perm filename A04.TEX[BKG,PHY] blob sn#844717 filedate 1987-08-14 generic text, type C, neo UTF8
COMMENT āŠ—   VALID 00002 PAGES
C REC  PAGE   DESCRIPTION
C00001 00001
C00002 00002	\magnification\magstephalf
C00006 ENDMK
CāŠ—;
\magnification\magstephalf
\input macro.tex
\def\today{\ifcase\month\or
  January\or February\or March\or April\or May\or June\or
  July\or August\or September\or October\or November\or December\fi
  \space\number\day, \number\year}
\baselineskip 14pt
\def\fnc#1{\mathop{\rm #1}\nolimits}
\rm
\line{\sevenrm a04.tex[bkg,phy] \today\hfill}

\bigskip
\line{\bf The Plausible Improbable Scenario\hfil}

On his way to the quarterfinals in the 1984 US Amateur Championship, my
friend and sometime doubles partner Sam Hanna ($X$) was leading
Randy Prater ($O$) by~5 to~1 in a seven point match in the position below.
Sam declined to double, fearing that Prater would take, redouble, hit
a blot, and equalize the score. Sam should have doubled and prayed for a take.

\vskip 2truein

\noindent
If Prater drops a double, experts estmate that he has about a 9\% chance
in the 6--1 match against an equal opponent. To take and win the game requires
a long parlay of favorable events:

\smallskip
\display 20pt:$\bullet$:
Sam leaves a shot 44\% of the time.

\smallskip
\display 20pt:$\bullet$:
Randy hits it 33\% of the time.

\smallskip
\display 20pt:$\bullet$:
Randy contains the checker and wins the bearoff half the time.

\smallskip
\display 20pt:$\bullet$:
Randy wins the resulting tied match half the time.

\smallskip\noindent
So his chance is $0.44\times 0.33\times 0.5\times 0.5=0.037$, a bit under 4\%.
He must drop a double.

Sam saw that he had a plausible, and therefore scary, way to lose the match.
He forgot for a moment what he tells his Boston University classes about
probability:

\smallskip
\display 20pt:$\bullet$:
The probability of a parlay is the product of the probabilities of its stages.

\smallskip
\display 20pt:$\bullet$:
If each stage has a probability no more than 50\%, the probability of 
a two-stage parlay is no more than 25\%; three stages, 12\%; four stages, 6\%;
five stages, 3\%.

\smallskip\noindent
Sam's scenario for losing seemed plausible, because every step has a
substantial probability, yet the product law makes it hopelessly improbable.


\bigskip
\parindent0pt
\copyright 1987 Robert W. Floyd

First draft August 14, 1987

\bye