The Old New Thing

# Pivot points and financial numerology

Raymond Chen

I’m baffled by financial numerology.

For example, the pivot points of a stock are âŒ©choose your favorite fancy formulaâŒª, producing a support price below the current price and a resistance price above the current price. These pivot points are important because when the price of a stock reaches a pivot point, you can confidently predict that one of three things can happen:

1. The stock price goes up, or
2. The stock price goes down, or
3. The stock price stays about the same.

Okay, maybe this isn’t so much numerology as it is horoscopes, where you make a prediction so vague that it it matches everything.

If the market is directionless (undecided), prices may fluctuate greatly around this level until a price breakout develops.

This sentence sounds cool until you take it apart and realize that it’s saying “If prices are not consistently moving up or down, then they will stay near their original positions, until they start moving away from them.” Tautology disguised as deep insight.

I like how the article I linked at the top is titled Using Pivot Points For Predictions, but if you read the text, it says that the pivot point is an indicator, not a predictor.

Let me make up my own financial number: The birthday price. Take the birthday of the company’s CFO, add together the day month, and year, and take the last two digits of the result. Take the current stock price and replace the fractional value with the value you calculated above. For example, if the CFO’s birthday is July 4, 1776, then adding the day, month, and year produces 1787, the last two digits of which are 87. If the stock is currently trading at 18.20, then the birthday price of the stock is 18.87.

The predictive power of the birthday price is that when the stock value reaches the birthday price, one of three things will happen:

1. The stock price goes up, or
2. The stock price goes down, or
3. The stock price stays about the same.

I am a financial genius!

Popular
Oldest
• Scarlet Manuka · Edited 0

“Pivot points can be used in two ways. The first way is to determine the overall market trend. If the pivot point price is broken in an upward movement, then the market is bullish. If the price drops through the pivot point, then it’s is bearish.”

Such deep insights we are getting here. (It does say before that that the most rapid price movements are expected near the pivot point, which does make this slightly more meaningful. But not much.)

The second way is “to use pivot point price levels to enter and exit the markets”, which is what Peter Cooper was talking about.

I did find these two (consecutive!) paragraphs interesting, however:

“While at times it appears that the levels are very good at predicting price movement, there are also times when the levels appear to have no impact at all. Like any technical tool, profits won’t likely come from relying on one indicator exclusively.

“The success of a pivot point system lies squarely on the shoulders of the trader and depends on their ability to effectively use it in conjunction with other forms of technical analysis. […]”

So you’re saying the tool sometimes works and sometimes completely fails (because it’s essentially an exercise in assuming past trends will continue into the future), but it’s the trader’s fault if it desn’t work for them?

• Rez Zircon 0

Something I wrote a few years ago:

04.07.07 Stock market: glorified loansharking. You lend companies money in return for what you hope will be a usorious interest rate.

And that ‘interest rate’ is actually set by what other investors think of the stock.

Consider it an experiment in mob psychology, with your money riding on the outcome.

Thanks for making me laugh

• cheong00 0

If the first point of “Using Pivot Points For Predictions” is “Don’t use Pivot Points For Predictions” than it sounds right to me.

• Letao Wang 0

I occasionally glance through financial news, and each day, you can expect to see headlines of one of the following forms:
– Markets surge after positive earnings reports
– Markets end mixed amid positive earnings reports
– Markets plunge despite positive earnings reports
or whatever factor somebody thought had influence – jobs report, trade talks, volcanos, some world leader sneezing, etc. It must take a great deal of insight to do “analysis” of the stock market and to write these headlines.

• David Streeter 0

Many of those financial news reports are written by algorithms.

• Peter Cooper Jr. · Edited 0

My best understanding of “technical analysis” is that to the extent it “works”, it’s that if a lot of people buy expecting a price to go up then that’s what will happen (and vice-versa, if a lot of people sell because they expect the price to go down then that’s what will happen). So if you can get a large-enough group to people to agree on “rules” on when to buy or sell, their collective action can, to some degree, be self-fulfilling. (Note that I’m making no claim about whether this group of people would actually make more money trading than an average person not following their “strategy”.)

But yes, it seems to me that there’s a lot of double-speak, trying to make people feel like they’re in on some amazing “secret of the market” and that it’s a fun gambling game, when people should just invest in boring broad-market index funds and be done with it.