Facebook, Inc. (FB) missed the earnings per share (EPS) estimates when it reported results after the closing bell, on the 29th of April. The stock gapped higher on April 30 and moved above its monthly pivot at $202.60 as May began. The upside potential is at its semiannual risky level at $225.99, which would be a new record if the trial.
The social media giant has missed EPS estimates in two of the last four quarters, primarily on the warnings related to advertising revenue. The stock is not cheap, as its P/E ratio is high at 28.94 without offering a dividend, according to the trends.
The stock ended last week at $212.35, up 3.5% year to date, and in the bull market territory, 54.9% above its March 18 low of $137.10. The stock is 5.3% below its intraday high of $224.20 game on Jan. 29.
The daily chart for Facebook
Refinitiv XENITH
The daily chart of Facebook shows that the stock had been above a cross of gold since April 3, when the 50-day simple moving average rose above the 200-day simple moving average to indicate that the price increase to come. The stock has slipped to a test of its 200-day simple moving average at $161.33 June 3, 2019, as a buying opportunity. The 200-day simple moving average was tested again to $175.30 on Oct. 2 as another buying opportunity.
The stock made its intraday high of $224.20 on Jan. 29 in anticipation of a rise in the earnings report released after the close of the day. The good news was assessed to this rate, and the stock gapped lower on Jan. 30. This price gap has been filled on the force and on Feb. 20.
Facebook stock and then crashed below its 50-day simple moving average on Feb. 24 and then below the 200-day simple moving average on 4 March. This led to the March 18 low of $137.10.
The V at the bottom of the stock has resulted in a test of the quarterly risky level at $188.32, and then a test of May risky level $202.60 the 1st of May. The upside potential is now at its semi-annual risky level at $225.49.
The table of the week for Facebook
Refinitiv XENITH
The weekly chart for Facebook is positive, with above its five-week modified moving average of $189.23. The stock is also above its 200-week simple moving average, or the return to the average, to $166.25. This average provided a buying opportunity between the weeks of March 13 and April 10.
The 12 x 3 x 3 weekly slow stochastic reading is increased to 57.23 last week, an increase of 46.38 1 May. In the month of January, this reading has been above 90.00, which has put the stock in a “swell parabolic bubble” of the training, and bubbles always pop.
Trading strategy: Buy Facebook shares on the weakness of the monthly and quarterly value levels at $202.60 and $188.31, and reduce holdings on strength to semiannual risky level at $225.49.
How to use my value levels and risky levels: The closing price on Dec. 31, 2019, were inputs to my proprietary analytics. Half-yearly and annual, the levels are still on the charts. Each calculation uses the past nine closes in these time horizons.
The second quarter of 2020 the level has been established based on the 31 March, close by, and the level of the month of May was established based on the April 30 close. New weekly levels are calculated following the end of each week, and new quarterly levels occur at the end of each quarter. Half-yearly levels are updated in the middle of the year, while the levels remain in play throughout the year.
My theory is that nine years of volatility between the farm are to assume all possible bullish or bearish event for the stock are taken into account. To capture the volatility of stock prices, investors should purchase shares of the weakness of a plan, the value and reduce the holdings of the strength to a risky level. A pivot is a level value of the risk level that was violated in its time horizon. Hubs act as magnets that have a high likelihood to be re-tested prior to their time horizon expires.
How to use 12 x 3 x 3 weekly slow stochastic readings: My choice of using 12 x 3 x 3 weekly slow stochastic readings has been based on the backtesting of many methods of reading the stock prices of the momentum with the objective of finding the combination that resulted in the least number of false signals. I’ve been following the stock market crash of 1987, so I was happy with the results of more than 30 years.
The stochastic reading covers the duration of 12 weeks, high, low, and close for the stock. There is a row of calculation of the difference between the highest and the lowest compared to the farm. These levels are modified for a fast reading and slow reading, and I found that the slow playback will work best.
The stochastic reading scales between 00.00 and 100.00 with readings above 80.00 considered as overbought and readings below 20.00 regarded as oversold. A reading above 90.00 is considered to be a “swell parabolic bubble” of the training, which is usually followed by a decline of 10% to 20% over the next three to five months. A reading below 10.00 is considered to be “too cheap to ignore,” which is usually followed by a gain of 10% to 20% over the next three to five months.
Want to learn how to integrate levels of trading in your day trading strategy? Check out my new publication, the 2-Second Operator.
Source: investopedia.com