Mathematics Of Correctness : When To Give Up ?

[petrushkagoogol]

Consider an outcome based on 5 sequential decisions :

N=5 (mandatory)

 

Probability of Success P(x) >= 0.5 / step

 

0.5, 0.5, 0.5, 0.5, 0.5 Total = 2.50 Person A

0.4, 0.3, 0.4, 0.8, 0.7 Total = 2.60 Person B

 

Person B takes a hit initially, but gains in the long term.

 

0.1, 0.1, 0.1, x, y Total = NA Person C

 

Person C should give up after event 3 because he cannot reach 2.5 (0.5 * 5), even if he perfects outcome 4 and 5. (1/1).

 

Is giving up not as stupid as it seems ?     

 

P.S. Application could be a sequential manufacturing industry like diamond processing.

 

Steps are -

 

• Cutting
  
• Polishing
  
• Interim QA  - introduce a micro QA for minimizing time of subsequent steps for non-eligible diamonds 
  
• Bagging
  
• Setting
  
• QA
  

Edited July 15, 2017 by petrushkagoogol

[Buffy]

You may wish to look up "combined probabilities" and "Bayes' Theorem"... 

 

It is, however, true that people rarely know when to "give up" when they really should.

 

 

The 50-50-90 rule: anytime you have a 50-50 chance of getting something right, there's a 90% probability you'll get it wrong, :phones:

Buffy

[Maine farmer]

 

 

It is, however, true that people rarely know when to "give up" when they really should.

 

 

Is this why gambling is addictive to some?

 

If the weather forecast calls for a 20% chance of showers, it seems there is a 90% chance it will rain on the hay field that is almost ready to bale.

 

I don't need to gamble because I farm.

[exchemist]

Is this why gambling is addictive to some?

 

If the weather forecast calls for a 20% chance of showers, it seems there is a 90% chance it will rain on the hay field that is almost ready to bale.

 

I don't need to gamble because I farm.

Just seen this and I am sure it is all too true. Though I suspect we all tend to remember disproportionately those occasions on which a forecast was wrong, rather than the more routine incidences of them being right.