When Forecasting Goes Too Far





I, like hopefully some of you, am a huge basketball fan.  Of course the early exit from the playoffs by our beloved Chicago Bulls was demoralizing, but I still find myself glued to the television and often cheering or cursing the players on screen during the current NBA final series.  While right now the Oklahoma City Thunder is down three games to one against the Miami Heat in a best of seven series, I have been listening to a great deal of sports talk radio over the last two days and most everyone is saying the same thing…  because no team in NBA history has come back to win a championship when down three games to one in the finals, Miami is going to win.

This sounds like forecasting gone too far.

Simply because something hasn’t happened before doesn’t mean it wont eventually happen.  No one can possibly know the outcome of the future, so why should we even try to predict it?

This makes me think about how some companies, especially start-ups, may have an extremely difficult time coming up with initial sales forecasts.  Who could have predicted things like the meteoric rise in demand of products from companies like Apple and Microsoft, or the rapid decline in domestic auto sales years ago?  According to MaRS, a Canaidian consulting firm, initial forecasting of sales is paramount to a strong start in a new business venture.  http://www.marsdd.com/articles/sales-forecasting-for-start-ups/  For a company to be able to properly allocate resources to things like distribution and storage of finished inventory, accurate sales forecasting can have a major impact on a company’s bottom line.  Forecasting sales too tentatively could cause a great deal of lost potential revenue whereas too strong a sales forecast can result in major expenses related to the excessive production.  This leads to what in my opinion is an interesting debate over what type of forecasting method to use, but also how much weight should you assign to those forecasts?

As we enter the final minutes of the NBA final’s games five, I have a feeling that in this case, the forecasting consensus may be right. Congratulations Miami.

Central question:  Knowing that even the most accurate and thoughtful forecast could be completely inaccurate, how much stock should be put into its creation?  What are some instances where forecasting is crucial?  When is forecast simply unimportant?


References:  “MaRS Discovery District.” Sales Forecasting for Start-ups. N.p., n.d. Web. 21 June 2012.


6 thoughts on “When Forecasting Goes Too Far

  1. It seems to me that forecasting or looking at the trends of what has happened in the Basketball playoffs previously has showed to be effective because as we all now know MIAMI are the new champs and no matter what you OKC fans believed they definitely were unable to get back in the game. SURVEY SAYS….. MIAMI HEAT all the way. CONGRATS GUYS!!!!

  2. The NBA has plenty of statistics to back up the predictions they make. I think that when startup companies are forecasting they need to focus on external factors rather than internal ones. There are many external factors that can affect a business and most of them are not immediately obvious.

  3. Forecasting may not be 100% reliable, but it is a great way to predict what may happen in the near future. It probably is not a good idea to bet everything a company has, on a forecast, because it is true that not everything can be consistent. But in this case, most situations where a team has won already three games, has won the final. And it was not a surprise that the HEAT won due to this prediction. Forecasts can be just as accurate as the research and information a company has.

  4. I think the amount of stock that a forecast should have depends on the industry in particular, the context, and credibility of the forecaster. For instance, forecasts are always made on the rise and fall of stock prices. However, stocks operate within a notoriously volatile market and are constantly changing. As a result, even with (what appears to be) the most credible forecaster, a healthy dose of doubt should always follow stock forecasting.

    Forecasting can also be extremely crucial when predicting, say, the demand for a certain piece of inventory in the holiday season. If the business did not accurately forecast due to lack of care or simple error, then the company could have lost out on a potentially large profit. For example, my aunt owns a flower shop and every Valentine’s Day she has an insane amount of demand for red roses. She needs to use her past knowledge to make an educated guess as to how much she think she will need. If she orders too little, she will have very angry customers and lost profit. Conversely, if she orders too much she will lose money as well due to wasted excess. Forecasting is important for all businesses, but I think it is especially special for small businesses.

    I think forecasting is basically always important, especially when money is involved. However, in certain small cases forecasting might not be worth the effort. For instance, a farmer selling extra crops on the side of the road doesn’t really have use for forecasting. The farmer has either sold all his crops to the markets or been subsidized by the government and thus a few extra crops by the road really has no effect on him. It doesn’t matter if an unexpected exodus of people come to buy up all his street side products or not, because he is dealing with surplus. Anyways, those are just my 2 cents.

    P.S. I talked with a lady who worked with LA Lakers and the NBA association the other night. This may not be true, but she told me that the finals are basically rigged. From a logistics standpoint, businesses and police have to know when a team is going to win the finals (think of the rioting that occurs afterwards) and have to prepare accordingly. Its way too expensive to have all the cops out on high alert just for a potential victory. Refs rule the whole thing…

  5. In business, forecasting is very important because it is the process of predicting a future event and an underlying basis of all business decisions. Do the proper forecast is the method to help the companies development. Also, it can help the companies save a lot of money and reduce the risk of investment. From your example, most people thought HEAT would take the championship in this year. Of course, they did. However, OCK still had a chance to win the series; people could not strangle the chance of OKC to win the series. Forecast is just can be a reference; the companies should consider all the risks when they do the forecasting. In business, forecast can be wrong, so people need to realize what risks they are taking. I still believed OKC can win the championship even though most people think it was not possible.

  6. It’s interesting that in the beginning of your post, you wrote that the idea that the Heat would win because no team ever lost after being up 3 games to 1 in the NBA Finals sounded like “forecasting gone too far,” but towards the end you wrote that you basically agreed with the forecast, as you seemed to contradict yourself. I’m wondering what made you doubt it at the beginning. It’s true that simply because something has never happened doesn’t mean that it cannot happen, but isn’t a 100% track record so far a good indicator to you?

    Of course, how long of a record there is to base a forecast on is very important. I don’t know how many years the NBA has been in existence, but if, for the sake of argument, it had only been around 5 years, to say that no team had ever come back from a 3-1 deficit would not be saying very much. This is similar to your point about start-ups, which do not have any history on which a forecast can be made. That doesn’t mean it is not important, however, since as you said, too small and they miss out on revenue, and too large and they create too many expenses. As a start-up, though, I think it would always be smarter to err on the small side, since at the worst your company would be in high demand, rather than in the red.

Leave a Reply

Your email address will not be published. Required fields are marked *