Thursday, May 12, 2016

The Amazing Statistics

I have a math degree: a Bachelor's of Science from Wagner College. I didn't have any idea as to how to use it. I definitely did not want to get into teaching. I spent so many years suffering in classrooms leading up to college . . . why would I go back? A few years later, I would get a Master's in Journalism from New York University. I should probably apologize to all the math professors I bothered at Wagner . . . especially Dr. Rahim Shambayati and Dr. Steve Sessions. This entry is dedicated to those two.

Math comes out in weird ways for me. If you've come to this blog in years past, you might remember that I came out with tables based on The Challenge. I think it goes back to a childhood poring over baseball statistics. Prior to Free Agents, I kept looking for patterns in performances, using Wikipedia to compile stats on many long-time Challengers. I went as far as differentiating between team efforts and those with a focus of individuals, pairs, and foursomes (see the second Battle Of The Seasons). That's how I discovered that Aneesa wasn't that good when playing for herself or with a partner. I stopped keeping track after Free Agents kept flipping the script in terms of format. I can probably do something on endgames (dubbed "elimination rounds" on the show), but I'm not breaking my neck doing that. If you've never seen me geek out, here are my posts of the men and women after Rivals II.

Survivor is more problematic to keep track, and the views of experts are based on subjective views based on overall gameplay. For instance, I consider Sandra Diaz-Twine to be the best lifetime player ever. Sure, Parvati Shallow competed in all but three days (a total of 114) in her three stints on the show, but not only is Sandra the only two-time champion (Pearl Island, Heroes Vs. Villains), she spent 77 of her combined 78 days with two of the biggest scumbags ever cast: Jon Dalton and Russell Hantz. The fact that she did not strangle either of them in their sleep makes her a saint. For excellence in a single season, I go with Tom Westman in Palau. He was the unofficial leader of the dominant Koror tribe that won all Immunity Challenges. He won all but two individual Immunity Challenges, and he did not get a single vote both times he was vulnerable. He got six out of seven votes on Day 39, with a player-hating Coby casting the sole dissent. Sure, Tom wound up breaking Ian into itty-bitty pieces by Day 38, and those two had an agreement that the winner of the final Immunity Challenge would take the less-favored Katie to the final day, but I feel that Tom was the best single season player. I just can't put it into intricate stats, making it subjective. Like how I feel that Scot Pollard and Kyle Jason are two of the ugliest motherfuckers in Survivor history, inside and out. That goes beyond numbers.

As for The Amazing Race, my reality show of choice? That is easily made for stats. In 2013, I ranked teams based on their average placements. I just went to Wikipedia, added  up their placements, then divided those by the number of legs run.I found out that the best with the best total was couple Rachel & Dave from the twentieth season. They averaged 1.83 per leg (22/12). On the flip side were Josh & Brent from TAR21, whose average was 4.58. That is still the worst total for a winning team, eclipsing Amy & Maya's 4.50 from TAR25. I kept trying to find meaning in the averages, going as far as deducting the best and worst performances for teams, then doing the calculations again. For instance: if you do that with Amy & Maya, the total is still 4.50. Do it again . . .still 4.50. While they were a mediocre team that lucked out in the final leg, they were very consistent about it.

I can't help but fool around with statistics and TAR. It is in that vein that I introduce something new: Strength of Final Three, or SOFT. The formula is easy to conceive, and the goal is to determine how good the remaining teams are going into the show's stretch run. Here's how it works.

1. Take the final three teams going into their season's finale. In older seasons, the last two legs should be discounted.

2. Give them three points for each first-place finish, two for second, and one for third.

3. Add up the points for the teams.

4. Divide that total by the product of legs run and six (the number of points that are "up for grabs" per leg).

5. Do NOT factor in the final leg of a season, since it would be redundant.

As you probably would have guessed, I do not have much of a life. Nevertheless, I feel that I'm onto something. Let's apply this to the first season's final three: Rob & Brennan, Frank & Margarita, and Joe & Bill:


Rob & Brennan
Frank & Margarita
Joe & Bill
Total
First
3
3
1
21
Second
1
4
5
20
Third
5
1
1
7
Total
16
18
14
48

48/(11)(6) = 48/66 = .727

Now, we look at the latest season. Heading into the final few legs, it looked as if the finale would consist of Brodie & Kurt, Tyler & Korey, and Burnie & Ashley. Brodie & Kurt had managed to finish first in half of the opening eight legs, but then they had a few bum legs and got into a Double U-Turn trap, as Tyler & Korey did the deed, while Burnie & Ashley prevented the Frisbee Bros from doing that to another team. Eventually, the boys would bow out after ten legs. Burnie & Ashley had not finished in first place throughout the season. They seemed on pace to match the feat set by Kelsey & Joey from the previous season, finishing in second place five times in a row. However, the pair wound up faltering on a Detour and switched tasks, which led to them getting narrowly eliminated by Sheri & Cole in the penultimate leg. Here are the SOFT results:

Tyler & Korey
Dana & Matt
Sheri & Cole
Total
First
5
1
0
18
Second
3
1
0
8
Third
1
3
2
6
Total
22
8
2
32

32/(11)(6) = 32/66 = .485

It turns out that TAR28 ties three other seasons in terms of overall lowest SOFT. It may have been due to the casting of social media stars, most of whom did not bicker as hard as teams from earlier seasons. As bad as it has been from Dana & Matt, they cannot hold a candle to Logan & Chris from TAR27. I could have lived with Justin & Diana winning, even with Justin's braggadocio, but not Logan & Chris. Here is the breakdown:



Season
Score
Legs
SOFT
Winners
20
53
11
.803
Rachel & Dave
9
48
10
.800
BJ & Tyler
1
48
11
.727
Rob & Brennan
7
42
10
.700
Uchenna & Joyce
5
49
12
.681
Chip & Kim
15
44
11
.667
Meghan & Cheyne
17
44
11
.667
Nat & Kat
23
43
11
.652
Jason & Amy
25
42
11
.636
Amy & Maya
27
42
11
.636
Kelsey & Joey
8
38
10
.633
Linzes
14
38
10
.633
Tammy & Victor
26
41
11
.621
Laura & Tyler
2
40
11
.606
Chris & Alex
24
40
11
.606
Dave & Conner
13
36
10
.600
Nick & Starr
3
39
11
.591
Flo & Zach
22
38
11
.576
Bates & Anthony
12
33
10
.550
TK & Rachel
10
36
11
.545
Tyler& James
6
34
11
.515
Freddy & Kendra
18
34
11
.515
Kisha & Jen
21
34
11
.515
Josh & Brent
11
35
12
.486
Eric & Danielle
4
32
11
.485
Reichen & Chip
16
32
11
.485
Dan & Jordan
19
32
11
.485
Ernie & Cindy
28
32
11
.485
TBD

Bold indicates highest score of final three.
The final legs of TAR25 and TAR26 are not counted, even though a team was eliminated midway through.

I'd like to know what you think, whether I'm onto something or if I need a life. Feel free to leave a comment below. And if you have your own metric for The Amazing Race, I'd want to hear about it.