7 ms·

The article doesn't provide some key details regarding the one-day returns, particularly the precise timeframe used. If the price of a stock rises sharply the i

The article doesn't provide some key details regarding the one-day returns, particularly the precise timeframe used. If the price of a stock rises sharply the instant it's mentioned on the show then decays gradually, and the basis price being compared is the pre-mention price, then the returns would largely be illusory---you'd have to beat the near-instantaneous initial rise to make any money. Otherwise, you'd be likely to lose money by pursuing this strategy because your buy would be executed at or near the local peak. On the other hand, if the basis for the buy is the price say, an hour after the show, and the sell is priced 24 hours later, then it would be a viable trading strategy for an individual investor.

PeterCorless 5y ago
He noted that the show is only aired after the close of the market (6 - 7 pm Eastern, to be specific). So the stock could not rise the moment it was mentioned except in after-hours market trading.
- vannevar 5y ago
  That still presents the same problem, just deferring it to the start of the next trading day. It's not really a viable strategy for the ordinary day trader if almost all of the gains come in the first milliseconds of trading. As an aside, it also looks like there's a decimal error in the 1-day SPY Buy row, it's unlikely the SPY Buy return would be 10x the SPY Positive Mention return---the difference between them should be small, since for SPY it's largely a random 1-day sampling over the 5-year timeframe.
  - toast0 5y ago
    > That still presents the same problem, just deferring it to the start of the next trading day. It's not really a viable strategy for the ordinary day trader if almost all of the gains come in the first milliseconds of trading. If these are stocks traded on major exchanges, you can place a market buy at open order, and get the opening price. You can then place a market sell at close order and get the closing price. Not that either of those are a particularly good idea, but they are available if that's what you want to do.
    - vannevar 5y ago
      Right, but the opening price will be gapped up from the previous day's closing price, which was before the show. It actually makes the situation worse, because the rise is truly instantaneous, without even milliseconds to get an order in before it goes up.
      - toast0 5y ago
        Sure, the opening price will be different than the previous day's closing price, but if Cramer comes on after the market closes and tells you to buy, and you're sensible enough to ignore the after/before market, then the earliest price you can get is the opening price (you could do an analysis with early/late trading, but I don't think the data is available, and only crazy people trade outside of market hours). And that's the price that was used for the analysis. So fair's fair. Would Cramer's advice be better if you could retroactively buy or sell before his show? Maybe, but that's not a realistic question.
        vannevar 5y ago
        Ah, got it---I completely missed that they used open-close for the day after.
jacobr1 5y ago
The article presumes that show airs after market close. So they model they one day timeframe as the market open-close the next day. This doesn't work with international stocks, but I suspect they tossed those out with the "we didn't find the data" picks.

PeterCorless 5y ago

He noted that the show is only aired after the close of the market (6 - 7 pm Eastern, to be specific). So the stock could not rise the moment it was mentioned except in after-hours market trading.

vannevar 5y ago
That still presents the same problem, just deferring it to the start of the next trading day. It's not really a viable strategy for the ordinary day trader if almost all of the gains come in the first milliseconds of trading. As an aside, it also looks like there's a decimal error in the 1-day SPY Buy row, it's unlikely the SPY Buy return would be 10x the SPY Positive Mention return---the difference between them should be small, since for SPY it's largely a random 1-day sampling over the 5-year timeframe.
- toast0 5y ago
  > That still presents the same problem, just deferring it to the start of the next trading day. It's not really a viable strategy for the ordinary day trader if almost all of the gains come in the first milliseconds of trading. If these are stocks traded on major exchanges, you can place a market buy at open order, and get the opening price. You can then place a market sell at close order and get the closing price. Not that either of those are a particularly good idea, but they are available if that's what you want to do.
  - vannevar 5y ago
    Right, but the opening price will be gapped up from the previous day's closing price, which was before the show. It actually makes the situation worse, because the rise is truly instantaneous, without even milliseconds to get an order in before it goes up.
    - toast0 5y ago
      Sure, the opening price will be different than the previous day's closing price, but if Cramer comes on after the market closes and tells you to buy, and you're sensible enough to ignore the after/before market, then the earliest price you can get is the opening price (you could do an analysis with early/late trading, but I don't think the data is available, and only crazy people trade outside of market hours). And that's the price that was used for the analysis. So fair's fair. Would Cramer's advice be better if you could retroactively buy or sell before his show? Maybe, but that's not a realistic question.
      - vannevar 5y ago
        Ah, got it---I completely missed that they used open-close for the day after.

jacobr1 5y ago

The article presumes that show airs after market close. So they model they one day timeframe as the market open-close the next day. This doesn't work with international stocks, but I suspect they tossed those out with the "we didn't find the data" picks.