The article and all the comments I’ve seen really miss the essence of winner’s curse. Let’s say there’s an auction for a gold bar. Everyone has to use their ow

The article and all the comments I’ve seen really miss the essence of winner’s curse. Let’s say there’s an auction for a gold bar. Everyone has to use their own scale. Our scales don’t have bias, that’s easy to detect and fix. But, they do have noise. Let’s say +/-1%, all equally likely. There are 50 people at the auction. The bar is 100g. The two highest readings are going to probably be near 101g, and those two are going to bid each other up. The winner will have overpaid. Simulating this, the mean second highest estimate will be 100.96. The essence of this is that valuations have a randomness to them. It doesn’t have to be a scale. It could be a model or just the ideas of a person. Because of the ways auctions work, the highest valuation gets chosen. So, the noise in your valuations doesn’t cancel out. The winner is cursed by overpaying. You can correct for this. There are formulas. But there are also things you can use intuitively. This is only for cases where you care about the value of something, like you may resell it. Bidding on a taxi to make it to a wedding doesn’t apply. Never bid what you think is a fair price, like you might in an open market. You need to be getting discount. The more bidders and the noisier the valuation, the higher the discount. If you’re bidding on something where it could be worth 1/2 or 2x easily and there are 100 others, you need to think you’re getting a screaming deal. Cause the most likely reason you’re the higher bidder is you were unluckily high in your valuation. Again, if you have a reason to bid high (that was my dad’s corvette!) this doesn’t apply.