Insurance Expected Value: Why I Cancelled My Laptop Protection Plan Using Math

Three years ago, I bought a mid-range laptop for my university work and walked out of the store having also agreed to a two-year protection plan worth roughly 15% of the purchase price. The salesperson was persuasive, I was tired, and the phrase “accidental damage coverage” sounded comforting. Last year, when it came time to renew, I did something different: I actually ran the numbers. What I found changed how I think about almost every insurance product I encounter.

Related: index fund investing guide

This is not a post telling you that all insurance is a scam. It is not. But some insurance products — and consumer electronics protection plans sit near the top of this list — are structured in ways that make them deeply unfavorable purchases for most buyers. Understanding why requires spending about ten minutes with a concept called expected value. After that, you will never sign up for a protection plan on impulse again.

What Expected Value Actually Means

Expected value (EV) is a way of calculating the average outcome of a decision when that decision involves uncertainty. It is used everywhere from actuarial science to poker strategy to public health policy. The formula is simple: multiply each possible outcome by the probability of that outcome occurring, then sum all those products together.

Formally: EV = Σ (Probability of outcome × Value of outcome)

If you flip a fair coin and win $10 on heads but lose $5 on tails, the expected value is (0.5 × $10) + (0.5 × −$5) = $5 − $2.50 = $2.50 per flip. Over many repetitions, you expect to profit $2.50 per coin flip on average. That is a positive expected value bet, and you should take it every time if you can.

Insurance works in the opposite direction — deliberately. When you buy insurance, you are accepting a negative expected value transaction. You pay a premium that is mathematically higher than the expected payout you will receive. The insurer must charge more than the expected cost of claims, otherwise they cannot cover operating costs, pay employees, or generate profit. This is not deception; it is arithmetic (Kahneman, 2011). The question is never “does this insurance have positive expected value?” It doesn’t. The question is “is the negative expected value worth paying for the protection it provides?”

For life insurance, health insurance, and catastrophic coverage, the answer is often yes — because the downside scenario (death, medical bankruptcy, losing your home) is genuinely catastrophic and not something a normal person can self-insure against. For a laptop protection plan on a $1,200 machine, the math tells a very different story.

The Numbers Behind My Protection Plan

Let me show you exactly what I calculated. My laptop cost approximately $1,200. The two-year protection plan cost $180, which works out to $90 per year. The plan covered accidental damage, hardware failures, and certain liquid spills. It did not cover theft.

To calculate the expected value of buying this plan, I needed two things: the probability that I would make a claim, and the expected value of that claim if I did.

Step 1: Probability of a claim. Consumer electronics failure and accidental damage rates are actually well-documented in industry and academic literature. SquareTrade (now Allstate Protection Plans), which underwrites a large share of retail electronics protection plans in the United States, published data suggesting that roughly 30% of laptops experience some form of damage or failure over a three-year period (SquareTrade, 2010). However, that headline figure is somewhat misleading for our purposes. First, many of those failures occur in year three, after a two-year plan expires. Second, most manufacturer warranties already cover hardware defects for the first year, meaning the incremental protection a third-party plan provides in year one is minimal. A more conservative estimate for the probability of a claim that would be covered exclusively by the protection plan and not the manufacturer warranty — in years one and two — sits closer to 10–15%.

I used 15% as my estimate, erring on the generous side for the protection plan. [4]

Step 2: Expected payout if a claim occurs. This is where most people make a critical error. They assume that if something goes wrong, the plan will simply replace the laptop. In reality, protection plans typically repair the device first. If the repair cost exceeds a threshold, they may replace it — but often with a refurbished unit or a voucher equivalent to the depreciated value. For a laptop that has been owned for 12–18 months, that depreciated value might be 50–60% of the original purchase price. So the realistic expected payout in a claim scenario, not the best-case scenario, is closer to $600–$700 for a $1,200 laptop. [1]

I used $650 as my estimate. [2]

Step 3: Running the calculation. [3]

Probability of a meaningful claim: 15% (0.15)
Expected payout if claim occurs: $650
Expected value of the plan’s benefit: 0.15 × $650 = $97.50
Cost of the plan: $180
Net expected value: $97.50 − $180 = −$82.50

Buying that protection plan had an expected cost of $82.50. Not a huge sum, but a reliable one. And this calculation was done with generous assumptions in favor of the plan. If I used a 10% claim probability or a $550 expected payout, the loss grows considerably worse. [5]

Why Retailers Push These Plans So Hard

If you have ever noticed that the pitch for a protection plan seems almost more rehearsed than the pitch for the product itself, you are not imagining things. Protection plans and extended warranties are extraordinarily profitable. Consumer Reports has repeatedly noted that retailer margins on protection plans can exceed 50–70%, meaning that for every dollar consumers collectively pay into these plans, they receive back only $0.30–$0.50 in claims (Consumer Reports, 2017). This is a far worse return ratio than, say, automobile insurance or home insurance, which typically have loss ratios of 60–80%.

The reason consumers keep buying them despite the poor expected value comes down to well-documented cognitive patterns. Loss aversion — our tendency to feel losses roughly twice as intensely as equivalent gains — makes the prospect of a broken, unreplaced laptop feel disproportionately threatening (Tversky & Kahneman, 1992). We anchor on the retail price of the laptop when imagining a loss, not the depreciated value or the realistic claim payout. And we buy these plans in an emotionally heightened state (at the register, during the excitement of a new purchase) rather than through calm deliberation.

Salespeople are also trained to make the pitch when you are least likely to do the math. “What if you drop it?” they ask, just as you are imagining all the productive work you will do on your shiny new machine. The question triggers vivid mental imagery of disaster, which psychologists call the availability heuristic — we judge the probability of an event by how easily we can picture it happening, not by its actual frequency (Kahneman, 2011).

The Self-Insurance Alternative

Here is what I did instead of renewing the protection plan: I created a dedicated savings category in my budget called “electronics replacement.” Every month, I set aside the equivalent of what I would have paid for protection plans across all my devices — laptop, phone, tablet — combined. For me, that was about $25 per month, or $300 per year.

After two years, I had $600 sitting in that fund, earning modest interest. That is almost exactly the depreciated replacement value of my laptop. If it had broken during that time, I would have been covered. If it had not broken — which, statistically, is the more likely outcome — I still have $600. This approach is sometimes called self-insurance, and it works well when three conditions are met: the potential loss is affordable, the probability of loss is moderate or low, and you have the discipline to actually set the money aside rather than spend it.

For most knowledge workers earning a stable income, these conditions are met for consumer electronics. They are explicitly not met for health emergencies (potential losses can be catastrophic and unpredictable), disability (loss of income is not self-insurable for most people), or property destruction (a house fire is not something you save your way out of). The self-insurance logic applies specifically to items where the worst-case financial scenario is painful but survivable.

When Insurance Actually Makes Mathematical Sense

The expected value framework does not lead to the conclusion that all insurance is irrational. It leads to a more nuanced principle: insurance makes sense when the downside scenario would be financially catastrophic relative to your assets, and when you cannot afford to absorb the loss out of cash flow or savings.

Economists and behavioral researchers have formalized this through the concept of risk aversion and marginal utility. A dollar means more to you when you have very few dollars than when you have many. This diminishing marginal utility of money means that a large loss (say, a $200,000 medical bill) is not just twenty times worse than a $10,000 loss — it can be functionally catastrophic in a way that destroys your quality of life entirely. Paying a negative expected value premium for protection against that outcome is rational because you are buying something real: the elimination of a life-altering risk (Thaler & Sunstein, 2008).

Apply this lens to different insurance products:

Health insurance: Rational even at high premiums. A serious illness or accident can generate hundreds of thousands in bills. Most people cannot self-insure against this.
Term life insurance (with dependents): Rational. The financial devastation to a family without income support is genuinely catastrophic.
Comprehensive car insurance on a financed vehicle: Often required by the lender, and the vehicle represents a large enough asset that coverage makes sense.
Laptop protection plan: For most working adults with any savings buffer, the expected value is too poor and the worst-case scenario too manageable to justify.
Travel insurance for a cheap domestic flight: Almost never rational. The ticket price is small, refund policies often exist, and the “loss” is easily absorbed.

How to Run This Calculation Yourself

You do not need a spreadsheet. You need four numbers and about five minutes. Here is the process I now use before agreeing to any optional insurance product:

1. Find the realistic failure probability. Do not rely on the salesperson’s estimate. Search for independent data. SquareTrade data, Consumer Reports reliability surveys, and academic studies on product failure rates are all accessible online. For newer product categories, manufacturer return rates are often published in SEC filings or consumer advocacy reports.

2. Calculate the realistic payout, not the best-case payout. Read the fine print. What does the plan actually cover? What is excluded? Does it replace at retail value or depreciated value? Are there deductibles? The realistic payout is usually meaningfully lower than what the pitch implies.

3. Multiply probability by payout. This is your expected benefit from buying the plan.

4. Compare to the premium. If the premium exceeds the expected benefit by a large margin, and if you could absorb the loss out of savings, decline the coverage. If the gap is small or if the loss scenario would genuinely be catastrophic for your financial situation, the purchase may be justified despite the negative expected value.

The critical mental shift is moving from “what if something goes wrong?” to “what is the mathematical cost of this protection, and is it worth that cost given my specific financial situation?” The first question triggers anxiety and availability bias. The second question is tractable and answerable.

What Changed After I Ran the Numbers

Cancelling that protection plan was a small financial decision. But applying expected value thinking more broadly has compounded meaningfully. I now keep a running total of optional insurance and warranty products I have declined over the past three years — phone screen protection plans, appliance extended warranties, rental car excess coverage when my credit card already provides it — and the total premium savings exceeds $800. More importantly, I have had no uncovered losses during that period, which is consistent with the probabilities.

I still carry robust health insurance, renter’s insurance (where the premium is low and the downside — losing everything to a fire — is devastating), and adequate coverage on my car. These pass the expected value and catastrophic loss tests. The difference is that I now choose insurance products with the same deliberateness I apply to any other financial decision, rather than defaulting to “yes” whenever someone offers me protection in a moment of post-purchase vulnerability.

Expected value is not a perfect framework. It assumes you have reliable probability estimates, which you often do not. It assumes rationality in the face of uncertainty, which our brains resist. But as a forcing function — a way of demanding that you think quantitatively before you commit money to a product — it is remarkably effective. The calculation does not need to be precise to be useful. Even a rough expected value analysis will reveal that most consumer electronics protection plans ask you to pay two to three times what you can reasonably expect to get back. Once you see that clearly, the salesperson’s pitch stops sounding like protection and starts sounding like what it actually is: a highly profitable product designed around your fear of loss, not around your actual financial interest.

Last updated: 2026-03-28

Your Next Steps

Today: Pick one idea from this article and try it before bed tonight.
This week: Track your results for 5 days — even a simple notes app works.
Next 30 days: Review what worked, drop what didn’t, and build your personal system.

References

Li, Q. (2025). Self-Protection and Insurance Demand with Convex Premium Principles. North American Actuarial Journal. Link
Wang, H., et al. (2026). Robust Investment-Driven Insurance Pricing under Correlation Ambiguity. arXiv preprint arXiv:2603.18969. Link
Cameron, C. (n.d.). Theory of Health Insurance. UC Davis Economics. Link
NAIC Center for Insurance Policy and Research. (2025). Modeling Behavioral and Attitudinal Drivers of Life Insurance Selection. NAIC CIPR Working Paper 25-04. Link
Krvavych, Y. (2007). Enhancing insurer value through reinsurance, dividends and capital management. Actuaries.org. Link
KPMG. (2024). Insurance – Assessing the impact (IFRS 17). KPMG Insights. Link

What is the key takeaway about insurance expected value?

Evidence-based approaches consistently outperform conventional wisdom. Start with the data, not assumptions, and give any strategy at least 30 days before judging results.

How should beginners approach insurance expected value?

Pick one actionable insight from this guide and implement it today. Small, consistent actions compound faster than ambitious plans that never start.

Insurance Expected Value: Why I Cancelled My Laptop Protection Plan Using Math