OHLC quantitative relationships, intra-day

         Use the OHLC quantitative relationships question to examine the magnitude of moves based on relationships between pairs of these four key price levels. You can also use this question to screen trades for minimum profitability and/or risk.

Example:

         In an uptrending market, a long position taken on the opening makes at least an eighth of a point profit by the close of the next day.

Questions from Traders:

         How often can I take out an eighth of a point by holding a position taken on the opening in an up market? What is the pain threshold on this type of trade? How often will I lose money?

Setting Up the Search:

Click on the parts of the box below where appears for explanations of features.

Answers:

         Previously (in the OHLC relationships question example), we tested the period from 11/01/92-03/01/93 to see how many times the bond market closed higher than or equal to the opening of the same day. We now refine the search to find out what percentage of these higher closes are more than 4 ticks higher than the opening, as shown above. On 76.6% of the days, the market closed more than an eighth of a point higher, open to close. We now have some idea of the profitability which can be expected from this type of trade. To find the risk associated with the trade we do the opposite search process. We first search the 81 day time period for the percentage of days that the market closed down on the day: 34.6%. Refining the search, we find that 53.6% of the time the market closed down more than 4 ticks.

         We now have some powerful information about the risk/reward characteristics in an uptrending market. In our example, 58% of the time the market will close higher on the day and almost 35% of the time it will close lower. But, over 76% of the time the higher closes will be more than 4/32nds, whereas only a bit more than half of the down closes will exceed this figure.

         Note: Our traders questioned the results of our initial search which showed that of the 81 days searched, 47 days closed at or above the opening and 28 of the 81 days closed below the opening. Why is the total 75 rather than 81 days? The answer lies in the manner in which we asked the question. We looked for the closing price in period <L>. On six of the days in the time frame that we examined, the market closed early and, therefore, those days do not show up in the search as we described it, i.e. there was no period <L>!