Thursday, April 16, 2009

Correlation does not imply Causation

In a recent discussion, I was told that to establish the cause of a trend, market research has to be performed. The writer indicated that correlation does not imply causation. You need to do market research to find out what the real cause of the trend is.

Correlation is the measurement of the relationship between two variables. When A changes, so does B. But that does not mean that A causes B. It’s a common mistake made by many. But the fact remains that it doesn’t exclude the chance that A affected B.

The discussion touched on the ROF (rate of fire) changes that have been implemented in various venues within the paintball industry. The PSP drop in ROF in the past two years is the most talked about and widely known. The Higher Ups in the paintball industry came to the conclusion that the increase in ROF in the past years has been a major contributor for the decrease in the paintball growth rate to the point where it has become a negative growth rate or a decline rate.

But were they right? Is the fact that markers became available and were allowed to be used with higher ROF’s the reason we have seen a decline in paintball? Are the high ROF’s the cause, keeping new players from coming back a second time? Have a high number of first time players actually said, “I would have enjoyed this more and would probably come back again if only everyone were shooting at a slower rate.”?

As a field owner, I ask first time players every day if they are having fun. Most answer with a resounding “Yes!” But every once in a while, I’ll get a player that will just say, “It was OK.” I can tell by the tone in their voice, that it wasn’t their thing and I’m probably not going to see them again after they leave. Very rarely do any of those customers actually give me a reason why they probably won’t be back again, with the single exception being, “It hurts too much.” Fair enough. Some people can’t, or don’t want to, deal with the discomfort of getting shot with a paintball.

Paintball isn’t for everyone. Some know, or at least believe, they are not going to like it before they even try it, so they don’t. Others are skeptical, but are persuaded by their friends to give it a try. They’ll try it, but find out it’s not the kind of activity they like to take part in. I can certainly live with that. All of us have refused to try or maybe have tried something once and found out it wasn’t of interest to us, even though it interests others we know.

But I’m getting off topic. Back to the discussion of whether or not the higher ROF’s are the cause of new players not returning. What about all the other potential causes that have been thrown around in the past? What about things like regular players taking advantage of fresh meat and overshooting them, or regulars making fun of new players, or the cost to play paintball being too high? I’m sure those reasons and others all have some merit.

So what is our market research going to reveal to us? If given a blank page, most of those customers that decide not to come back a second time are probably not going to be able to explain exactly what it is about their experience they didn’t like. They just know they spent money to have a good time, and in the end, the experience they had wasn’t worth the money they spent and they aren’t going to do it again.

We could get more specific and ask them direct questions. We could ask things like, “Do you think if everyone was shooting slower you would have had a better time?” “Do you think if the regular players that were there would have been nicer to you, you would have had a better time?” Do you think if the prices were lower, it would entice you to come back again?”

Do we really need to ask these questions though? Would a customer even know if they would enjoy the experience more if all the players were shooting slower without experiencing it? Would most of the answers we would receive not be predictable? Would we not be putting thoughts in people’s heads just by asking the questions? For instance, I am sure that if I asked my customers if they would come back more often if the prices were lower, I would get an almost 100% “yes” response. So if I listened to my customers (a good businessman should always listen to his customers, right?), if I lowered my prices, I would get more customers (or at least have the same customers come back more often). Whether or not I would make more money is a different matter, but at least it would help solve the player decline rate and I might make up for it in higher attendance volume. Or would it?

Market research (asking our customers) would tell us that a field would get more business (either more customers or the current customers would come more often) if the lowered their prices (I’m making an assumption here that that is indeed what the market research would show us – I haven’t actually asked the question. I make this assumption because thousands of posters on paintball forums are constantly telling us that they would play more often if prices were lower). But we already knew this, didn’t we? Isn’t this one of the basic laws of Supply and Demand? Lower prices and more people will buy and/or individuals will buy more of it. If it’s a law (law of economics, not State, Province or Federal), does that mean in this case correlation does imply causation? One variable affects another, right? Drop prices, sell more, raise prices, sell less.

So tell me then, why it is that a few years ago, during a price war one of our competitors started, when we dropped prices considerably of the main component (paintballs) to play the game, after a few months we started to get a decline in attendance over the same period in previous years? And why, when we raised the prices back to what they were before, that trend reversed and we started to see growth in the attendance rate again after a few months? I say a “a few months” because initially, exactly what we expected to happen, did happen. When we first lowered our prices, attendance did increase for a little while, then leveled off and after a few months started to drop. When we returned our prices back to the original “high” prices, initially attendance dropped (only by “regular” players though), and after a few months, attendance started to increase again.

So in the short term, exactly what the market research predicts did indeed happen. But in the long term, exactly the opposite happened.

I’ll leave you with that conundrum for now and let you ponder why this might be. I will just add that when we dropped the price of our paintballs (by close to 40% by the way, not an insignificant amount), our individual customer spending did not decrease significantly at all. And no, prices other than the price of paintballs were not changed during this time.

1 comment:

  1. attendance droped because newer players were being shot to shit cause there was no worrie for paintball preservation! Yes? No?