Thursday, September 12, 2013

Better is Better than Best


            A long time ago, I remember reading an article about product marketing in which it stated that you'll never (rarely?) see a company say that there product is better than a similar product.  Why?  Because if you say something is better you have to prove that it is better.  If you, on the other hand, say your product is the 'best' product then it is possible for the other product to be 'best' too.  Best, in marketing parlance, simply means as good as, whereas, better means superior.  So much for those Run, Spot, Run books which talk about good, better and best.  In marketing it is more like good, best and better.

            As bad as it is in the English language, I think it gets even worse when it comes to math.  You can make numbers say just about anything you want them to.  In the past couple of years, much has been made about the 1% (or is it 2%) of the country that is the wealthiest members of our society.   Then there are those that focus on the remaining 99% (or is it 98%).  I don't get political in this column, so please understand I am making no political commentaries here - only mathematical ones.  A friend of mine posted up on Facebook the other day some statistics about what happens if you look at the top 1% of the world instead of just the U.S.A.  All of a sudden a very significant portion of the country is in the top 1%.   Which statistic is more relevant?  I know people who are millionaires who think they are doing 'ok' and I know people who make very modest salaries who are as happy as can be.   The relevance is more likely in the message that someone is trying to send rather than anything absolute. 

            In the case I just discussed, I'll assume that all data presented was reasonably accurate.  A such, no lies were told.  No misinformation was disseminated.   Data was simply presented in a way to try and get some particular message across.  A few months ago, I wrote a column in which I asked if the U.S. has had a Democrat or Republican President more.  In the past 5 years, it has been a Democrat.  But in the last 8 out of 13 years, we had a Republican.  But, in 13 of past 21 years, a Democrat.  In 20 of Past 33 years, a Republican.  I can keep going backwards through the 20th century.  Someone attempting to make a political point is likely to use whichever statistic backs his point the best, even though it may have only minor relevance to the point.

            When it comes to gambling, the numbers can be manipulated just as much, if not more so.  If I had to take a guess, I'd say that the average video poker machine in Las Vegas probably has a payback of 97%.   Now, if I'm a casino whose average video poker payback is only 96%, I might put together some advertising that simply says 'come to Vegas where the video pokers pay an average of 97%!' .  The implication is that the casino pays this as well, but that's not what they said.  On the other hand, a locals casino that likely has a significantly higher average payback is much more likely to say 'come to the XXX casino, where OUR video poker machines payback 98.5%'.  Of course, we don't really know how they calculate these averages.   With slot machines they simply can present how they paid in the prior month because there is no human error involved.  With video poker machines do they simply take a straight average of all their video poker paytables?  Do they weight $1 machines more than nickel machines?  Maybe there is a uniform method for doing this or maybe they have enough wiggle room to give you whatever number they feel will send the right message. 

            Of course the 'average' payback has very limited value to an Expert Player.  In the simplest example, let's assume we have 2 casinos with 2 video poker machines each.  One casino boasts a 98% average and the other a 97% average.  So, should you head over to the one with the higher average?  What if their 2 machines each have a 98% payback.  But the other casino has one paying 94.5% and the other paying 99.5%.   The only machine out of the bunch worth playing is this last one, even though the casino's average is lower than the others.

            One of the local casinos here boasts of having the most machines paying over 100%.  I'll assume that they are telling the truth with this, but that doesn't mean that their machines don't pay 100.01% and that all of the machines below 100% pay 96%.  Now, to the experienced Player you will still happily seek out the 100.01% machine, but this doesn't mean that you can just show up and sit down at any machine and know that you are playing 100.01%.  At the same time, you would be better off finding a 101% machine at another casino that may have only 3 machines of this type and whose average machine pays 96%!


            So, what's my point?  Don't be sucked in by numbers that can be made to say anything they want.  The payback of a machine is an absolute number, calculated with mathematical certainty.  It doesn't mean that every time you play (even if it is for a few hours), that you will experience this payback.  It does mean that over time, your experience should begin to approximate this theoretical payback, assuming you are using the right strategy.  When you look at the paytable on a machine, that tells you the payback of the machine.  This is all that matters.  Marketing numbers that the casinos produce are just that - numbers they produce.  After all, why trust the group that has actually made better, better than best?

No comments:

Post a Comment