This is very neat, but if 1 is best and 2-3 equally matched can 2 and 3 tie/draw? Or is there noisy wins? These isomorphs are so interesting
-
-
-
A tie would instantly tell you that 1 is the strongest.
- Još 9 drugih odgovora
Novi razgovor -
-
-
If find this intuitive, and use stackexchange, you can weigh in here, I just posted this response: https://math.stackexchange.com/a/3360686 I think the current #2 is fine, but #1 has got to go.
Prikaži ovu nit -
Prikaži ovu nit
Kraj razgovora
Novi razgovor -
-
-
It's an interesting version but not very intuitive for me, my mind starts to nitpick. For instance with boxers rock paper scissor - types of advantages are common. But if there'd be one who defeats 98 then prob is different. With boxers hierarchies are not so linear..
-
Try Mike Tyson vs two emeritus professors, works for me!
- Još 4 druga odgovora
Novi razgovor -
-
-
This is really clever—a good device for explaining it.
-
Thanks! It's been on my list of things that maybe have research potential for awhile, but I never did anything with it. Hopefully twitter gives me, or someone else, some inspiration.
- Još 2 druga odgovora
Novi razgovor -
-
-
I have been thinking. Your example is intuitive because it makes the options 'fight'. Maybe I am too much trained in Bayes, but it is the 'dynamics' that gave me intuition. Compare likelihood you were right on your first vs. second guess. First guess, 1/3. Second, 1/2 -> switch.
-
this approach is general, does not require 100 doors, just underlines the fact that you are actually comparing your former self (who had less info) to your actual self (with more info) and former self had lower chances of being right.
- Još 3 druga odgovora
Novi razgovor -
Čini se da učitavanje traje već neko vrijeme.
Twitter je možda preopterećen ili ima kratkotrajnih poteškoća u radu. Pokušajte ponovno ili potražite dodatne informacije u odjeljku Status Twittera.
| behavioral economics | experimental & statistical methods