Takeaways

#
  • Hill’s momentum selection combined a market filter, an individual-stock trend filter and a volatility-adjusted ranking.
  • His five-day versus 200-day trend rule used a 3% buffer on each side to reduce repeated signals around the longer average.
  • He treated breadth as a market-regime tool and accepted whipsaws as a cost of following the model.
  • Technology weakness worried him, but he still described the September 10 decline as a correction within a bull market.

Arthur Hill still regarded the market as bullish on September 10, 2024. Technology and semiconductors were correcting, and their weakness concerned him, but his broader market model had not turned bearish.

Appearing on The Morning Show with JC Parets, Steve Strazza and Spencer Israel, Hill described the rules he used to distinguish an uncomfortable pullback from a change in trend. Those rules could produce false starts. He accepted that cost as part of following trends.

Rank momentum after checking the market

Hill’s stock-selection process began with three questions: Was the market bullish? Was the individual stock in an uptrend? And where did it rank against other momentum leaders?

He looked at roughly nine months, or 200 trading days, of price change. To keep the highest-volatility stocks from automatically dominating the rankings, he described adjusting that rate of change by a 20-day standard deviation. The adjustment was intended to make comparisons more useful across stocks with different behavior.

His rotational approach held the top ten stocks and replaced a holding when it fell out of the top twenty. He described a weekly process, with the development and testing done beforehand and orders entered on weekends.

Simple rules mattered to him. Adding complexity to make a historical test look better could leave a strategy poorly suited to the next market. He favored testing across several market cycles, generally using about twenty years of history, and sometimes starting just before the 2000 bear market to examine a difficult entry point.

A moving-average cross with a buffer

Hill used a 200-day moving average for his long-term trend filter, but he made two adjustments to reduce frequent reversals around that line.

First, he compared a five-day average with the 200-day average instead of using the daily close. Second, he required a meaningful percentage difference between them.

A five-day average moving more than 3% above the 200-day average supplied the bullish signal. Once that signal occurred, he stayed bullish until the five-day average moved more than 3% below the longer average. Simply crossing back below the 200-day line did not reverse his position.

The intervening range was a buffer against small fluctuations. Hill said he found this approach more useful for large-cap stocks and the S&P 500 than for the choppier Russell 2000. He based that distinction on his own testing.

Strong breadth was a reason to participate

Hill also described a composite breadth model using the S&P 500 and S&P 1500. Its inputs included the percentages of stocks above several moving averages, new highs and lows, and a thrust measure.

He used those readings to identify the market regime. Strong participation made stock selection more attractive to him. Weak participation called for greater caution, cash and defense.

Asked how he dealt with a bearish signal that quickly reversed, Hill said his momentum strategy moved to cash when the model turned bearish. False signals were an expense he expected to encounter. Preserving capital during periods of elevated risk was the reason to follow the model even when an exit later proved premature.

The model also needed renewed strength before turning bullish again. It was not designed to buy the exact bottom.

Technology weakness had to be weighed against other leadership

Hill placed more weight on weakening technology and semiconductors than on the strength elsewhere. Parets challenged that emphasis, pointing to financials, industrials and the equal-weight market, and recalling an earlier bull market in which semiconductors had not led.

Hill nevertheless described the current decline as a correction within a bull market. Software, fintech and cybersecurity were holding up better, while his breadth model remained bullish.

For a discretionary entry, he wanted a tradable pullback inside an established uptrend. August’s sharp decline and rapid reversal had been less appealing to him. If the market fell toward its longer-term average, he would look for stocks showing relative strength during the decline. A bearish turn in his breadth and trend filters would require a different response.

Questions and answers

#

Questions about the original 2024-09-10 episode.

How did Hill try to reduce moving-average whipsaws?

He compared a five-day average with the 200-day average and added a 3% threshold. After a bullish signal above the upper threshold, he waited for a move below the negative 3% threshold before turning bearish.

Link to this answer

How did his momentum rotation choose and replace stocks?

After applying market and stock-trend filters, Hill ranked momentum leaders. He described holding the top ten and replacing a stock when it dropped out of the top twenty, with the strategy run weekly.

Link to this answer

Why did Hill still call the market bullish?

His breadth model remained bullish, and he saw areas such as software, fintech and cybersecurity holding up. He regarded technology and semiconductor weakness as a correction at that point, while continuing to watch his trend and breadth signals.

Link to this answer

Adapted by Stock Market TV Editorial from the original video. Source date: . Article published: .