TD Perfected 9 Buy (Daily)

How the TD Perfected 9 Buy (Daily) screen works

In plain words: a DeMark-style count looks for a selloff that has gone on long enough to tire out. The "setup" counts bars in a row that close lower than they did four bars earlier; when it reaches 9 without a break, sellers may be running out of steam. The "countdown" then keeps a slower tally of weak closes (close at or below the low two bars back, not necessarily in a row); a qualified 13 marks a deeper, later exhaustion point. A "perfected" 9 is a stricter 9 where the last bars also undercut the lows of bars 6 and 7. Signals averaged +0.21% one session later and +0.31% five sessions later, versus +0.26% for SPY over the same five sessions. These are timing signals, not trend calls: a 9 or 13 can fail, and the trend can keep going.

"TD Sequential" is a trademark of Tom DeMark. SetupHits is not affiliated with or endorsed by DeMark; this is our own DeMark-style Sequential count implementation, and other platforms' versions can differ.

The TD Perfected 9 Buy (Daily) screen is a DeMark-style setup: the stricter DeMark-style 9: the last two bars of the count also undercut the lows of bars 6 and 7. Daily bars. A stock qualifies on a given day when all 6 rules are true at the close (the exact thresholds are part of Pro). To keep results tradable, only S&P 500 and Nasdaq-100 members above $5 with more than $5M of average daily dollar volume are included.

Across 518 tickers from 2021-10-13 to 2026-10-14, the screen fired 5,637 times on 517 different stocks. 53.8% of signals were higher five sessions later, ranking #11 of 30 screens on SetupHits by hit rate. The average 5-day move was +0.31% (median +0.36%), ahead of SPY's +0.26% over the same days, and the equal-weight trade curve's worst drawdown was -23.6%.

Pro members get copy-paste TradingView, ThinkorSwim versions of this screen by email. In TradingView use the string in the Stock Screener filter or a Pine condition; in Finviz append the filter codes to the screener URL; in ThinkorSwim paste the thinkScript into a Stock Hacker custom study filter. Platforms compute indicators slightly differently, so expect small differences in the hit list.

Treat these numbers as context, not a forecast. The backtest uses daily bars, ignores commissions and slippage, overlaps signals on the same stock, and tests today's index members back in time, which adds survivorship bias. A mostly rising 2021–2026 market also lifted most long setups.

Full test rules: methodology. Compare every setup in our ranked list of stock screeners.