Past performance is not indicative of future results. This disclaimer is used by financial companies when they reference the performance of products such as mutual funds. But this disclaimer is a bit of an odd contradiction, because the past is all that we have to work with. If the companies that comprise the mutual fund have consistently had positive quarterly results for the last eight quarters straight, does that guarantee that they will also have a positive set of results for the next eight quarters and that their public valuation will continue to rise? Probability could be on the side of that being the case, but that might not be the whole story. And, before we get too wishful in thinking that Elastic ML's ability to forecast is our key to making a fortune in the stock market, we should be realistic about one key caveat.

The reason financial companies use this disclaimer is that there are often unknown, uncontrollable factors that emerge and can be very influential on the trajectory of something. For example, the government could change regulations or trade policies that greatly help or hinder the company's ability to operate and be profitable. There could be an internal fraudulent accounting scandal in which the executives conspired to falsify corporate performance, which becomes untenable to maintain and ultimately bankrupts the company.

These factors are deemed unknown and external because of the following reasons:

• They are outside of the control of the entity itself (as in the example of the government dictating policies independently of the company's activities)
• They are hidden from the available information about the system (an outside investor, in real time, only has access to publicly available performance reports, and not to the knowledge of the fraudulent activities that may be fabricating those performance reports)

As such, predictions are only as good as the information you have and your ability to eliminate or mitigate external unknown factors that will affect your prediction. The same is true in the world of IT data. It's not always possible to predict a trend or a failure if an unknown, external factor is at play (someone incorrectly making a configuration change, a failing piece of hardware, and so on). However, we can use probabilistic analysis to give us our best guess at the future, aside from those possible external factors. Understanding this caveat allows us to satisfy some good forecasting use cases without getting hung up on the expectation of prophecy.