kissingerj payoff table final
TRANSCRIPT
Joseph KissingerCase 1 – Payoff Table
Executive Summary
I recommend Tom Roberts.
Background
I have completed my training rotation. I will soon transition into a permanent role and a
new trainee must be chosen to take my place. This trainee will also serve as my junior analyst
during the beginning of my new assignment. I narrowed down the choices to five candidates and
invited each to the office for an interview and aptitude test, which measured the candidates on a
Likert scale of 1 to 7, with 7 being the best, in the four areas in which I rotated during my time in
the training program. Of the four areas, there is a 35% chance I will be permanently placed in
International Investing, a 20% I will be placed in Real Estate, a 35% of being placed in Personal
Finance, and a 10% probability of being assigned to Health Care. The raw data collected from
the aptitude tests can be found in the appendix of this report, in Figure 1: Aptitude Test Results.
The five candidates, all recent MBA graduates, are Samantha Lindle from Wake Forest, Andy
Karoly from Duke, Wendy Lee from UNC, Tom Roberts from Virginia, and Bill Williams from
Clemson. I considered who would be the most qualified candidate for the company, and who
would be the most helpful to me during my first months in my permanent role. This criteria
presented tradeoffs that had to be made to determine who I would recommend to take my place
and be my junior analyst.
A nalysis
The tool used to analyze the data in this report and for this recommendation was a payoff
table. Payoff tables are useful decision making tools when a scenario encompasses a number of
possible outcomes and a variety of possible responses. Payoff tables are also used when looking
at a single decision. To make a decision using payoff tables, we use decision rules to compare
decision alternatives to one another for a given state of nature. The decision rules give us
different looks at the data and a method of ranking each alternative against one another. The
rankings are analyzed and eventually used to make a single decision from a number of
alternatives and possible future states. Below, labeled Table 1: Payoff Table, is the original
payoff table used in this analysis.
Table 1: Payoff Table
35% 20% 35% 10%
International
InvestingReal
EstatePersonal Finance
Health Care B of B B of W E.L. E.V. M/M R
Samantha 20 14 10 21 21 10 16.25 15.4 10Andy 30 9 15 10 30 9 16 18.55 13Wendy 13 19 18 23 23 13 18.25 16.95 17Tom 16 15 18 19 19 15 17 16.8 14Bill 14 11 13 11 14 11 12.25 12.75 16
In the table above, the candidates’ names along the side are the alternatives while the four
business areas along the top of the table are the possible states of nature. The percentages above
the states of nature are the probability of each state occurring, given the chosen alternative. The
payoffs were derived by summing the candidates’ scores on the aptitude tests to get a total score
for each business area. Given the 1-7 scale of the aptitude test, the range of payoffs for each
candidate and business area was 5 to 35. The decision rules can be seen in the last five columns
and from left-to-right read: Best of the Best, Best of the Worst, Equally Likely score, the
Expected Value score, and the MiniMax Regret. Best of the Best shows the highest score for
each candidate, while the Best of the Worst shows each’s lowest score. When we take the
difference between the Best of the Best scores and the Best of the Worst scores we find the range
of payoffs for each candidate. The Equally Likely rule shows the average of each candidates’
scores. We can compare the Equally Likely rule to the midpoint of the range to determine where
the data is clustered. That is, an Equally Likely score greater than the midpoint tells us that most
of the data is on the high end of the range. The Expected Value rule is the weighted average of
each candidates’ payoffs, given the probabilities of each state of nature occurring. When
compared to the Equally Likely score, the Expected Value can tell us the relationship between
the weights and the payoffs. That is, an EV greater than the EL suggests a favorable match of
high payoffs and high probabilities. The MiniMax Regret rule tells us how much, in opportunity
costs, we would be giving up if a certain alternative is chosen. Because the MiniMax Regret and
Best of the Worst rules are conservative measures, they can be compared on a basis of rankings
to see if there is a relationship between the regret and the lowest payoff for each candidate. The
rules are ranked to analyze comparisons and make tradeoffs between payoffs. The rules here
were ranked using green, yellow, and red shading to indicate the winner, second place, and the
loser, respectively. The colors make reading the rules much easier and allow us to begin making
assumptions.
To analyze the payoff table most effectively and timely, I first checked to make sure there
was nothing on the table that did not need to be analyzed. I noticed that Bill Williams could be
eliminated from the table and my analysis because of domination. An alternative that is
dominated is one that, for each given state of nature, is consistently the loser when compared to
the other alternatives. If an alternative is dominated by at least one other alternative it can be
discarded, as I would never choose a lower payoff given the same probability and state of nature.
For example, in Table 1 we see that Tom has a higher score in all areas than Bill. When this is
the case we would choose Tom over Bill every time. Further, if we were to choose Lindle,
Karoly, or Lee over Tom Roberts, and Tom Roberts is consistently better than Bill Williams,
then any time we prefer a candidate over Tom Roberts we also prefer the same candidate over
Bill Williams. Therefore, Bill can be eliminated from our analysis and consideration. The payoff
table showing Bill to be dominated, with new rankings for the decision rules, is shown below as
Table 2: Domination. While the green, yellow, and red shading again indicate the winner,
second-best, and loser for each decision rule, notice the gray shading indicates Bill is dominated.
Table 2: Domination
35% 20% 35% 10%
International
InvestingReal
EstatePersonal Finance
Health Care B of B B of W E.L. E.V. M/M R
Samantha 20 14 10 21 21 10 16.25 15.4 10Andy 30 9 15 10 30 9 16 18.55 13Wendy 13 19 18 23 23 13 18.25 16.95 17Tom 16 15 18 19 19 15 17 16.8 14Bill 14 11 13 11 14 11 12.25 12.75 16
The first alternative to be analyzed is Samantha Lindle from Wake Forest University.
Samantha has the next-to-last Best of the Best (BoB) and Best of the Worst (BoW) scores, at 21
and 10, respectively. This relationship gives us the second largest range, at 11, which could be
concerning. The midpoint of this data range, 15.5, falls just below the equally likely (E.L) score,
at 16.25, indicating that the payoffs are slightly grouped above the midpoint, which is favorable.
The expected value (E.V.) score of 15.4 ranks last and is less than the E.L., indicating an
unfavorable matching of probabilities and payoffs. The regret score of 10 ranks the best on the
table, while our other conservative decision rule, BoW, ranks next-to-last for this alternative. The
disagreement between our conservative rules indicates the need for further inspection of our data
set. The regret table, which can be found in the appendix of this report titled “Table 3: Regret”,
shows why this relationship is in disagreement. With the exception to Andy, each alternative’s
regret score comes from the same state of nature, International Investing. Andy’s score of 30 in
International Investing is significantly greater than any other data point on the table, causing the
regret scores to be skewed by this outlier. The regret score, then, is more of a measure of how
much we would regret recommending someone other than Andy if we ended up in International
Investing, or how much we would regret choosing Andy if we ended up in Health Care. The
major takeaway from the MiniMax Regret score is that it is being adversely affected by an
outlier and thus should not be analyzed at the surface level, but requires further investigating to
fully understand. I will make reference to this throughout my analysis. This alternative causes
concern as the BoB and BoW scores are next-to-last and the range of 11 is the second biggest.
The best payoff for Samantha is 21 and the worst is 10, as shown in the BoB and BoW
decision rules. The other payoffs are 20 and 14, giving us the E.L. score of 16.25. The midpoint
of the range is 15.5, just below the E.L. The E.V. is 15.4 and is lower than the E.L. due to the
highest probability, 35%, of getting the lowest score of 10, and the lowest probability, 10%, of
getting the alternative’s highest payoff of 21. The remaining payoffs of 20 and 14 have
probabilities of 35% and 20%, respectively, thus bringing the E.V. closer to the E.L. and
midpoint. Though the regret score of 10 was the best on the table it is being affected by an
outlier, causing disagreement with our other conservative decision rule, BoW, and therefore is
not accurately measuring regret.
The highest payoff for this alternative is 21 and is a major reward. The lowest payoff is
10 and, conversely, is a major risk. Another risk in Samantha’s lowest payoff is that it is the
second lowest score on the entire table. There is also a major risk in the probabilities, since there
is a 35% chance of receiving Samantha’s lowest payoff. Additionally, there is only a 10% chance
of receiving her highest payoff, a minor reward. The second highest payoff of 20 is certainly a
reward, especially given its 35% chance of occurring, which is also a reward. The remaining
payoff of 14 is considered a very minor risk, as well as the 20% chance of this payoff occurring.
Thus, there is a 45% chance of receiving a high reward, which is a reward, and a 55% chance of
receiving a risk, which is a risk. From another point of view, there is an equal chance, 35%, of
receiving Samantha’s lowest payoff and second highest payoff. The 35% chance of receiving her
lowest payoff is a risk. This, and the only 10% chance of receiving Samantha’s highest payoff,
causes me to consider this alternative to be moderately risky.
The next alternative to be analyzed is Andy Karoly from Duke University. This
alternative has the highest BoB score, at 30, and the lowest BoW score, at 9, giving us the widest
possible range of payoffs. This is an early concern. The E.L. score of 16 ranks last on the table,
and is well below the midpoint of 19.5, which is significantly influenced by Andy’s exceptional
BoB score of 30. Because the E.L. is lower than the midpoint, the payoffs should be grouped to
the low end of the range. This alternative has the highest E.V., at 18.55, which looks great at first
but requires further analysis to completely understand. Nonetheless, an E.V. greater than the E.L.
indicates a favorable relationship between high payoffs and high probabilities. Finally, the regret
score of 13 is second best on the table, while the BoW score ranks the worst. This again causes
concern because of the disagreement between conservative rules. However, it is worth noting
that this alternative is the cause for skewed MiniMax Regret scores. While the other alternatives’
regret scores came from International Investing, Andy’s came from Health Care. As mentioned
earlier, this is not accurately measuring regret.
In the data, we see the highest payoff of 30 and the lowest payoff of 9. The other payoffs
are 15 and 10, giving us the worst E.L. score on the table at 16. Though the E.L. is the worst on
the table, the E.V. for this alternative ranks the best because of the low probabilities associated
with the two lowest payoffs, and the high chances of receiving high payoffs. This relationship is
very favorable. Though we would expect a high regret score for the alternative with the worst
BoW score, this alternative’s regret score is second best. This is again because of Andy’s
exceptional score of 30 in International Investing which is causing the regret scores to be
skewed.
This alternative has a reward of a high payoff of 30 and a risk of a low payoff of 9. These
scores are the highest reward and biggest risk on the table. The second highest payoff for Andy is
15 and is considered a very minor reward, while the remaining score of 10 is only marginally
less risky than the low payoff of 9. Looking at the probabilities, there is the greatest chance of
receiving a reward, with each payoff having a probability of 35%. The risks have probabilities of
20% and 10%, giving us a 30% chance of receiving a risk. While the probabilities are in favor
with our payoffs, which is a reward, we must be careful not to focus solely on the data presented
in this way. When looking at the data differently, we see there is a 70% chance of receiving a
reward. This seems great. However, the rewards vary significantly, creating a risk. That is, it is
just as likely to receive a score of 15 as it is to receive a score of 30. While 30 would be the
greatest payoff possible, receiving a payoff of 15, now half of our hopeful payoff of 30, would
hardly seem rewarding. Therefore, I would consider receiving anything other than the payoff of
30 to be a loss. Even if we receive a rewarding payoff of 15, the opportunity cost of losing out on
the other 15 (30-15=15) is too significant for me to overlook. Overall, I would classify this
alternative as a moderately high risk, given a 35% chance of a desirable payoff and a 65%
chance of missing out on the ultra-rewarding payoff of 30.
The third alternative to analyze is Wendy Lee from UNC. This alternative provides us
with both the second-best BoB and BoW scores and this is a favorable agreement. The BoB and
BoW scores give us a range of 10, which is the second best among alternatives. These payoffs
give us a midpoint of 18, which is just below the E.L., at 18.25, which indicates the data is just
slightly grouped above the midpoint. This too is a favorable agreement. However, when we look
at the E.V., though second-best among alternatives, we see it is less than the E.L. This indicates
an unfavorable match of probabilities and payoffs. Finally, this alternative has the worst regret
score due to Wendy’s lowest payoff occurring in International Investing, the area where Andy is
skewing the data.
In the data, we see a high score of 23 and a low score of 13, as shown in the BoB and
BoW rules. The remaining scores are very close to one another, at 18 and 19, which brings the
E.L. score to be 18.25. However, because there is not an equally likely chance of each state of
nature occurring, and because this alternative’s lowest two payoffs are associated with the
highest probabilities, the E.V. of 16.95, though second best on the table, is less than the E.L.
Despite the close grouping of this alternative’s payoffs, Wendy has the worst regret score on the
table. It is interesting that the alternative with the worst regret score has the second-best BoB and
BoW scores. This is because Wendy’s lowest payoff of 13 is in International Investing, where
the outlier lies. Because of the outlier, the regret score was the worst on the table. However,
when looking at the regret table, which again can be seen in Table 3 in the appendix of this
report, we see that this alternative had the highest regret in International Investing, but no regret
in any of the other areas. That is, Wendy has the lowest payoff in International Investing, but is
at least tied for the highest payoff in the other three areas. Because of this, the regret score should
not be considered too heavily.
This alternative has one outstanding reward of 23, its highest payoff, and one very slight
risk of 13, its lowest payoff. The 10% probability of receiving the outstanding payoff is a minor
reward. The remaining scores of 18 and 19, and their 35% and 20% respective probabilities of
occurring are all rewards.. Thus, there is a 65% chance of receiving a reward, which is a reward
in itself. Conversely, there is a 35% chance of receiving a very manageable risk, which is still a
risk but not a major one. While it may be a concern that there is a 70% chance of receiving this
alternative’s worst two payoffs, it is worth noting that Wendy’s next-to-worst payoff of 18, in
Personal Finance, is actually tied for the best score in that area. Actually, while there is a 35%
chance of receiving Wendy’s worst payoff, in International Investing, there is a 65% chance of
receiving the best score possible in the other three areas. This is a major reward. Given the major
rewards and minor risks associated with this alternative, I would classify this alternative as a low
to moderately low risk.
The last alternative to be analyzed is Tom Roberts from the University of Virginia. This
alternative has the lowest BoB score, at 19, but the best BoW score, at 15. This relationship
creates the smallest possible range of payoffs, at 4. With such a small range, we expect the
payoffs to be close to one another. The midpoint for this alternative is 17 and is equal to the E.L.,
suggesting the payoffs may be uniformly distributed around the midpoint. Next, we see that the
E.V., at 16.8, is just below the E.L. and midpoint, indicating a slightly unfavorable match of
probabilities and payoffs. Finally, the regret score, at 14, is next-to-last on the table and is again
caused by the outlier in International Investing.
In the data, we see a high payoff of 19 and a low payoff of 15. This small range suggests
Tom has general knowledge in all of the tested areas, as is reinforced by his remaining scores of
16 and 18. Interestingly, this alternative’s highest and lowest payoffs have the smallest
probabilities of occurring at 10% and 20%, respectively. It is also interesting to see the uniform
distribution of payoffs, as the payoffs lie equally below and above the midpoint. If each state of
nature were equally likely to occur, the E.V. would equal the E.L. However, there is a 55%
chance of the payoff falling below the midpoint, and a 45% chance of receiving a payoff above
the midpoint. This is slightly unfavorable and is the reason the E.V. is just less than the E.L. The
regret score is again being skewed by Andy’s exceptionally high score in International Investing.
This alternative’s highest payoff is a reward, at 19, while the lowest payoff, at 15, is the
risk. The high and low payoffs for this alternative give us the lowest range on the table, which
represents a degree of consistency and can be considered a reward. The 10% probability of
receiving Tom’s highest payoff is a minor reward. Because 15 is the lowest payoff for this
alternative it is considered a risk, though it is not undesirable when compared to other scores in
the Real Estate area. The remaining two scores, 16 and 18, can be considered rewards. The 35%
probabilities associated with the payoffs of 16 and 18 are also rewards. Thus, this alternative
provides us with three rewards and one very manageable risk. When looking at the data
differently, we see that there is an 80% chance of receiving a reward, and a 20% chance of
receiving a risk. I could even argue that the risk of 15 in Real Estate is one worth taking, since it
is the second best score in the Real Estate area. It is also worth noting that this alternative’s
second lowest payoff, 18, is tied for the best payoff in Personal Finance. This is a reward, as is
the 35% chance of this payoff occurring. For all of the reasons listed above and because of the
conservative nature of a low range alternative, I would classify this alternative to be a
conservative, low risk choice.
Recommendation
Comparing the first two alternatives analyzed, Samantha and Andy, the highest payoff for
each is 21 and 30, respectively, and Andy wins. When looking at the lowest payoffs, we see
Samantha has a low of 10 while Andy has a low of 9, making Samantha the winner of the BoW
scores. Considering this, I prefer Andy over Samantha as I would risk losing one point to
possibly receive a nine point higher payoff. When we look at the probability of the highest
payoff for each alternative, we see Samantha’s high score of 21 has only a 10% chance of
occurring, while Andy’s high score of 30 has a 35% chance of occurring. The preference here
again favors Andy, as both his payoff and probability are higher than Samantha’s. Looking at the
chance of the lowest payoff, we see Samantha’s lowest payoff of 10 has a 35% probability of
occurring. Meanwhile, Andy has only a 20% chance of receiving his lowest payoff, at 9. This
comparison again favors Andy as there is a greater chance of receiving Samantha’s low payoff.
Though Samantha’s lowest payoff is higher than Andy’s lowest payoff, Andy’s probabilities
give us a better chance of receiving a payoff other than his lowest. The second best payoffs for
Samantha and Andy are 20 and 15, respectively. Both alternatives’ second highest payoffs have
35% probabilities associated with them, favoring Samantha as we would have the same chance
of receiving an additional 5 points with Samantha. Though this favors Samantha, it is not enough
to change my overall preference for Andy because, if I were to choose Samantha and receive her
payoff of 20, in International Investing, I would actually be missing out on an additional 10 in
payoffs if we had chosen Andy and been placed in International Investing. While there is a 45%
chance of receiving Samantha's highest two payoffs, there is a 70% chance of doing so with
Andy. Thus, there is a 55% chance of receiving one of Samantha’s lowest two payoffs, and a
30% chance of receiving one of Andy’s lowest two scores. This favors Andy. Also, with Andy,
we have a good chance to receive an outstanding reward, while we maintain the same chance at
receiving significantly lower payoffs with Samantha. We also see that Andy gives us a 70%
chance of receiving a payoff of 15 or higher. With Samantha, there is a 55% chance we receive a
payoff of 14 or less. Further, there is a 45% chance of receiving a payoff of 20 or higher with
Samantha, but I would be willing to increase my probability to 70% to receive a payoff of 15 or
higher, with potential for the highest payoff on the table. Overall, my preference is for Andy due
to the reasons listed above, and his better match of strong probabilities and high payoffs when
compared to Samantha.
Next, I will compare Andy to Wendy. The highest payoffs for Andy and Wendy are 30
and 23, respectively. Andy again wins this comparison, as he will continue to do because of his
exceptional score in International Investing. The lowest payoffs for Andy and Wendy are 9 and
13, respectively, and this favors Wendy. Though Wendy’s lowest payoff has the best chance of
occurring, her score of 13 is only two points lower than Andy’s second best score of 15.
Therefore, Wendy’s biggest risk is only marginally worse than Andy’s second best reward. This
heavily favors Wendy. The ranges for these alternatives are 21 for Andy and 10 for Wendy.
Because of the better low-end payoff and the smaller range, I prefer Wendy over Andy. When
we consider the probabilities, there is a 35% chance of receiving Andy’s highest payoff and a
10% chance of receiving Wendy’s highest payoff. This favors Andy greatly as his highest payoff
is significantly greater than Wendy’s, with a much better chance of occurring. Also in favor of
Andy is the probability of receiving each alternative’s lowest payoff, as Andy has a 20% chance
and Wendy has a 35% chance. This would be enough to change my preference to Andy if we did
not realize that Wendy’s worst payoff is not that risky when compared to Andy’s payoffs, and if
we did not look further into the probabilities. Though the probabilities are in Andy’s favor when
we look at the highest and lowest payoffs, the overall probabilities favor Wendy. While there is a
70% chance of receiving a reward from Andy, and a 65% chance of receiving a reward from
Wendy, there is also a 65% chance of receiving a better payoff from Wendy. Actually, Wendy’s
payoffs are the best scores in Real Estate, Personal Finance, and Health Care, while Andy has the
highest score in only International Investing. Because of this, I would take my chances with
Wendy as I would rather have the best score in three areas than a significantly better score in one
area. This is another reason why I prefer Wendy over Andy, as the training program suits
generalists better than it would specialists as trainees are rotated through different business areas.
Andy may be exceptional in International Investing, but his other scores suggest he may be more
of a specialist and could be disengaged in areas other than where his strengths lie. Wendy brings
more stability to the table, which is probably best since I do not know which area I will be placed
in, and because I would much rather make a conservative first hire decision, rather than a high-
risk, high-reward choice that may not work out. Also, it is important to realize that though there
is a 70% chance of receiving one of Wendy’s two lowest payoffs, her second lowest payoff of 18
is the best in Personal Finance. For these reasons, my overall preference is with Wendy. Next,
we will compare Wendy to Tom to make a final recommendation.
The highest payoffs for Wendy and Tom are 23 and 19, respectively, and this favors
Wendy. The lowest payoffs for Wendy and Tom are 13 and 15, respectively, and this favors
Tom. While these comparisons do not tell us a lot just yet, I prefer Tom over Wendy because his
small range of only 4 is much better than Wendy’s range of 10 and I can therefore count on Tom
to be a more stable generalist than Wendy. When we consider the probabilities, we see that
Wendy has a 35% probability of receiving her lowest payoff, while Tom has a 20% chance of his
lowest payoff. Therefore, not only does Tom have a higher low score than Wendy, but we are
also less likely to receive Tom’s lowest payoff than we are Wendy’s lowest payoff. We can also
see that there is an 80% chance of receiving a reward from Tom, and a 65% chance of receiving
a reward from Wendy. While these are both favorable probabilities, Tom’s chances are better.
We also see that, for International Investing and Personal Finance, Tom has an equal chance as
Wendy at receiving better payoffs. Wendy’s better payoffs lie in areas with the lowest
probabilities. This places favor on Tom as his scores are not much worse than Wendy’s in Real
Estate and Health Care and there is the lowest chance of being placed in these areas. That is,
Wendy’s best rewards lie in areas we are less likely to be placed in, creating opportunity costs if
we chose Wendy and were not placed in one of her best areas, which has a 70% chance of
happening. Because of Tom’s small range and distribution of payoffs, the 80% chance of
receiving a reward, the equal chance of receiving Tom’s better payoffs in International Investing
and Personal Finance, the 70% chance I am placed in one of those two areas, and because he has
the highest BoW score, a conservative measure, my overall preference is for Tom as he presents
himself as a capable generalist, scoring fairly well in all four areas despite the difficulty of the
aptitude tests. With this in mind, I believe it would be best to choose Tom as he is capable of
making an impact in any of the four areas in which I may be placed.
Appendix
Figure 1: Aptitude Test Results
International Investing Real Estate
Personal Finance
Health Care
Aptitude 7 3 1 5Analytical Training 2 3 3 2Experience 1 2 1 3Creativity 4 5 4 6Interest 6 1 1 5
International Investing Real Estate
Personal Finance
Health Care
Aptitude 6 1 5 2
Analytical Training 7 2 4 3Experience 5 2 4 1Creativity 5 3 1 3Interest 7 1 1 1
International Investing Real Estate
Personal Finance
Health Care
Aptitude 2 3 4 5Analytical Training 3 3 4 3Experience 1 3 2 5Creativity 4 3 3 4Interest 3 7 5 6
International Investing Real Estate
Personal Finance
Health Care
Aptitude 1 5 6 2Analytical Training 6 1 4 6Experience 2 3 1 2Creativity 4 2 4 6Interest 3 4 3 3
International Investing Real Estate
Personal Finance
Health Care
Aptitude 3 2 3 2Analytical Training 3 2 2 1Experience 2 3 1 2Creativity 3 2 4 3Interest 3 2 3 3
Samantha Lindle from Wake Forest:
Andy Karoly from Duke
Wendy Lee from UNC
Tom Roberts from Virginia
Bill Williams from Clemson
Table 1: Payoff Table
35% 20% 35% 10%
International
InvestingReal
EstatePersonal Finance
Health Care B of B B of W E.L. E.V. M/M R
Samantha 20 14 10 21 21 10 16.25 15.4 10Andy 30 9 15 10 30 9 16 18.55 13
Wendy 13 19 18 23 23 13 18.25 16.95 17Tom 16 15 18 19 19 15 17 16.8 14Bill 14 11 13 11 14 11 12.25 12.75 16
Table 2: Domination
35% 20% 35% 10%
International
InvestingReal
EstatePersonal Finance
Health Care B of B B of W E.L. E.V. M/M R
Samantha 20 14 10 21 21 10 16.25 15.4 10Andy 30 9 15 10 30 9 16 18.55 13
Wendy 13 19 18 23 23 13 18.25 16.95 17Tom 16 15 18 19 19 15 17 16.8 14Bill 14 11 13 11 14 11 12.25 12.75 16
Table 3: Regret
International Investing Real Estate Personal Finance Health Care
Samantha 10 5 8 2
Andy 0 10 3 13Wendy 17 0 0 0Tom 14 4 0 4Bill 16 8 5 12