A view on the picture of TV watching is a view on data, typically in a graphic form obtained using the so-called measuring panel.
The charts themselves are a mine of information, they can tell a lot to those who know how to handle this information. This article is about how to build a clear chart and how to read it.
Our new approach is that we decided to look at the picture of TV viewing through the eyes of an stock analyst.
The technical analysis of stock indicators appeared long before the television itself, and now this field is not only full of theoretical works, but also includes a bunch of practical developments. It was the stock where traders, using technical analysis, learned how to determine current tendencies, trends and predict their future behavior. This prompted us to check whether these techniques can be applied in the analysis of TV watching.
The following assumption was fundamental — the ratings of TV channels are the basis of the advertising cost for a particular channel. However, ratings are based on past broadcasts, whereas money for advertising are spent for future ones. And this is exactly the commodity futures traded on the market. That means that, most likely, the technical analysis applicable to the commodity futures markets shall also be suitable for the analysis of TV watching ratings. In addition, to be sure, we should say that our hypothesis was justified.
Theory, demand, supply and technical indicator in our model.
First, we will describe the basis that underlay our model. So to say, we will set a coordinate system in which we plan to work.
Since we planned to use the tools used by traders in various financial markets, we had to make sure that our model lay on the same theoretical basis on which the graphical analysis used in the stock market was originally built.
The basic fundamentals of the market is that everything in the market is determined by supply and demand, and the basis of technical analysis is the rule that market indicators take everything into account (this was first mentioned by Charles Dow in his theory, he was so cool that an indicator was named after him, the well-known Dow Jones index).
Thus, in our model, we must first determine what to be considered as demand and what to be considered as supply.
Offer is a number of channels that a communication service provider (a TV provider) offers in their network. Each provider is a separate trading platform with its own specific package offer, whereas the package itself is understood to be able to include only one channel (degenerating to a la cart).
Demand is determined by the subscriber base of these providers, and corresponds to how many subscribers watch a certain channel, i.e. the current demand for the channel. In our panel, this indicator is named as IPTVPORTAL INDEX.
The Iptvportal index is a technical indicator, the primary measurement result in our panel, it can be used to build charts of both shares and ratings (TVR) that are familiar to TV workers, but in further technical analysis we will employ only IPTVPORTAL INDEX. And here is why: the share does not take into account the demand level, it does not depend on how many subscribers are watching TV at a certain moment, one hundred or one million, there is no difference, thus, when applying the technical tools for market analysis to the share, we discard the key parameter of market itself, the demand, thus making the demand not suitable. As for the rating, we ignore it due to other reasons, the rating is an indicator that is calculated by approximation of the measurements, i.e. the same views, but any approximation is only approximate. Such approximation inevitably introduces an error, which naturally cannot affect the shape of the chart obtained with such an approximation, and therefore this error will be added to the results of the technical analysis performed on the chart. Thus, when working with the primary data, we do not introduce additional error into the forecasts as a result of approximation.
By introducing our technical indicator, we offer analytics that, in addition to Share and TVR, includes another projection that reflects the real picture of TV watching and is intended for further processing using graphical technical analysis tools.
IPTVPORTAL Research Panel
The quality of TV watching research is directly determined by the quality of research panel used for the measurements. Similarly to a video camera, the higher the camera matrix resolution, the better the result of the produced image.
The telemetry research panel also has its own resolution. It is a number determined by the number of controlled subscriber screens.
In addition to the number of controlled subscriber screens, it is very important how these screens are distributed and their type (smart TVs, set-top televisions, smartphones, tablets).
The key distribution parameter is the coverage area. One city, two, hundred cities. Obviously, if the panel measures only one city, it is safe to say that the results of its measurements can be representative only for a given city, and even so may be not necessarily right. It is important that the panel distribution in a given city reflect the TV watching of different groups of people, in particular how they consume in provider networks with different TV package offers.
Below are two exemplary figures showing TV watching in networks that offer different packages, one can clearly see the viewers’ interest in the Mult Channel.
Fig. 1: Share, channels, IPTVPORTAL measuring panel, platform No. A
Fig. 2: Share, channels, IPTVPORTAL measuring panel, platform No. B
After analyzing the picture of TV watching at these two platforms, one can clearly see the level of subscribers’ real interest in the Mult Channel, which is not that clear in the generalized picture of TV watching produced by a panel that provides only average data for the two projects. This is due to the fact that a simple territorial distribution of measuring control points can only show average consumption for a particular channel, i.e. average across the board, but does not show the potential demand for a particular channel. This can also be explained as follows: imagine that Channel One became inaccessible to half of the communication service providers, for whatever reasons. Eventually, the overall nationwide rating of Channel One will fall by half or more, or even getting lower as compared to other channels, but in those networks where Channel One broadcasting is not interrupted, the rating will remain the same. For advertisers, the value of Channel One is based on its overall rating (after all, this is an indicator that shows how many subscribers will see a commercial), the potential value of Channel One is however shown by its specific rating/share in a network in which it is broadcast. The same goes for the Mult Channel. This should be taken into account by the carriers, TV providers who draft their package offer; in addition to a channel’s overall rating in the country, they should also take into account the demand indicators for a particular channel in a network where it is broadcast. It is advisable to analyze the level of demand for the channel in different networks, the more the better.
To see the potential value of a television channel, content is a part of an analyst job. After all, it is the value that corresponds to the popular trends in TV watching. It is the high potential value of a channel that indicates its compliance with the TV watching trend.
Thus, it is preferred that the measuring panel can monitor one hundred providers in one city, rather than one provider with one package offer in one hundred cities. The last example is typical for an OTT provider, in spite of its large territorial coverage, this is just one and the same model of TV watching viewing behavior because the package offer is the same. Whereas a high-quality measuring panel should have a good distribution in terms of both territory and providers.
Our research panel encompasses more than 300 providers providing TV services using IPTV, including OTT providers. Each provider is a separate measuring subpanel, which we call a trading platform or simply a platform in our model. There are operators/platforms with both unique package offers, and same package offers.
Japanese candlesticks charting
To demonstrate how technical analysis for TV watching data is performed, let us consider an example with Japanese candlesticks as the most vivid method of graphic analysis. Only the terms such as «hanging man pattern», «bullish hammer», «evening and morning stars» deserve a special notice.
The method has been developed for centuries in the East. For the record, the first futures market was founded in the Japan. At first, the merchants in the Japanese stock exchange traded rice, but it was hard to transport, so they invented «rice coupons» and traded them, and then they realized that it was possible to write receipts for further yields, which were called «empty coupons», thus organizing the first futures exchange. By the middle of the 18th century, there was already a brisk trade in «empty coupons». At the same time, there appeared a need to predict the future price of rice. Prediction of future prices was based on past prices, thus forming the method of graphical analysis, which is now called Japanese candlesticks.
The figure below shows the candlestick chart for Channel One during New Year’s holidays.
A candlestick consists of a body, the thick part, and of shadows just above and below the body. The candlestick body represents the price range between the open and close of trading; in our case, the technical indicator value range between the beginning and end of a measurement interval. If the technical indicator value has grown during the measurement interval, then the candlestick body turns red, if the indicator value has fallen, then the candlestick body turns green. The upper shadow (a blue line above the candlestick body, not always shown) shows the maximum value reached by the indicator during the interval. Accordingly, the lower shadow (a blue line below the candlestick body, not always shown) shows the minimum value reached by the indicator.
For example, the largest rising candlestick of the chart shown in Fig. 1 (Putin’s candlestick) shows how the Channel One audience increased dramatically during the New Year speech of the President, and immediately reversed thereafter (a large falling candlestick).
Candlestick analysis can be applied for any intervals. In this article, however, we will only demonstrate charts with hourly intervals.
reversal signals, «hanging man candle»
The graph in Fig. 1 clearly exhibits the reversal points; no technical analysis skills are required to predict the reversal at the beginning of the day when everyone wakes up and starts to consume content and, accordingly, at the end of the day when everyone, on the contrary, gets ready for bed. However, to predict the reversal in the daily interval based only on well-known fundamental data, which may include those from program timetable, is quite difficult. However, with the technical analysis skills, in particular knowing the reversal signals in the candlestick analysis (appearance of candlesticks of a special shape and in a corresponding sequence), one can significantly improve the quality of prediction of market behavior.
Look at the candlestick chart iptvportal index for Channel One on January 2 Fig. 2
In the middle of the day, after 3 pm, a trend change can be observed: the rise is replaced by a sharp drop, the bull trend is replaced by a bear trend. The question is, could this be predicted in advance?
A technical analyst would say yes. The drop was preceded by two reversal signals. These are candlesticks 1 and 2 with a short body and a long lower shadow, they are called «hanged man candlestick» and most likely predict a reversal from a bullish to a bearish trend.
This is just one example; it is not possible to show all the signals in this article given the large amount of information and the short format of the article. The aim here is only to outline the approach itself.
Rule of multiple techniques
Financial analysis includes a bunch of techniques, Dow theory, trend lines, moving averages, oscillators, Elliot wave theory etc. It is recommended that you use them all together. The technique of confirmation of one indicator by another is constantly used by graphic analysts. It is called the Rule of multiple techniques.
Let us go back to our chart for Channel One. Fig. 2. Some kind of drop can be observed in the middle of the day (which we predicted in the previous section). A hasty conclusion: program directors made a mistake somewhere and what caused the drop should be analyzed. However, by using a technical analysis tool called the trend line (Fig. 3), in particular by drawing a downward resistance line reflecting the “bearish” trend (not surprisingly, after the New Year programs the demand is gradually lowers), we can see that this is not a pit, but two mountains. That is, Channel One outwitted the market twice, so it is worth analyzing how the two successes occurred rather than how the drop occurred.
Those who want to try analyzing real TV watching charts, please go to our page.
P.S. Unexpected discovery
Those who are familiar with the theory of technical analysis are, of course, also familiar with criticism concerning this theory, in particular to the well-known «self-fulfilling prophecy». The essence of this prophecy is that the technical analysis itself does not work, and the high number of come-true forecasts is due to the fact that traders themselves, following their own rules, adjust the market in the right direction. However, in our case, the buyers of the goods (TV channels) are not traders who know the rules of technical analysis, but ordinary people, therefore, we cannot talk about any «self-fulfilling prophecy», that is, the methods of technical analysis work, work on their own. The debate concerning «self-fulfilling prophecy» can be put to rest.