We should have another look at what paying "market rate" means.

It sounds reasonable. Safe, even.

I'm starting to think it's a trap.

Market rate assumes interchangeability. One analyst equals another. For average performers, this holds. For A-players, it collapses.

The real question isn't "what does the market pay?" It's "what does it cost if this specific person leaves?"

When an A-player leaves, you pay three compounding costs.

The lame duck phase. Mental checkout to last day. Projects only they could drive begin to drift.

The recruitment black hole. Six months to find another A-player. A year is common. Your remaining team absorbs the load. Your next-best people start questioning if this is the environment they signed up for.

The onboarding lag. Even genuine replacements need 12-18 months to internalise institutional knowledge. You're funding a year of learning while competitors operate at full capacity.

When Apple bought NeXT for $500 million, they were paying to rehire Steve Jobs. Brad Jacobs calls it the deal of the century. That's the dynamic range of elite talent.

A-players attract A-players. They set the intellectual tempo.

When a linchpin departs, the tribe dissolves. One resignation becomes two. Two becomes an exodus.

Paying $150,000 for an analyst isn't about that analyst. It's an insurance policy on the stability of your entire high-performance unit.

The most dangerous thing about losing an A-player isn't their absence. It's their destination.

They join a competitor. Bring your blueprint. Know your bottlenecks. They're incentivised to a rival's problems. The same ones that politics prevented them from solving with you.

Michael Jordan understood this. He paid his trainer Tim Grover with an explicit clause: "I don't pay you to train me. I pay you not to train anyone else."

Jordan wasn't buying training. He was buying exclusivity.

As major events approach, Olympics, World Cups, the supply of elite practitioners evaporates. Local benchmarking creates 'fair' contracts at obsolete prices.

Meta and OpenAI are offering $1 billion packages for top-tier AI researchers. To a layperson, absurd. To a CEO, rational; if one researcher determines your survival, $1 billion is a steal.

Sport operates at different scales. Same principle.

Trading $2 million in results for $100,000 in savings isn't frugal. It’s self-sabotage.

You cannot overpay for talent that creates a value multiple. You can only under-invest and absorb the consequences.

The industry may not value an analyst at $150,000. But when you have one of the best, someone whose work changes direction, not just reports on it, can you afford not to lock them in?

The arithmetic makes the answer clear.