TPIR doing Plinko stunt week of Apr 9th..

[MikeK]

Is all of this going to be on the exam?

And you must fail to reject the null hypothesis just to get a passing grade.

[BrandonFG]

Now I remember why I majored in journalism.

[Fedya]

I was told there would be no math.

That's a relief to Dick Clark.

[Mr. Armadillo]

Using historical averages gleamed from Scorpz's site, your typical Price is Right contestant will hit the center slot 14.2% of the time, which means there's a slightly better than even chance (roughly 55%) that the jackpot will not reset before Friday.

Though I doubt there's any analysis to back this up, even among the analysis-obsessed fans of the show, the difference between what we'd expect (22%) and the actual (14%) is probably most attributable to the players who don't drop from the center.

Also, the Plinko board has 12* rows of pegs, not nine, which I believe is the number that was previously used to construct that 22% figure.  The fact that the Plinko chip is bound by the laws of physics and not a binary left/right choice at each peg also increases the board's variance, which I would hypothesize would bring the probability of landing in each slot closer to the uniform distribution.


*technically 13, but the top-most row of pegs never comes into play

[dale_grass]

The fact that the Plinko chip is bound by the laws of physics and not a binary left/right choice at each peg also increases the board's variance, which I would hypothesize would bring the probability of landing in each slot closer to the uniform distribution.

Huh?  How does one contradict the other?  A chip hits a peg and either falls left or right.  And if you have time this week, build yourself a Galton box and check out how uniform the distribution of BBs is.

[TLEberle]

Huh?  How does one contradict the other?  A chip hits a peg and either falls left or right.  And if you have time this week, build yourself a Galton box and check out how uniform the distribution of BBs is.

Or a chip bounces twice to the left, or back up to the previous row of pegs.

/Built a bean machine and got an inverse of the expected bell curve

[dale_grass]

Or a chip bounces twice to the left, or back up to the previous row of pegs.

/Built a bean machine and got an inverse of the expected bell curve

Those cases are so few and far between they're negilible in the simplistic calculations we're dealing with.

/Built a bean machine and got an inverse of the expected bell curve

Jumping beans from south of the border tend to skew the results.  Jelly Bellies are far more trustworthy, although I am biased toward some more than others.

/See what I did there?

[clemon79]

Those cases are so few and far between they're negilible in the simplistic calculations we're dealing with.

Um, I defy you to show me an instance where a Plinko chip DID either choose an immediate left or right path from every peg it hits, all the way down the board. Having a chip roll is WAY more common than it falling in a mathematically natural way.

(That said, I suspect that the rolls even themselves out over time as well, but certainly it introduces enough of an element of randomness over the relatively short drop the chip is making as to explain why the spread might not match up completely evenly with what you would expect from Pascal's Triangle.)

[dale_grass]

Holy crap, I just watched a slew of Plinko clips on the YouTube.  It does shoot over more that one space on the majority of drops.  I sit corrected.

Now the big question: What's the correct dirstribution?

[Mr. Armadillo]

Going back to the site I sourced earlier (Link), there were 1,319 chips dropped that successfully reached the bottom of the board between seasons 29 and 39.  Of those, 187 hit $10k, 322 hit $1k, 289 hit $500, 152 hit $100, and 369 hit $0.  (22 got stuck along the way.)  If we assume the distribution is symmetrical, we get:

5.8% - 11.0% - 12.2% - 14.0% - 14.2% - 14.0% - 12.2% - 11.0% - 5.8%

Relatively flat everywhere except the far sides, which makes sense when you look at the board and realize the corners are tucked away and hard to reach.

[Dbacksfan12]

Relatively flat everywhere except the far sides, which makes sense when you look at the board and realize the corners are tucked away and hard to reach.

One reason I was a fan of the plinko setup on TnPiR '94.

[Matt Ottinger]

That said, I suspect that the rolls even themselves out over time as well

Actually, we all seem to be assuming an absolutely perfectly designed game board.  A fraction of an inch here and a fraction of an inch there, and real-world physics starts rearing its ugly head again.  Could be that the thing is tilted just enough to throw off the results in a meaningful way.  Almost certainly not deliberately, but enough to skew abstract predictions.

[clemon79]

Little something for the naysayers: I've enjoyed this all week; I don't think it's oversaturated the show, and Drew's done a fine job of pimping today's $80K drop.

Now, I'd rather they not do stunts like this every week, or even every month, but for a special thing I think it's been totally effective.

[SRIV94]

Relatively flat everywhere except the far sides, which makes sense when you look at the board and realize the corners are tucked away and hard to reach.

One reason I was a fan of the plinko setup on TnPiR '94.

Could you refresh my memory about that setup?  Had trouble finding a clip on Youtube.

[TLEberle]

Could you refresh my memory about that setup?  Had trouble finding a clip on Youtube.

2500-500-1000-zero-5000-zero-1000-500-2500. There was another that had two 5000s and the rest were zero or 500.