For 2 block card user: I did 1 000 000 deck simulation

Hey guys,
I am currently moving to 2 block card system. I think most of memory athletes know about 2 block card system.

Here is I simulated 1 million decks:

I briefly explain that.

That numbers in left side it’s images per location
inRow in order repeat that image ex: When put one image to the fist location and the second location get again one image
Alone is like put one image to fist locus, two or more image to second locus, some thing like that

2 Likes

Guys this is true? I mean you know about more than I

@Rajadodve786
@BradenExplosion
@AlexM @Chicc8 @katiek

Interesting to see it laid out with actual simulations. This is exactly what probability predicts for a 50/50 event. Each pair is essentially a coin flip: if it’s black-first you move on, if it’s red-first you stay in the same location. Every new location, there’s a 50% chance of it being a “single”. So the results line up perfectly with that.

2 Likes

Yeah you are right, It is still flip a coin.

But I tested red-first move.
Here is black-first move, but result almost the same.

Now I add how many locus needed, and count occurence but i cant print in one-line, need to short words

Looks about right to me

2 Likes

I generally use a 13-loci memory palace for cards in the memory league, and most of the time it always works out. If there are a few extras, I just imagine there are some more spaces and start placing things in imaginary locations.
If it’s a 10-minute deck event, then I use the 26 loci memory palace for each deck, since I’m always going for the worst-case scenario that technically will not appear, but if someone purposely arranged the deck in that order, knowing I use a 2-card block.

At this point, it doesn’t matter what color is coming next. I am slowly getting adapted to it. If it’s the same, put on the same locus, if it’s different, move on.

1 Like

i think you’ll probably figure this out (or already have), but as Katie said it’s just a series of coin flips BUT you can see that it doesn’t go 50-25-12.5-6.25-3.125, it falls off much faster because you’ve already seen some number of the cards and at least half of them are red (if red-first means stay at the locus in your system). so fewer red cards in the deck means it’s increasingly likely you get a black-first pair

2 Likes

First thing itʼs actual simulation. And I tested in ML cards, i just checked 5 time. counted it manually. And understand why itʼs happening. But I could not explain it. Itʼs whole math even I canʼt imagine
And I just tested 1 million and 10 million deck it’s very similar, what if 100 billion deck? we don’t know that it’s fully random.

But you are right if you get 2 card red image, thatʼs increase your getting more black-fist-pair , but one card red image it’s not effect to getting more black-fist-pair

Yeah this is pretty much exactly what I would expect as well! Very cool that you did this @onehand

1 Like