Pelican Lake, Manitoba

Let experience inform my judgment, evidence correct it, and reason guide what follows. The wise man holds firmly to his principles, but never so tightly to his opinions that reality cannot change them.

A thought: A pattern is evidence, not proof. Understand the relationship, question the cause, and never surrender your judgment to the certainty of a prediction. Correlation tells us whether two variables move together and how strongly. Regression translates those relationships into predictive equations. Multiple regression incorporates additional predictors and may improve how much variation a model explains. But a strong correlation, a statistically significant result, and a high R² do not independently establish causation. This lesson may be one of the most important in the statistics because it deals with something I have spent most of my adult life doing: trying to understand relationships and anticipate consequences. Farming required prediction. We looked at historical yields, rainfall, soil conditions, input costs, commodity prices and interest rates. We tried to establish relationships that would help us make better decisions. But one thing farming taught me was that relationships are not guarantees. A particular fertilizer might produce consistently better yields. That relationship could be strong. Yet a different growing season could produce an entirely different outcome. The challenge was to understand which variables mattered, how they interacted, and how much confidence I should place in my expectations. That is exactly what regression attempts to accomplish. I am particularly interested in the distinction between correlation and causation. I see people make this mistake constantly, especially when discussing economics and politics. Government spending rises while employment increases. Does that prove the spending created the jobs? Taxes are reduced while investment increases. Does that prove the reduction caused the investment? Neither conclusion necessarily follows from the correlation alone. I may have strong views about what government policy should accomplish, but those views do not relieve me of the responsibility to examine competing explanations. Economic conditions rarely depend on one variable. Interest rates, international demand, productivity, regulations, labour availability and investor expectations can all influence the same outcome. A simple correlation might tell me something interesting is happening. It does not tell me that I have discovered the cause. I should be just as demanding of evidence that supports my political beliefs as I am of evidence that challenges them. That is intellectual consistency. R² also speaks to something I have learned through experience. We naturally prefer simple explanations. A successful business must have an excellent manager. A struggling economy must have a bad government. An investment that performs well must have a superior strategy. There may be truth in those statements, but reality is generally more complicated. Adding experience dramatically improves the model. This reminds me that good judgment requires examining enough of the relevant variables to understand the situation properly. I also see an important connection to investing. I have always been interested in generating dependable income. But a historical relationship between an investment's yield and its past performance does not prove that the distribution is sustainable. A regression model might show an impressive historical pattern, but changes in market conditions, volatility, interest rates or the underlying strategy can alter the relationship. A model is useful because it organizes evidence and makes assumptions explicit. It becomes dangerous when I mistake its predictions for certainty. Epictetus taught that we should examine our impressions rather than immediately accept them. Statistics offers a mathematical way to practise that discipline. When I see two things moving together, my first question should not be how I can use that observation to support an argument. My first question should be whether the relationship is genuine. Then I should ask whether it is significant, whether other variables explain it, whether the model meets its assumptions, and whether its conclusions remain valid under different conditions. There is a kind of humility required here. I have lived long enough to know that experience does not make me infallible. Sometimes it makes my assumptions more deeply established. That can be useful when those assumptions are correct. It can also make me resistant to evidence that challenges them. The purpose of education at this stage of my life is not simply to accumulate more information. It is to improve the quality of my judgment. Correlation and regression give me additional tools to do that. They help me distinguish patterns from explanations, evidence from certainty, and prediction from knowledge. I cannot control whether the future follows the pattern I expect. I can control how carefully I examine that pattern and how willing I am to change my conclusions when new evidence appears. That is both statistical discipline and practical Stoicism.

A difference is not an explanation, and a result is not the whole truth. Examine the conditions, question the assumptions, and let reason determine the judgment.

Borrowing can create value when the money is invested productively and generates a return greater than its cost. Borrowing to finance ongoing consumption simply pushes today's bills onto tomorrow's taxpayers. I've spent a lifetime running businesses and farming. I understand the difference between productive debt and debt that merely keeps the lights on. The real question isn't how much you can borrow. It's what you have to show for it, what it costs to service, and who's ultimately responsible for paying it back. A government that confuses borrowing capacity, like NDP/Liberals, with economic prosperity is fooling itself, and eventually the taxpayers get stuck with the bill, through higher taxes, reduced public services, inflationary pressures, or a heavier debt burden passed on to future generations.

Reality owes nothing to my expectations. Observe what is, question what I assume, and never confuse a significant difference with a meaningful truth.

“4th” Reich seems perfectly comfortable condemning concentrated power when Trump exercises it, then advocating much greater government control when Musk is the target. You cannot defend economic freedom only when you approve of the people exercising it. Reich is entitled to dislike Musk, criticize his political influence and question government contracts. But using the power of the state to take over a successful private enterprise because you distrust its owner is a dangerous precedent. The real test of your principles is whether you defend them when the person benefiting is someone you despise.