The problem
You quote 98 / 102 and someone lifts your offer. You are now short 1 at 102.
If you re-quote 98 / 102, you are equally willing to get shorter as to get flatter. That is wrong - you already have risk you did not want.
Skewing
Move the whole quote in the direction that reduces your position.
Short? Quote higher - say 99 / 103. Now buyers find you less attractive and sellers find you more attractive, so the next trade is more likely to flatten you than to extend you.
Long? Quote lower.
The size of the skew should scale with how large your position is relative to what you are comfortable holding.
Skewing versus widening
Two different tools for two different problems, and interviewers like the distinction:
- Skew (shift the mid) when you have inventory to manage.
- Widen (increase the spread) when you face more uncertainty or adverse selection.
You may need both simultaneously - after a large fill from someone who looks informed, you would skew away and widen.
Why inventory risk grows non-linearly
Holding twice the position does not merely double the risk in practice: your ability to exit without moving the market degrades, and the position becomes correlated with the reason you acquired it (usually informed flow). Desks therefore run hard position limits rather than pricing risk linearly.
The interview form
An interviewer will trade against you on the same side three or four times. What they are watching:
- Do you know your position? (Say it aloud after every fill.)
- Do you skew?
- Do you skew in the right direction? Getting this backwards is common under pressure and is fatal.
- Do you eventually refuse to keep trading at all? Saying "I'm at my limit, I'll only quote a buy side now" is a strong answer.
The mantra
After every trade, state your position and your new market. "I'm short 300 at an average of 101.5, so my market is now 100 / 104."
Practise with the make a market game, which tracks inventory and P&L for you.