I hate to break it to you, but the 4th digit in my analysis of TPIR car prices is even less exciting than the 3rd digit. As you can see, it is fairly evenly distributed, so I’ll keep this post short and sweet. Like the 3rd digit, the 4th digit rarely features the number 0 (3% of the time, which translated to 6 out of 200 cars). 1, on the other hand, is the only number that I’m guessing is significantly over-represented in the 4th spot. So I guess that’s good to know, especially since 1 is often an unpopular guess, for any digit. Everything else looks, to me, randomly distributed, and the minor fluctuations are likely not statistically significant. For analysis on how the third digit might predict the fourth, or how the fourth might predict the fifth, I can only say the same thing I said yesterday, which is that my sample set is not large enough to make conclusive predictions of that kind, and the repetition of certain prices for the same car over and over cancels out any effect we might see.
I suppose that, with this database, you could start to memorize the various prices of every car, and for a short time, you’d do very well if you were on the show and played a game for a car. I will try to make this dataset available to you all soon (comment below if you have a suggestion on how to do this).
In the meantime, don’t forget to check out my analyses on the 1st digit, 2nd digit, and 5th digit, all of which are much more interesting and useful than the 3rd and 4th.
