Hard to Price

In Liar Game, Akiyama wins every round the same way. He does not outplay his opponents by being a better liar. He reads what they want — money, safety, trust, survival — and uses it to predict their moves before they make them. His opponents cannot do the same to him, because they cannot tell what he wants. The game is not about deception. It is about information asymmetry — and the advantage is not just being hard to read. It is the gap between how well you model them and how well they model you. Maximise your read of others; minimise theirs of you. The delta is the edge.

A friend asked me how I think in adversarial situations. I told him: my first thought is always what does this person want, and how do I take it away from them. He was shocked. I said: sufficient analysis of someone else's wants gives you insight into how they operate. Being able to say "no I don't need it" is a very useful skill.

He thought the point was defence — nothing to use against you. That is close, but it misses the mechanism. Akiyama is not invulnerable. He is hard to price. His opponents cannot calibrate their plays against him because they do not know what he is optimising for. The same dynamic operates in every employment relationship you have ever had. Nobody tells you the rules or that you are playing.

I default to Akiyama. This essay will teach you to do the same. It will also tell you why that is not enough — but you need the teeth before the lesson about teeth means anything.

Different games, different chapters. Same principle throughout.

Your employer runs performance reviews, salary benchmarking, leveling frameworks, exit interviews. These are not administrative overhead. They are legibility infrastructure — tools for modelling your incentive structure so they can pay you precisely enough to retain you and not a dollar more.

Your reservation wage. Your career ambitions. Your risk tolerance. Your identity attachment to the role. Every piece of information you volunteer is a data point in a pricing function that works against you.

Platforms model your willingness to pay. Lenders model your creditworthiness. Recruiters model your desperation. The system converts information about you into leverage over you. That is what it does. That is what it is for.

The rational response is to stop cooperating with the pricing function. Be legible about one thing — the value you create — and opaque about everything else. Your mechanism. Your capabilities. Your price.

When your employer can see what you produce but cannot model what it takes to lose you, they face an asymmetric problem. Overpaying you by 20% is a line item nobody audits. Losing you because they underbid is a crisis — the client relationships, the institutional knowledge, the replacement search. The loss function is not symmetric. Under uncertainty, they bid high.

Legible employees get priced in a tight band around their market value. Illegible employees get priced under uncertainty — high variance, wide band. But the variance is not symmetric either, because the employer's downside is not symmetric.

Valuable illegible employees get paid a lot more than they should.

The employer's rational response is to collapse the uncertainty. Performance reviews. KPI frameworks. Salary bands. Comp benchmarking. These are attempts to make you legible so they can price you tighter.

The problem is Goodhart's Law. The moment you tie compensation to measurable outputs, every employee optimises the metric at the expense of everything the metric does not capture. You have replaced illegible employees extracting surplus with legible employees destroying value — and on the dashboard, it looks like high performance.

The opaque-but-valuable employee is a better outcome for the employer than the system the employer builds to defeat them.

Your manager says I'll cut your pay. You say go ahead. You make plans to leave. The threat only worked if they correctly modelled that your survival or your identity was tied to the position. Most employment relationships run on that assumption. Most of the time it is correct.

When it is not correct — when your position is genuinely orthogonal to what the employer can offer or threaten — the standard playbook fails. Not because you are hard to read. Because your walkaway point is real, and their toolkit does not reach it.

The people who are hardest to deal with are not unpredictable. They are predictable — you know they will walk — and you cannot beat their price.

Be legible about your value. Be opaque about everything else. In a market like Singapore's, where no union does this work for you, it is the strongest lever you have. The worst that happens is you end up where you would have been anyway.

Your friend's success is not just good news. It is a price signal that enters every negotiation you have without you saying a word.

When employees compete with each other, the employer's legibility infrastructure works. When employees cooperate, it breaks — and it breaks in the employer's favour too, because the dysfunction disappears. The only player who loses is the pricing function.

But Liar Game does not end with Akiyama winning through superior information control. It ends with Nao — the trusting one, the naive one, the player everyone exploited — being the reason anyone gets out at all. Akiyama's opacity keeps him alive. Nao's irrational trust is what actually saves people. The manga's thesis is not that the best player wins. It is that the game itself is the enemy. Your opponents are not the other people struggling. Your opponents are the structures that force you to compete when there is enough for everyone.

The real winning strategy is not information asymmetry. It is this: stick by one another.