Saturday, December 16, 2017

Bowl Pool - First two game breakdown



OK.  It looks like we've got the kickoff to the first game of the 2017-2018 Bowl Season!!  That means that all entries are in, and it looks like this year we have 1 more entry than last year, for a total of 57.  Last year, I only paid out 2 spots, which I think is a little strange, since usually that cutoff is 50 entries for a 3rd place payout.  I'll remedy that this year and pay out 1st, 2nd and 3rd.

Full disclosure, I entered 3 times this year.  One is listed under my name, another is an entry I pulled from the internet.  I think the site was "College Football News".  They polled thier "experts" and made an ranking based on an average of about 6 people.  I like to put these kind of entries in the pool because I like it when they do bad.  Most years, they do.  Last year, however, the entry I made like that actually won, but I'm hoping that was an anomoly.  The third entry I made was something I found on-line where they somehow got a chicken to pick games (and points somehow), so I claim ownership of the "Clucko the Chicken" entry.  A few years ago I made a random pick generator entry, and it did horrible.  I'm expecting similar results for Clucko, but time will tell.

Fun facts:  Every team was picked by at least one person.  The team with the most points wagered on it is Auburn in the Peach Bowl.  I was a little surprised by this, since the 4 teams in the playoff could get an extra 30 points for that extra game.  It looks like the Final 4 team that has the most points wagered on it is Clemson (8th).

Given that Auburn fact, it should come as no surprise that the Peach Bowl is the game with the most points wagered on it.  Somewhat surprisingly to me, the Championship game is 2nd.  I'm used to seeing some more lopsided games near the top, especially considering that Florida Atlantic is a 3+ TD favorite in the Boca Raton Bowl.  That game comes in 3rd in points wagered.

People who aren't that familiar with the scoring for this pool, I will be sending out 2 sets of rankings.  The first will be only by points.  I kind of don't like this ranking, since you game it to look like you are doing well if you front load your points.  So for instance you wager 40 points on Game 1 and 39 on Game 2 and only win 1 of the 2, you will be near the top, but you probably aren't better off than someone that won both games and wagered 20 & 18 points.  The reason is because the front-loaded point sheet already lost 40 or 39 points, while the other sheet can still win get those points.  That's where the second ranking comes in.  The second ranking is in net points (points won minus points lost).  This ranking is more indicative of the odds of winning.  I'll send out both until 12/30.  At that point, I can calculate the probabilities of winning and who has been mathematically eliminated.  At one point, they used to have a game a day towards the end of the season.  I liked that, because it made it easier for me to send out elimination scenarios for each game and not have to worry about interaction effects, but that doesn't look like the trend in college football, so we are left with what we have now, which is 8 days of nothing between the last game and the Championship Game.  Last year, the Championship game didn't even determine anything.  That happens some times if someone gets a big enough lead.

For each game, I'll give some descriptive statistics about the game before the start (except for the Celebration Bowl, which is going on right now, and maybe the New Mexico Bowl, which starts soon).  For me, the most important of these is the median pick.  This is the pick where half the field took above that number, and the other half below.  I like this better than the average since there are several games where outlier picks skew the average.  Take, for instance, the aforementioned Boca Raton Bowl.  The Median pick in that game is Florida Atlantic for 40 points.  The average pick is Florida Atlantic for 22.9 points.  The reality is if you didn't put the maximum amount of points on Florida Atlantic and they win, you are losing ground to the field.  So, that person who picked Florida Atlantic for 30 points might think they are doing good, since they were above the average, but they will lose position in the standings because more than half the field won more points.

I'm going to try to get a graph to work.  I seem to remember having a problem doing this last year, but I'm undaunted.  I think it's interesting.

All right, let's get to the games.  I love the first game.  It may be my favoirite game of the entire season, especially since Western Michigan isn't in a Bowl game this year.  The Celebration Bowl is great because depending on where you looked, there may not have been a point spread listed.  I searched the internet for "experts" picks for the Bowl games, and virtually all of them left this game off.  Given that, it makes it interesting to see how people handled this game.  There just wasn't much to go on.  Usually the game is pretty good as well, which is a bonus.  If I'm right, I think this game is between HBCU teams, which are typically not on TV ever, so there's a certain underdog feel to this game, but lots of enthusiasm.   Anyway, here's the breakdown:


As you can see from the graph, most people just defaulted to 1 point, with a lot of people going hard for North Carolina A&T.

The point spread was NC A&T (-9.5), which would make it one of the 15 games where one team is favored by a TD or more.

Median: NC A&T for 5 points
Mean: NC A&T for 12.2 points
Picks for NC A&T: 43 (Average: 19 points)
Points for NC A&T: 816 (21st overall)
Picks for Grambling: 14 (Average: 8.5 points)
Points for Grambling:  119 (77th overall)
People with NC A&T as a top 10 pick: 12
People with Grambling as a Top 10 pick: 1 (Jay H. #2)

As you can see from the graph also, if Grambling wins, that will give Jay H. #2 a huge head start on the field.

This game ranks 34th in total points wagered, but so there's not a ton riding on it, but it's got a good spread of picks and is a great way to start the pool.

The second game of the morning is the New Orleans Bowl.


This is a little more lopsided than the Celebration Bowl.  This game has the most points wagered on it, but despite that, this game only ranks 20th in total points.  Here's the stats:

Vegas Line: Troy (-7)
Median: Troy for 18 points
Mean: Troy for 14.2 points
Picks for Troy: 43 (Average: 21.35 points)
Points for Troy: 918 (18th overall)
Picks for North Texas: 14 (Average: 7.79 points)
Points for North Texas:  109 (78th overall)
People with Troy as a top 10 pick: 10
People with North Texas as a Top 10 pick: 1 (Joel S. #3)

Joel must have a good feeling about the Mean Green (which is a great nickname, by the way).  He has 40 on North Texas, which is 26 points clear of the next highest points pick for them.  Good luck with that Joel.  Everyone should know by now that I really like picking like this.  It's risky, and I think the winner of this thing should be the person that takes the best risks.  The tough part is knowing what best risk is.  Sometimes it's risking a big loss to make 58 points on the field.

Good luck to everyone.  I'll be back in a little with the breakdown for the Cure Bowl (sponsored by Robert Smith, I think).

2 comments:

  1. Ooops. I mislabeled the graph for the Celebration Bowl. The teams should be opposite of where they are.

    ReplyDelete
  2. I think it’s the same graph as the second one too

    ReplyDelete