Well, that was interesting.

For weeks, it seemed that the market was looking for a reason to take an upward swing.

It finally happened, including the famed follow-through day, which we covered last week. Now some analysts are wondering how high the market can go. The same indicators that were only recently describing a broken market are now tilting bullish.

Others believe the bounce back may have already fizzled out. They want to see more stocks benefitting from the current rally. The intermediate trend is still pointing down for many stocks and indexes.

FWIW, the Swingex Index has been negative for the last two days. The last previous negative reading was a -1 on February 26th.

The signals are mixed. So we decided to use this week’s newsletter to discuss a quirk in a common market statistic that might soon have some relevance.

New Highs and New Lows

Even if you have never cared about “new highs” and “new lows”, you surely have heard others talking about it every now and then. We think it might matter in the coming weeks, so this is a good time to look into what people actually mean when they use those terms.

Most of the time, when someone mentions a stock making a new low or the number of stocks making new highs, they implicitly mean a rolling 52-week time period. Call it a year, if you want to. Other time periods are legitimate to use and they are - or should be - explicitly stated: an all-time high, a 1-month low, etc.

It wasn’t always this way. Before 1980, when technology was less advanced, when data publishers noted annual or yearly highs and lows they actually were referring to the calendar year. Sort of. You probably are already thinking that the data would be a useless mess in January.

The way publications partly solved this problem was, in the first few months of a year, to use data starting from the beginning of the prior year. By March, new high/low statistics were based on 14+ months of data. Then, in April, they would switch to using only the current year’s data. Not exactly a robust way of doing things, is it?

Now a rolling 52-week number, updated daily, is standard. However, there are still some devils in the details. Some data providers only evaluate the end-of-day closing price to determine if something has made a new high or low. Others look at the entire range of prices throughout the day.

On top of that, there are “corporate actions”. This includes stock splits and declared dividends. Most data sources, especially charting services, will adjust their data for stock splits. They may or may not adjust for dividends.

And then there is the issue of which listings to count if you want to add up the new highs and lows for a particular stock exchange or market. Dow Jones, publisher of The Wall Street Journal, excludes ETFs from their statistics. Other sources include them.

Why does all of this matter now?

Remember that new high/low statistics are understood to be based on a 52-week lookback period? And remember what was happening 52 weeks ago? Yes, the tariff meltdown. The market hit bottom on April 8th last year.

Few stocks have been making new lows because last April’s debacle set a pretty low floor for the annual data. Now that the anniversary is behind us, it will become “easier” for stocks to set a 52-week low.

If the market softens a bit from here, we could see an expanding list of stocks officially reaching new lows. And people will freak out about it. Needlessly, we think. It’s all just a matter of the calendar and how “new lows” are defined.