Postscript to College Football Predictions 2022 Season

This article summarizes what did and didn’t work when predicting college football games for the 2022 season. Entered into this exercise with low expectations, due to coming off the 2021 season with a 53.4% success rate. Biggest questions are how to improve the accuracy of predictions for the 2023 season and ensure the postings are also entertaining to read.

2023 CFP National Championship Game Pick Recap

CFP National Championship Game – 1/1 correct due to Georgia dominating TCU in all phases of the game for sixty minutes

2022 All Picks Recap

Week 1 to Week 13 Top 5 games to watch – 48/65 correct

Week 1 to Week 13 Top 5 upsets that could happen – 26/65 correct

Top 10 conference championships – 6/10 correct

Bowl Games Part I – 8/11 correct

Bowl Games Part II – 4/10 correct

Part III Bowl Games – 6/12 correct

Part IV Bowl Games – 5/8 correct

CFP National Championship Game – 1/1 correct

Overall picks – 104/182 correct which is one standard deviation above the mean

College Football Predictions 2022

1. Top 5 Games to Watch

Statistical Analysis: 48/65 correct, 73.9% accuracy rate, one standard deviation above the mean, +7.7% improvement from the 2021 season

Highlights: Had perfect 5/5 predictions in Weeks 4, 6 and 12. Spread those correct picks across eleven different teams (Michigan, Clemson, Tennessee, USC, Kansas, NC State, Kansas State, Alabama, TCU, Georgia, Oklahoma) and five conferences (Big Ten, ACC, SEC, PAC-12, Big 12). This blogger should have bought a lottery ticket those weeks, but ran out of time.

Lowlights: Went 2/5 in Weeks 11 and 13. Hard to maintain the pace needed to approach 80.0% success rate for the entire 2022 season and had two bad weeks late in the season. In Week 11, picked Illinois to beat Purdue by 10 points and they lost by 7 points (31-24). Thought Oregon would take a step towards the CFP and beat Washington by 10 points at home, and they lost by 3 points (37-34). In Week 13, picked Ohio State to beat Michigan by 10 points and they lost by 22 points (45-23). Thought Clemson would sneak into the CFP yet again and beat South Carolina by 10 points at home, yet they laid an egg and lost by one point (31-30).

Comments: This section was the easiest to write on a weekly basis. Determined the five best games based on the Top 25 rankings and then inter-conference matchups. Several weeks this section could have listed up to ten games. Hard not to focus on the dominant teams (Georgia, Michigan, Ohio State, TCU) to simply pad the prediction statistics. Tried to use a variety of teams to ensure the weekly postings were as fresh as possible to read.

2. Top 5 Upsets that could happen

Statistical Analysis: 26/65 correct, 40.0% accuracy rate, one standard deviation below the mean, +6.2% improvement from 2021 season

Highlights: Went 4/5 in Week 5 thanks to UCLA, Wake Forest, Illinois and Kansas. Always nice to spread picks across four conferences (PAC-12, ACC, Big Ten and Big 12). Went 3/5 in Weeks 4, 7, 10 and 11.

Lowlights: Going 0/5 right out of the gate in Week 1. Didn’t help that East Carolina had two chances to upset NC State in the final three minutes of the game, to then miss an extra-point and a potential game winning field goal. Houston beat UTSA in three OTs thanks to having fresher legs to run in a two point conversion. That’s football these days.

Comments: This section was the most difficult to write on a weekly basis. Several weeks this section could have listed less than five games. As my prediction statistics started to crater, I looked for betting lines that were favoring an underdog that was ranked in the Top 25 and I would then pick the “better” team. This may be considered gaming the statistical system, but it opened my eyes to how betting lines can and do change on a daily basis.

3. Top 10 Conference Championships

Statistical Analysis: 6/10 correct, 60.0% accuracy rate, one standard deviation above the mean, -10.0% deterioration from 2021 season

Highlights: Predicted Georgia would beat LSU by 14+ points and they won by 20 points (50-30). Predicted Michigan would beat Purdue by 14+ points and they won by 21 points (43-22).

Lowlights: Predicted USC would win the PAC-12 Championship game by 3 points and they lost by 23 points to Utah (47-24). Picked Boise State to have a big game on their home blue turf in the MWC Championship game against Fresno State. Thought the Broncos would win by 8 points and they lost by 12 points (28-16).

Comments: 60% accuracy rate is nothing to sneeze at, but this was based on the twenty best teams in the country during Week #14. Teams trending up late in the season won their conferences and most talking heads could see the final scores easily.

4. Bowl Games Part I

Statistical Analysis: 8/11 correct, 72.7% accuracy rate, one standard deviation above the mean

Highlights: Picks started at 1/3 and then rebounded to 7/8 to equal 8/11. Helped that Oregon State was predicted to beat Florida by 7 points and they were impressive in winning by 27 points (30-3). Predicted Southern Mississippi would beat Rice by 10 points and RB Frank Gore Jr. helped lead a 14 point victory (38-24). Underdog Eastern Michigan was picked to beat San Jose State by 7 points and they won by 14 points (41-27).

Lowlights: Completely missed the Cincinnati vs. Louisville matchup, due to predicting the Bearcats to win by 6 points, and they were dominated by the Cardinals in a 17 point loss (24-7).

Comments: Tried to anticipate opt out players, transfer portal players and general team malaise in these early bowl matchups. Better to be lucky than good with a 72.7% accuracy rate.

5. Bowl Games Part II

Statistical Analysis: 4/10 correct, 40.0% accuracy rate, one standard deviation below the mean

Highlights: Picks started at 3/4 and then cratered to 1/6 to equal 4/10. Picked Air Force to beat Baylor by 3 points and they won by 15 points (30-15). Predicted Wake Forest would beat Missouri by 10+ points and they won by 10 points (27-17).

Lowlights: Forgot how good Western Kentucky’s offense really was, due to picking South Alabama to win by 8 points, and the Hilltopper’s won in a rout by 21 points (44-23). Picked Utah State to represent the MWC well and beat Memphis by 2 points, and the Tigers rolled to a 28 point victory (38-10).

Comments: Continued to track the opt out rule players and frankly my dartboard was just way off for these ten games. Kind of like a bad rebound relationship.

6. Bowl Games Part III

Statistical Analysis: 6/12 correct, 50.0% accuracy rate, right at the mean

Highlights: All of these games got much closer on the scoreboard and tighter on the playing field. Picked Washington to beat Texas by 4 points and the Huskies won by 7 points (27-20). Picked Minnesota to run past Syracuse by 9 points and they won by 8 points (28-20).

Lowlights: Picked Clemson to beat Tennessee by 3 points and they wore down badly in the second half to lose by 17 points (31-14). Dabo Swinney needs to reignite the Tigers in 2023.

Comments: Games were exciting to the final play and betting lines went out the window. This is why the college bowl season will never get old.

7. Bowl Games Part IV

Statistical Analysis: 5/8 correct, 62.5% accuracy rate, one standard deviation above the mean

Highlights: Picked Alabama to beat Kansas State by 10 points and they won by 25 points (45-20). Picked Mississippi State to beat Illinois by 4 points and they won by 9 points (19-10).

Lowlights: Missed badly on the first CFP semifinal game, due to picking Michigan to beat TCU by 3 points and the Wolverines lost by 6 points (51-45).

Comments: CFP semi-final games were incredibly fun to watch despite no defense being played by Michigan, TCU, Georgia and Ohio State. No Alabama players opted out of the Sugar Bowl and they dominated Kansas State.

7. CFP National Championship Game

Statistical Analysis: 1/1 correct, 100.0% accuracy rate, three standard deviations above the mean, +100.0% improvement from 2021 season

Highlights: Wrote a thorough analysis that highlighted offense, defense, special teams, coaching and intangibles for both teams.

Lowlights: Picked Georgia to win by 10 points and the Bulldogs won the national title by 58 points (65-7).

Comments: Georgia repeated as national champions and now they set their sites on a potential three-peat. Head coach Kirby Smart has his work cut out for him in 2023.

8. Looking forward to 2023

Overall 2022 Statistical Analysis: 104/182 correct, 57.1% accuracy, one standard deviation above the mean, +3.8% improvement from 2021 season

Highlights: Was able to maintain a weekly posting cadence of ten games being analyzed and predicted.

Lowlights: Trying to end an 0/5 hitless streak for upsets in Week 2.

Comments: 2022 was an exercise in keeping pace and opening my eyes to more of the college football scene. That definitely happened and I can’t wait for the 2023 season hype to start in the early summer.

Same bat time, same bat channel for Week #1 of the 2023 season. There might be some changes, might roll with what I’ve got. Tune in and see what happens in the fall of 2023.

Never walk away from the table when you’re on a heater in Vegas!

Want to find highly rated beers near you? Check out this article that tells you how to locate great beer near you.

Comments

No comments yet. Why don’t you start the discussion?

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.