06aFear & Greed

The VIX measures how much movement the options market expects in the S&P 500 over the next 30 days. It is quoted as an annualized percentage, and because option premium is priced off expected movement, the VIX is effectively a read on what selling options pays right now.

The gauge on the screener and the AI Trader page turns that single number into six named levels. Each level carries a deployment range: the share of a book the AI Trader aims to have working in open positions while conditions look like that.

The gauge runs the opposite way to instinct. A red dial is not a warning to stop — for a premium seller it marks the richest conditions of the cycle. The colors follow the familiar fear/greed convention, but in this strategy fear is where the money is made.

The six levels

Levels are contiguous: every possible VIX reading falls into exactly one. The dollar columns show the midpoint of each range applied to the AI Trader's two books — $50,000 weekly and $250,000 monthly.

LevelVIXDeployedWeeklyMonthly
Extreme greed< 1430%40%$17,500$87,500
Greed14–1640%50%$22,500$112,500
Balanced16–2050%60%$27,500$137,500
Moderate fear20–2560%70%$32,500$162,500
High fear25–3070%80%$37,500$187,500
Extreme fear≥ 3080%90%$42,500$212,500

Where a reading lands within a range depends on which way the VIX is moving. A falling VIX means the scare is passing while premium is still elevated — historically the most favorable corner of the cycle — so the weekly book sits at the top of the range. A VIX spiking hard means the move is not finished, and it sits at the bottom. The monthly book uses the middle of each range: at 30–45 days to expiry there is enough runway to absorb a spike, so direction carries little information at that horizon.

Why more exposure into fear

Selling a cash-secured put means being paid to accept the obligation to buy a stock at a set price. What that obligation pays is a function of expected movement. When the VIX is at 12, the market expects very little to happen, and the premium reflects it — the same strike that pays $1.40 in a fearful market might pay $0.35 in a calm one, for the same capital tied up for the same week.

Two things happen when volatility rises. Premium expands, so each dollar of collateral earns more. And strikes move further from the current price for the same probability of assignment, so the cushion widens at the same time the payment increases. For a strategy whose entire income is premium, that combination is the opportunity, not the risk.

The reverse is the real trap. A becalmed market feels safe and pays almost nothing, which tempts sellers into strikes far too close to the money to make the week worthwhile. The gauge exists partly to name that condition out loud: extreme greed is when the wheel is least worth running at size.

Elevated volatility is not free money. High VIX means the market genuinely expects large moves, and assignment becomes more likely precisely when it is least comfortable. Deploying more capital into fear also means owning more stock if those moves go against you. The ranges here describe how the AI Trader allocates — they are not a recommendation for any individual account, and no allocation model removes the risk of being assigned well below your strike.

How the AI Trader uses it

The deployment range is a target for the total capital at work, not a per-trade size. Each Monday the select cron reads the VIX, finds the level, compares the target against what the book already has deployed, and only opens new positions up to the remaining room.

When a book sits above its target — usually because volatility fell while positions were still open — nothing is sold. The cron simply stops opening new positions and lets existing ones expire, which walks deployment back down on its own. Forcing exits to hit an allocation number would book real losses to satisfy a percentage, so it does not do that.

Two limits sit above the model at all times. No single position may exceed 15% of a book, so a high target can never become a concentrated bet. And each book has a fixed number of slots — six weekly, eight monthly — so the target is reached across several names or not at all.

The gauge, the screener strip, and the two trading crons all read the same set of levels from one place. If the gauge says a level targets 70–80%, that is the number the bots are actually sizing to — not a separate illustration that could quietly drift out of step with them.