The fall of 2016 may be best remembered for the election of Donald Trump as the President of the United States. He defied all odds as almost all “experts” predicted Hillary Clinton would win the election. In the aftermath of the upset, the methods used in forecasting the election outcome were scrutinized. On a far less serious note, in September, I predicted the outcome of the 23^rd season of Dancing With The Stars. Now that the results have been tabulated, I will assess my performance, and critically examine my methodology.

Following the first episode of the season, I used data from previous seasons of DWTS to predict the order of finish. Points were awarded to each of the 13 contestants based on 5 categories: judges’ score from the first show, age of the contestant, occupation of the celebrity, professional partner’s past performance, and presence of an inspirational story. After tallying the points, the predicted outcome was (from 13^th place to winner):

13: Rick Perry

12: Maureen McCormick

11: Babyface Edmonds

10: Vanilla Ice

9: Jake T. Austin

8: Amber Rose

6 (tie): Jana Kramer

6 (tie): Terra Jole

5: Marilu Henner

4: Calvin Johnson

3: Ryan Lochte

2: James Hinchcliffe

Winner: Laurie Hernandez

After an exciting and highly entertaining season, Laurie Hernandez was crowned the champion, and brought home the coveted mirror ball trophy, which she will undoubtedly display next to her Olympic gold medal (or maybe not). Hernandez’s victory at least partly validated my predictions, as she garnered most points on my analysis. But how accurate were the remaining picks from my forecast? To evaluate my performance, here are the official results:

13: Jake T. Austin

12: Rick Perry

11: Babyface Edmonds

10: Vanilla Ice

9: Amber Rose

8: Maureen McCormick

7: Ryan Lochte

6: Marilu Henner

5: Terra Jole

4: Jana Kramer

3: Calvin Johnson

2: James Hinchcliffe

Winner: Laurie Hernnadez

In addition to Hernandez, three other celebrities (Edmonds, Vanilla Ice, and Hinchcliffe) finished in exactly the same place as I predicted. Five contestants finished one spot away from my prediction (Perry, Rose, Jole, Henner, and Johnson). That’s a total of nine of thirteen dancers that my analysis predicted with high precision. Another contestant, Jana Kramer finished in fourth place, and was predicted to come in sixth place. However, there are three outliers in my evaluation (McCormick, Lochte, and Austin), all finishing 4 spots away from their predicted finish. So what can be learned from these deviations, and how can the analysis be changed to perform better in the future?

First, Maureen McCormick, the former child actor from the TV show Brady Bunch, finished in 8^th place, despite being picked to finish in 12^th place. It’s difficult to figure out where she could have earned more points in my analysis. She is 60 years old, with the oldest winner from the previous 22 seasons (Donny Osmond) being 51. She performed horribly on the first episode, receiving second to last lowest score from the judges. In addition, her partner, Artem Chigvintsev, has had little experience on the show, with only three prior seasons on his resume, where he finished in 11^th, 8^th, and 6^th places. It was clear she had little, if any, chance to win the competition, but there may have been a clue she would finish somewhere in the middle of the pack, and not in 12^th place. In season 22, two former child actresses from TV comedy shows, Kim Fields (Facts of Life) and Jodie Sweetin (Full House), finished in 8^th and 6^th places respectively. Season 18 winner, Alfonso Ribiero, also starred in a TV comedy (Silver Spoons) as a child. Perhaps former child actors should receive few extra points in my analysis.

Another disappointing prediction involved the Olympic swimmer, Ryan Lochte, who finished in 7^th place despite being picked to come in 3^rd. But unlike McCormick, there were more clues that suggested an unexpected outcome. First, Lochte (like Hernandez) accumulated massive points because he was an Olympic athlete. However, he is different than the five Olympians who raised the mirror ball trophy. First, four of the five demonstrated some element of dance in their sport, with two being gymnasts (Shawn Johnson and Hernandez) and two being figure skaters (Kristi Yamaguchi and Meryl Davis). The other Olympian to win the DWTS title was Apolo Ohno. He excelled in a sport (short-track speed skating), which requires agility and reflexes. Lochte, on the other hand, is a swimmer. Swimmers are not even required to walk, let alone use footwork to gracefully move around the floor or ice. Perhaps Lochte should not have received the maximum points in my analysis awarded to Olympians.

Lochte had another disadvantage that was not considered in my prediction. Although points were awarded for contestants with an inspirational story, no points were deducted from potential unlikable contestants. Just prior to the airing of this season, Lochte gained notoriety for his false robbery report at the Olympics. Although he seemed likable on the show, his reputation probably hurt his chances. Lochte’s negative persona coupled with mediocre dancing made his premature exit from the show inevitable. In future predictions, “unlikability” factor should be considered, although it would be hard to quantify.

Arguably the biggest surprise on the show was Jake T. Austin, who was the first to be eliminated. The 21-year-old former Disney Channel star was not predicted to contend for the title, mainly due to his poor performance on the first show, but his early exit was unprecedented. Of the 22 prior last place finishers, the youngest contestant was 29 years old (Josie Maran, Season 5). In addition, previous Disney channel stars have done well on DWTS. In season 17, Corbin Bleu (High School Musical) came in second. Zendaya, from the Disney Channel show Shake It Up, also finished in 2^nd place (season 16). With the available data, it seems impossible to have predicted that Austin would inherit the embarrassing title of DWTS “loser.”

As political analysts probably learned many lessons from their inaccurate prediction of the presidential election, I also learned from my first attempt as an amateur fortuneteller. There is no magic crystal ball to predict who will win the mirror ball trophy, but it was fun to try.