We must be able to measure a group or our own expectations in order to predict the outcomes. We are only dealing with two possible outcomes, so we are either building expectations or lowering them. After we identify an action we would like to predict with only two possible outcomes, we need to apply a numerical value and natural limit to it.
When we measure expectations of groups, we do not focus on a single member’s intensity but rather the sum of the whole group. If we are measuring the expectations of a group of people, we do not focus on an individual person’s intensity level of expectations instead we value high expectations as one and negative expectations as a minus one. Just because one person may be more passionate than another, positive is one and negative is minus one. Neutral expectations are impossible in a human. If a person has no expectations then they are not a part of the action and therefore should not be evaluated in the first place. Remember in order to evaluate expectations, we can only look at those who are a part of the action. Why would we ask a British citizen if they have high or low expectations of a United States politician in a political race? Their expectations play no role in the future outcome of who is elected because they are not able to vote.
Assume we would like to predict a future re-election of a United States politician applying this concept with only two candidates. Using this method of prediction outcomes, we would focus on the daily or weekly approval ratings of the candidate running. We would only focus on the expectations of people who are able to vote for this politician. If the election is for the Governor of New Jersey, a Virginia resident should not be evaluated. In the United States most, if not all, regional and national media outlets perform weekly or daily approval ratings. Also there are institutions that just poll Americans daily to determine sentiment. We would first find media outlets and or institutions that poll potential voters on approval and choose polls with the least amount of overlap and cover the majority of voters. If this election covers an entire state or larger we must ensure the polls cover most if not all regions. If the election is on a smaller scale such as a small town then we could look for polls of different age groups, affiliations, or regions of the town. Again our goal is to cover the most voters with the least amount of overlap. Then we would evaluate the best time frame for us to use. If the media or institutions mostly conduct daily polls then we should evaluate daily expectations. If equally they conduct daily and weekly polls, then we should just look at weekly numbers. Keep in mind that it is important that we use the most unbiased polls available. Understanding that most media outlets in the United States have some sort of political bias, we should avoid polls from political party driven media sources such as purely Democrat, Republican or other political party news agencies. Now we would gather and evaluate the data. Regardless of whether we are evaluating daily or weekly polls, we would set a value to either positive or negative approval ratings. If more people in a poll approve then disapprove of the candidate then that specific poll on that date is assigned a value of one. If more people disapprove then approve of the candidate then we would assign that poll, on that date, a minus one. Since most of these institutions give these voters a way out of committing to approval or disapproval there is a chance that they will have equal ratings, so in this case we would assign this poll a zero. Although this value of zero contradicts the idea of only two outcomes, the chance of this poll being neutral for every time period evaluated is extremely rare. We use the value of zero so that it does not contribute or take away from the period measurement. After gathering the scores we would tally them for that time period. The following is an example of what measurements could look like:
Poll 1: (name of institution here) If this poll had 48% approve and 44% disapprove than would be assigned a scored: 1
Poll 2: If this poll had 41% approve and 44% disapprove it would be scored: -1
Poll 3: If this poll had 44% approve and 44% disapprove it would be assigned a score of: 0
Poll 4: 1
Poll 5: 1
Poll 6: 0
Poll 7: -1
Poll 8: 0
Poll 9: 1
Poll 10: 1
In the above example the candidate’s expectations have a value of three. This number means that overall the voters are more positive than negative on the candidate. Therefore voters have higher than lower expectations for him or her. This means in this measured period that voters expect more from this candidate.
The above scenario just gives us a snapshot of voter opinion, and does not allow us to predict the future outcome of this election. If this poll was conducted the day before or on the day of the election we might assume that he or she would win. As discussed earlier, we understand the direction of expectations determines the direction of their momentum. To determine direction of expectations we must find the natural limit of the expectation. In the above example, the candidate expectations are valued at three. The best way to determine this candidate’s natural limit is to continue to evaluate the daily or weekly poll data and study the effects of the expectations values from period to period. If this candidate has a one value on the next period then we could start to consider the three as a possible limit. The only way to truly determine this limit is by observing and recording a complete expectations cycle. If the numbers over the next periods are: 1, 2, -1, 0, -2, -3, -1, 1, 3, 1, 0, etc. then we could begin to validate that the expectations at a positive three could be the upper limit of this candidates natural high limit. As discussed earlier, if we touch the limit of high expectations we will touch the lower limit of minus three. If this candidate has a three value now, then we could assume that if the election was four periods away, they would lose the election. Vice versa, if this candidate had touched a minus three in poll expectations and the election was again four periods away, then we could assume they would win the election.
By taking these measurements we are applying a value to a group’s expectations over time. By understanding the amount of pressure that weighs on a possible outcome we can surmise a probable conclusion.
Measuring our own expectations is more difficult than of a group, because of our own involvement. Instead of looking at separate group polls, we are polling our own expectations. If we were attempting to predict whether we will get a promotion at work, we would start by determining who or what will determine whether we get promoted. After we research what action will determine our promotion, we would need to start measuring them. Let’s assume we determined that our promotion depends on the following; our direct manager’s support, their manager’s support, our daily sales, tardiness, organization, good communication skills, company dedication and leadership skills. We would organize these eight expectations and give a plus one for exceeding, a minus one for missing and a zero for meeting them daily. Our tally for each day may look something like the following:
Manager Approval: Our manager is happy with our overall performance from the day before, so we would value this expectation at: 1
Manager’s Manager: Our Manager’s Manager is pleased since our previous week’s sales exceeded their weekly expectations earning us an expectations value of: 1
Daily Sales: We missed our daily sales expectations so we earned a value of: -1
Tardiness: We were on time so we neither exceeded of missed expectations therefore we apply an expectation value of: 0
Organization: Our desk was left unorganized the day before and our manager noticed so this expectation value is: -1
Good Communication Skills: We neither missed or exceeded this expectation thereby applying an expectations value of: 0
Dedication: Since we left on time even though our sales were poor we would assign a expectations value of: -1
Leadership Skills: Because we ignored a teammate who needed our help with a customer on our way to lunch we would record a value of: -1
In the above example our personal expectations for getting this promotion were lowered. Notice how each day, week, hour and minute affects the other expectations. From the above evaluation we would likely not be promoted if the decision was being made that day, but if we have 10 days until the decision is made this may be the spring board to get us promoted. Since most of these expectations are under our direct control, we could use this to help ensure we are on the right expectations slope as we approach the decision day. As in the group expectations measurement we must determine our limits. In the above scenario obviously a plus eight or a minus eight would be the best or worst we could achieve. Since most of these expectations are tied to another, in some way, reaching these extreme values would be highly unlikely. If in the next days leading up to the decision we recorded the following values: -3, -1, 0, 2, 1, 3, 2, -1, -3, we could assume that if there were 5 days left to the decision we made our probability of getting promoted greater. This is because we are at a low point in expectations and exceeding would be easier than failing. If we were at 3 then our chance of failure would be greater than success since our expectations are already at their limit.
In the stock market, equities do not trade on fundamentals alone; otherwise they would only trade several times a year. Instead, stocks trade on what the market expects of them and whether they exceed or miss their expectations. To accentuate this fact we will discuss a simple stock market expectations indicator, that was developed using this theory, which provides a accurate measurement of Wall Street and companies earnings expectations.
Taking measurements in expectations for our personal as well as observed group actions gives us the likely outcome.