Most traders have the entry problem. They find a good setup, size it right, enter at the pivot. Then the stock runs 30%, pulls back 10%, and they panic-sell before the real move begins.
William O'Neil held Mosaic for 581%. He held Schwab through the dot-com NASDAQ crash — the index dropped while Schwab kept climbing. He held Netflix through 2009–2011 for 750%. He did this consistently. The question is how.
The answer isn't willpower. It's a rule-based system applied to the RS line.
The three exit signals (Mike Webster's formulation)
The RS line has two moving averages: - Short MA: 21-day (daily chart) or 8-week (weekly chart) - Long MA: 50-day (daily chart) or 21-week (weekly chart)
| Signal | Trigger | Action |
|---|---|---|
| Quick | RS line drops below the short MA | Sell one-third of core position |
| Quicksand | RS line falls below the level where it first broke above the short MA | Sell another third (down to one-third remaining) |
| Grateful Dead | RS line crosses below the long MA | Sell the rest |
Webster's framing: "I designed this specifically to fix a flaw of mine — not being able to hold things for a long period of time because I get wiggled and wobbled out by looking at the price action and intraday action."
The system forces you to hold until the RS line tells you the stock is genuinely losing relative strength versus the market — not just because the price pulled back.
Why the RS line and not the price?
Price is noisy. A 10% pullback in a stock running 300% looks terrifying but is structurally normal — just a shakeout. The RS line shows you whether the stock is still outperforming the market during that pullback.
If price drops 10% but the RS line holds above its MAs, the stock is holding up better than the market. Hold.
If price drops 10% and the RS line drops below the short MA, relative strength is deteriorating. That's the first warning — sell a third.
This is why O'Neil could hold Schwab while the NASDAQ was crashing. Schwab's RS line was making new highs while the index fell. Every RS signal said hold. He held.
Case studies from the vault
Netflix (2009–2011, 750%)
O'Neil's first signal that Netflix was different: the RS line hit new all-time highs before Netflix's price hit all-time highs. RS leads price — it's the earliest tell.
Through three base sequences (cup/handle, then straight-up-from-bottom, then double bottom), the RS line stayed above both MAs. Each pullback that would have triggered a panic exit on price alone left the RS line intact. The final exit signal was two consecutive down weeks on above-average volume — the RS line confirmed distribution was happening.
Amgen (1990–2000, 50× gain)
Mid-run, Quick and Quicksand triggered during a sideways period. O'Neil sold two-thirds of his core. When the RS line recovered back above both MAs, he bought back. The position never hit Grateful Dead — he held the core through the full decade.
This is the critical point: Quick and Quicksand are partial exits, not full exits. They're early warnings that let you reduce risk while keeping a position if the stock recovers.
Schwab (1998–1999, 300%+)
The buy signal itself was an RS divergence: while NASDAQ was hitting two successive lows, Schwab was forming a higher low. "The biggest tell you'll ever see," Webster said. Schwab's RS line never triggered Quick throughout the hold. The final exit came when the RS line crossed below both MAs (Grateful Dead) after the stock had already moved 300%.
Three rules for applying this system
1. Only apply Quick/Quicksand/Grateful Dead when you are UP on the position. This is an exit system for managing gains, not a stop-loss system. Your entry stop handles downside from cost basis.
2. Apply only to a core position. O'Neil defined core as 15% of portfolio. Positions beyond 15% (up to 25%) use a different, more aggressive sell discipline — not this three-stage system.
3. If Quick or Quicksand fires but the RS line recovers, you can re-enter. The rule is: only re-enter if the RS line recovers back through both MAs. Don't re-enter on price alone.
Thai market application
The logic transfers directly to Thai stocks with weekly charts. Thai stocks can be more volatile intraday and weekly than US, so use weekly data only — not daily — to apply the MA signals on the RS line. Daily charts give too many false signals in thinner-float SET names.
Practical threshold changes: - Thai stocks can form the 8-week/21-week MA touch more often due to thinner float. The Quick signal may fire more frequently. Use it as originally intended — partial exit, not full exit. - The Grateful Dead signal (RS crosses below 21-week MA) is the reliable final exit. In Thai stocks, once RS line loses the 21-week MA, the move is usually over.
The biggest application: use RS divergence during SET corrections to build the "ready list." When SET is making lower lows, screen for Thai stocks making higher lows on the RS line. Those are the names ready to break out when the market recovers — the Schwab pattern in a Thai context.
What this means for the scanner
The current scanner uses RS≥80 (RS line slope >0, above prior-period level) as an entry gate. This system is the matching exit gate: enter on RS strength, exit when the RS line progressively loses both MAs.
Together they bracket the trade: enter when relative strength is strong, hold until relative strength deteriorates past the two-stage warning.
Source: IBD Model Book Stock Review — Mike Webster analyzing O'Neil's actual trades in Mosaic, Schwab, Amgen, Chrysler, Syntex, Netflix. Extracted 2026-06-12. See also: [RS Line Slope — The Filter That Passed](/articles/rs-line-slope-the-filter-that-passed.html) · [When to Sell a Big Winner](/articles/when-to-sell-a-big-winner.html)