# Thread: Success in Survivor - A Mathematical Model

1. You know this is a very good formula kind of stuff... but you forgot one important thing... emotions and emotions can't really be calculated with numbers. Every season really depends on how emotional the survivors are etc.

2. Originally Posted by lalol
You know this is a very good formula kind of stuff... but you forgot one important thing... emotions and emotions can't really be calculated with numbers. Every season really depends on how emotional the survivors are etc.
Absolutely correct. There are obviously things which cannot be calculated that factor in to the game.

For example, I'd love to be able to give extra points to a tribal leader as this would be an excellent indicator of who was in control of the game. But how can you, with any degree of objectivity, define who all the tribal leaders were? The answer is, you cannot. So, since I cannot give these types of qualities an unbiased value, I chose not to include them in the first place.

My intent was not to try to quantify everything, only to set values to those things that were quantifiable.

3. Originally Posted by Roz_City
So... What's the answers? Did I get it right
Spot on! Good work Roz_City!! See the next post for details.

4. ## How about that...Jenna DID deserve to win!

Okay. Here are the results for each season of the show along with some applicable notes. Each castaways name is followed by their finishing position followed by their point total.

Survivor - Pulau Tiga

Kelly (2).46.918
Rudy (3).40.977
Richard (1).40.887
Susan (4).39.214
Sean (5).35.707
Gretchen (10).29.769
Colleen (6).29.590
Jenna (8).29.381
Joel (11).25.778
Gervase (7).25.433
Dirk (12).23.635
Greg (9).22.984
Ramona (13).16.432
Stacey (14).8.266
Sonja (16).5.286
B.B. (15).3.048

Notes: Kelly gets the nod here because near the end, she was a challenge hog. Winning the last four IC's, in effect, wrested control of the game from the manipulative Richard. If I could award points for controlling an alliance, then perhaps Richard would have come out on top. It's interesting that Rudy edged out Richard for second place. Why? Rudy was more consistent in casting a vote at TC for the person who got the boot. While both attended the same number of TC's, Richard did not cast a vote for Sonja at the first TC while Rudy did. In fact, from that point forward, they both voted the same way and always for the person getting the boot. And look where Pagong member Gretchen finished! Again, she was much more consistent with voting correctly at tribal council and had fewer votes cast against her for the time she was on the island.

Survivor - The Australian Outback

Colby (2).53.639
Tina (1).42.922
Keith (3).40.834
Elisabeth (4).33.095
Rodger (5).32.885
Amber (6).32.814
Jerri (8).31.500
Nick (7).31.333
Alicia (9).30.429
Jeff (10).29.086
Michael (11).25.095
Mitchell (13).21.111
Kimmi (12).19.956
Maralyn (14).16.487
Kel (15).1.905
Debb (16).1.000

Notes: Was there any doubt as to who would come out on top this season? Colby wracked up 5 post-merge immunities and 3 post-merge rewards. He carried Tina to the final 2 only to hand deliver her the cheque for \$1 million. This season really seemed to fit well with the model in terms of where people actually finished compared to where they "should have" finished.

Survivor - Africa

Kim J. (2).47.917
Lex (3).46.402
Ethan (1).43.275
Tom (4).37.849
Kelly (9).32.571
Kim P. (6).32.144
Brandon (8).30.976
Teresa (5).29.378
Clarence (10).28.139
Frank (7).28.054
Lindsey (11).23.635
Silas (12).20.556
Jessie (15).15.154
Linda (13).10.709
Carl (14).7.000
Diane (16).2.429

Notes: Hmmmmm. Kim Johnson? The best player in Africa? Well, she did win the last two IC's which cemented her spot in the Final 2. Lex was an immunity hog in his own right, winning three post-merge IC's. Ethan won the first post-merge IC and RC and won nothing else from that point forward. Except, of course, the game itself!

Survivor - Marquesas

Neleh (2).45.617
Kathy (3).42.851
Paschal (4).42.429
Vecepia (1).39.626
Sean (5).35.972
Tammy (7).32.260
Robert (6).32.027
Zoe (8).32.003
John (9).29.000
Rob (10).28.681
Gina (11).21.752
Sarah (13).18.750
Patricia (15).15.154
Hunter (14).14.778
Gabe (12).6.476
Peter (16).3.857

Notes: Just as everyone suspected all along, Kathy got game! Or, at least, had it when she was in Marquesas. She played well, but eventually was no match for Neleh and Pappy's alliance which saw each of them with a 100% accuracy rating when it came to voting out members of the tribe. Vecepia, the eventual winner, finished with the fourth highest score making the Marquesas version of the show the biggest anomaly in my model.

Survivor - Thailand

Brian (1).54.484
Clay (2).40.910
Ted (5).39.954
Helen (4).39.467
Jan (3).38.449
Ken (8).29.580
Penny (7).29.286
Jake (6).28.713
Erin (9).27.624
Shii Ann (10).23.242
Ghandia (13).15.484
Tanya (15).15.154
Robb (11).12.881
Stephanie (12).10.455
Jed (14).4.976
John (16).2.429

Again, there should be no surprises here. Brian won the last three IC's. He knew who to vote for each time at TC. No one, not one single person, ever wrote his name on a piece of parchment until it was time to hand him the money.

Survivor - The Amazon

Jenna (1).44.511
Mathew (2).44.150
Rob (3).41.184
Butch (4).36.810
Heidi (5).36.474
Alex (7).33.560
Deena (8).31.960
Roger (10).28.319
Christy (6).28.072
Dave (9).27.000
Shawna (11).18.667
Jeanne (12).16.841
Joanna (13).11.042
Daniel (14).7.068
Ryan (16).5.286
Janet (15).4.619

Notes: Now, don't get me wrong. I was never a big fan of Jenna's either. But deep down I was hoping that this model might somehow justify her finishing where she did. And, believe it or not, it does. Jenna won four post-merge IC's including the last two which, as we all know, are the most important ones. She was quite consistent in correctly voting at TC's and she only had three votes cast against her IN TOTAL. That was over the course of 11 TC's. The fact that we percieved her to be such an awful person while she drew relatively little fire for her actions at TC is a testament to her gamesmanship. Amazon was another example of a season which provided a good fit for the model.

5. Interesting results, tmcrae. Your #5 criteria Casting a Correct Tribal Council Vote has always seemed to be one of the most important factors for staying in the game. A player who can stay on the right side of a majority and sense when to switch will usually go far. The results for Marquesas shows the uncertainty of the purple rock and then my favorite of that season Kathy going next.

6. This is really all we can calculate as fans of this show... and I got to agree except I would never think Butch should have been 4th and Heidi 5th

7. I think this is an interesting excercise, as it assigns points to those areas of the game that are likely important, but I wonder if it has any value as a predictive tool, given that it is mainly data compiled after each event, and those events in turn effect the total number attainable to the end.

8. I think this is pretty interesting results! Vee finished lowest of all the Survivor winners, truly showing that she should not have won that season! I guess Jenna really did deserve to win, it seems that way anyway.

9. I think that if somebody dosn't get a vote against them(untill they get voted out) it could be because they are a terrible player. Case in point Rodger(S2) or Jan(S6) which both are rated high on your lists and did finish high on the show just because they were dead ducks.

It's hard to argue who may be the best player ever in the history of the show but alot of the worst players in the history of this game finish about 5-8ish overall and in the jury just because they so useless as game players and other people can easily manipulate them because they know they are good for a vote but otherwise dead weight.

10. Originally Posted by Bill_in_PDX
I think this is an interesting excercise, as it assigns points to those areas of the game that are likely important, but I wonder if it has any value as a predictive tool, given that it is mainly data compiled after each event, and those events in turn effect the total number attainable to the end.
No. It has no value as a predictive tool at all and it was not meant to. It was only meant to be a means to show who played the game the best after all was said and done.

Astrogirl posted a couple of links above which show formulae one person has used to attempt to predict who will win based on factors from previous seasons.