc.) belongs to football.  Football also gets 10-0, but 11-0 belongs to baseball.
  • In fact, any game where the winner scores 11 belongs to baseball.
  • 18-8 is the lowest score combination to not be in my data set.  This means that this score has never occurred in the NFL and has (likely) not occurred in the NHL or MLB since 2000.
  • If the losing team score 10 points, the game belongs to football EXCEPT if the score is 11-10.
  • If the losing team scores 12 or more points, it’s always football.

Cheers!

Confusing Scotland pie chart

 

In the graphic below (from this article), the orange represent Scotland and the gray represents the United Kingdom.  So in the result bar this make sense: 55% voted for Scotland to be independent and 45% voted to stay in the United Kingdom.

For population and GDP, I think this also makes sense.  For instance, the United Kingdom has a population of 64.1 million and Scotland has a mere 5.3 million.

The map of the UK get’s a little bit more confusing.  The way this is colored Scotland is clearly the orange here, but it looks like to United Kingdom is only the gray part when, in fact, the UK is both the gray and the orange.  I would have made Scotland orange and gray striped to indicate that Scotland is both Scotland AND part of the UK.

Finally, the pie chart.  This is a puzzle to me.  I’m not really sure what they are trying to say here.  Maybe I’m missing something, but it looks like what this is saying is that Scotland, who has a small fraction of the total population of the United Kingdom has a full 97% of the eligible voters and the United Kingdom has only 3%. That’s clearly not true.  So what is that pie chart trying to say exactly?  I really have no idea.  (Please let me know if you know what it is trying to say.  I really have no idea.)

But no matter what the answer is, can’t we all agree we shouldn’t be using pie charts at all?

Cheers.

Screen Shot 2014-09-19 at 7.48.37 AM