The Journey Ahead — Part 3: The Price of Being Wrong

There is a particularly dangerous phrase in the watch business:

“That’s a good price.”

I have said it. I have heard other people say it. I still say it occasionally, but I am slowly learning that it is almost a meaningless statement.

A good price compared to what?

Compared to the cheapest one currently listed on Chrono24? Compared to what somebody is asking on eBay? Compared to MSRP? Compared to what another dealer wants? Or compared to the price at which somebody will actually take money out of his pocket and buy the watch from me?

Those are very different numbers, and somewhere along this journey, I realized that learning the difference between them may be one of the most important skills in this business.

When I first started buying watches more seriously, my thinking was still largely that of a collector. I would research a watch, look at the market, find what appeared to be an attractive example, and negotiate. If watches like it were listed for $5,000 and I could buy one for $4,000, that felt like a very good deal, and perhaps it was, but it did not necessarily mean that I had bought a $5,000 watch for $4,000.

I might simply have bought a $4,000 watch.

That sounds obvious when written down. It is considerably less obvious when a beautiful watch is sitting in front of you and every listing you can find seems to confirm that you have discovered a bargain.

Asking Is Not Selling

One of the first things I had to unlearn was my reliance on asking prices. The internet gives us an extraordinary amount of information about watches. Within minutes, I can find dozens of examples of almost any reasonably common reference. Chrono24, eBay, dealer websites, Facebook groups and forums collectively create the impression that the market is completely transparent.

It isn't.

What they mostly show us is what sellers want. What I increasingly wanted to know was what buyers actually did.

That distinction led me deeper into auction data, particularly Grailzee. It also led to many hours of spreadsheets, statistical experiments and conversations with my increasingly patient artificial-intelligence research assistant.

Let me digress for a moment here, because there is an important distinction to make. I do not rely on my AI friend to make decisions for me. I use it as a tool to mine data, run calculations, test assumptions and, occasionally, challenge my own conclusions. Like any powerful tool, however, it becomes dangerous when used outside its strengths—or when its output is accepted without supervision. Ultimately, the decision, and the money at risk, are still mine.

I wasn't trying to predict the exact price of the next auction. Markets don't work that way, and watches certainly don't. I was trying to answer a different question:

What happens if I'm wrong?

That changed everything.

No-reserve auctions became particularly interesting to me because they provide something ordinary listings cannot. A seller can ask anything, a reserve auction can fail, but when a legitimate no-reserve auction ends, somebody owns the watch, and that makes those transactions extraordinarily useful.

They don't necessarily tell me what a watch is “worth.” There may not even be one correct answer to that question. Condition matters. Age matters. Box and papers matter. Timing matters. Photography matters. Buyer enthusiasm matters. Sometimes two people simply decide they both want the same watch on the same evening. But enough transactions begin to describe a range, and near the bottom of that range is something extremely valuable to someone trying to trade watches with limited capital: a risk floor.

I stopped looking only at the average selling price and started becoming increasingly interested in the ugly outcomes. What did the weakest examples sell for? How frequently did prices fall below a particular number? What happened to full sets? Did newer examples actually command meaningful premiums? How wide was the distribution? Was I looking at twenty meaningful transactions or three?

Those questions don't tell me how much money I am going to make, but they tell me how much trouble I might be in if my assumptions are wrong, and that is far more useful. There is an old dealer expression that profit is made when you buy, not when you sell. I used to regard that as one of those pleasantly simplistic business sayings that sounds good over dinner.

I understand it much better now.

Once I own a watch, many things are outside my control. The market may soften. Another dealer may suddenly list three identical examples. A model may fall out of favor. My auction may end on a terrible evening. The two people who would normally fight over the watch may both be on vacation. I can take excellent photographs. I can write an accurate description. I can present the watch properly. I can answer questions quickly and behave professionally. But I cannot manufacture a buyer. What I can control is the price I pay, and that is where most of the risk is created.

If the realistic downside on a watch appears to be $3,500 and I buy it for $3,700, I have one kind of problem. If I buy the same watch for $3,000, I have an entirely different business. The watch hasn't changed. My risk has.

And that realization has gradually changed the way I negotiate. This is also where being both a collector and an aspiring trader becomes complicated.

I like watches.

That may seem like a rather obvious admission from someone building a watch business, but it creates a genuine problem. I can rationalize almost anything. A beautiful dial becomes “strong market appeal.” An unusual reference becomes “scarcity.” A watch I have always wanted becomes “inventory I wouldn't mind holding.” Sometimes those statements are even true, but I have learned to ask another question:

If I didn't personally like this watch, would I still buy it at this price?

That question can be surprisingly uncomfortable. There are watches I would happily own but should not trade. There are watches I would happily trade but have little desire to own. And occasionally there is the wonderful intersection where both are true. Those are dangerous too, because I can become more forgiving with the numbers.

Recently I looked at a watch that I genuinely liked and could probably have bought at what appeared to be an attractive price. The data suggested that the downside was manageable. I could wear it carefully while offering it for sale and be perfectly happy if it took several months to find the right buyer. That matters, but it doesn't magically turn inventory into a personal collection.

If I begin using “I wouldn't mind keeping it” as an excuse every time the numbers are marginal, I will eventually have a wonderful collection and no working capital.

Capital Has a Cost

This may be the biggest lesson so far. When capital is limited, the cost of a watch is not simply what I paid for it. It is also everything else I cannot buy while that money is sitting on my wrist—or in the safe. An $8,000 watch that might produce a $700 profit sounds better than a $3,000 watch that might produce $300, until I realize that the $8,000 watch might take four months to sell while the $3,000 watch could turn twice. Or that the larger position exposes me to considerably more downside. Or that while I am waiting for the expensive watch to sell, three better opportunities might appear and I no longer have the capital to pursue them.

I recently passed on precisely that kind of opportunity. I liked the watch. I thought I could buy it well. I thought there was money in it, and I let it go. Six months from now, with more working capital, I might make the opposite decision. That wasn't fear; it was allocation.

I suspect learning to walk away from good watches may ultimately be more important than learning how to find them.

Then I Did Something Slightly Crazy

Eventually, spreadsheets are not enough. A model can tell me what has happened. It can estimate probabilities. It can help establish reasonable boundaries. But sooner or later I have to find out whether any of this works when the watch belongs to me. This is something I know very well from my real job. I can design and simulate a circuit inside out, run simulations until my computer's CPU begs for mercy, and eventually convince myself that it works perfectly. Then I put that circuit into the actual system and discover that its behavior has changed—or that it doesn't work at all.

At some point, you have to stop simulating and start testing.

So I bought one of my analysis subjects. A very good-condition, full-set Omega at $3,700. Not because I expected a spectacular profit. Quite the opposite. My anticipated profit was roughly $200 after fees. Two hundred dollars is hardly an exciting return for the work involved in sourcing, inspecting, photographing, listing, shipping and assuming the risk of owning a luxury watch. But the watch wasn't really the investment.

The experiment was.

I listed it at reserve and now I get to watch what happens. If it sells immediately, wonderful. If the reserve isn't met but buyers begin negotiating afterward, that is useful. If it fails completely, that is useful too because for the first time, I am not merely studying somebody else's transaction history. I am generating one of my own.

And that may be worth considerably more than $200.

Learning to Lose Small

I think this is ultimately where my thinking has changed the most. At the beginning of this journey, I was primarily trying to figure out how much I could make. Now I spend much more time thinking about how much I could lose. That may sound pessimistic, but I don't think it is.

If I can consistently structure purchases so that being wrong is merely annoying rather than catastrophic, I can afford to keep learning. Some watches will outperform expectations and some will disappoint. Some will sit far longer than expected, and occasionally I will undoubtedly make a terrible decision despite all the analysis in the world. The objective isn't to eliminate those mistakes. It is to survive them cheaply. Perhaps that is what separates speculation from a business.

A speculator asks:

How high can this go?

A trader eventually learns to ask:

Where is the floor, and can I live with it?

I am still learning how to answer that question, but increasingly, when somebody shows me a watch and says, “You can buy this for $4,000. That's a great price,” my first reaction is no longer excitement. It is another question.

What happens if I'm wrong?

And, strangely enough, I think that question is making me a better buyer.