In a small judge meeting last week I learned that my mentor would issue a TE Insufficient Shuffling if I shuffled the way I do at a competitive REL event he's head judging. The shuffling method in question is just using ten riffle shuffles.

I've talked to some judges about this. Some say that ten riffle shuffles sound okay. Some echoed the concerns of my mentor and told me to use multiple methods when randomizing a deck instead. One judge even gave me a very specific answer of “3/2 * log2(N)” riffle shuffles being sufficient.

This apparent lack of consensus confused me so I tried to find something close to a definite answer. Here's an article I found: Shuffling Dos & Don’ts. It is about ten years old and therefore terribly outdated but does mention “When it comes time to shuffle, use different, alternating techniques”.

Is there some consensus among judges on the specifics of Insufficient Shuffling, specifically this case of ten riffle shuffles?

This seems odd to me, I believe Frank Carston (Probably butchered the spelling of his name) put forward the idea that 8 Riffle or Mash shuffles should mathmatically randomize a deck.

Yes, Frank Karsten is known to be a magic player who knows his maths. The notion that eight is enough can be traced back to this paper.

From your answer I get that you don't think that players have to use multiple methods when randomizing a deck?

I want to point out that the infraction is for “Insufficient Shuffling” not
“Insufficient Randomization”. This is intentional. We do not expect
players to shuffle to some mathematical point of perfect randomization. We
want players to make a reasonable effort to shuffle their deck so that they
don't know the order of cards in the library. If you look at the examples
for IS, they are “No shuffle”, “One riffle”, and “No shuffle”.
The Philosophy expects players to “shuffle their deck thoroughly”. I would
find it hard to believe that a deck that was not meeting that philosophy.

Shawn

On the issue of multiple methods:
The Philosophy sections states:
“A player should shuffle his or her deck using multiple methods”
I see that as a recommendation for better shuffling, but not a hard
requirement to avoid Insufficient Shuffling.

Your first clue about the currency of that article is right near the beginning, where “pile shuffles” are listed as a valid technique. (Well, that plus the outdated “zebra” judge shirt…)

As Shawn said, we can't require players to understand nor guarantee sufficient randomization, we just demand that they make a sincere effort to shuffle sufficiently; reasonable randomization is a likely result of such efforts.

d:^D

I want to point out that the infraction is for “Insufficient Shuffling” not
“Insufficient Randomization”. This is intentional.

I thought shuffling is the method we use to achieve randomization and therefore insufficient shuffling and insufficient randomization are basically the same. But maybe this is just semantics and beside the point.

The Philosophy expects players to “shuffle their deck thoroughly”.

Well that's what I do. I really believe that my ten riffle shuffles are enough for the deck to be shuffled thoroughly and that I don't know the order or position of any of my cards when I present my deck.


So all in all you'd say that I shouldn't get IS for doing only ten riffle shuffles on a deck and present it?

The standard is Insufficient Shuffling and not Insufficient
Randomization because we don't care about the ordering of the deck to
determine whether it was shuffled sufficiently. We only look at the
methods. Human beings are notoriously bad at judging randomness, but
we can tell whether you've shuffled sufficiently. We can't give you a
definitive answer about whether 10 riffles is sufficiently shuffled or
not. There's too much variation in the way people shuffle to codify
this exactly. Clearly 1 shuffle is insufficient and 1,000,000 shuffles
is sufficient. Where we draw line remains something of a judgment
call.

The penalty for Insufficient Shuffling is intentionally a Warning
rather than a Game Loss to help judges use the penalty more easily to
inform players that their shuffling may not be good enough. It allows
us to step in with a gentle nudge and improve player behavior, rather
than being a very blunt instrument.

On Wed, Oct 26, 2016 at 8:51 AM, Laurent Eriksen
<forum-31036-00d3@apps.magicjudges.org> wrote:

Well, 1,000,000 shuffles is insufficient if, for example, the same card remains on top of the deck the entire time.

To the OP: did your mentor mention anything like this, or was it simply “you didn't use multiple methods”?

Originally posted by Laurent Eriksen:

So all in all you'd say that I shouldn't get IS for doing only ten riffle shuffles on a deck and present it?

You are correct… IF you don't consider the human factor. From a pure mathematical point of view, it is sufficient. But sadly, we are humans, who by design are not precise and tend to chose the messy way of doing things even when there are more cleaner ways to do them.

Insufficient Shuffling also takes into account
- how you did the shuffles (pretty much everyone do them differently and I've seen people doing 10 shuffles without moving the top 5 cards)
- where were you looking when you shuffled the cards (if you look at the bottom cards, no amount of shuffling will be sufficient)
- etc, etc…

This is also one of the reason why there is a requirement and not a method (do this and you are in the clear) because a method would need to be very exhaustive to be of any use and would leave a lot of room to be abused (either by the one doing the shuffling or by the opponent noticing some inconsistency with the required method)…

When you(?) came into the IRC channel, you listed other methods your mentor
suggested you mix riffling with. One of those included pile shuffling. I'll
point out that if your mentor thinks riffle-shuffling on its own is not
enough, adding a pile shuffle into the mix also isn't, as it doesn't affect
the randomness of the deck and you could still know the position of a card.

2016-10-26 12:37 GMT-04:00 Jeremie Granat <

Originally posted by Jeff Morrow:

To the OP: did your mentor mention anything like this, or was it simply “you didn't use multiple methods”?

He said it could be the case that I shuffle in such a way that the top card or the bottom card stay the same. Not intentionally (that wouldn't be IS, that's cheating) but accidentally through habit. By including multiple methods you reduce the probability of this happening because the chance is way smaller that this happens for two separate shuffling methods. Even better if you add some cuts in between which makes it impossible for the top or bottom card to stay the same through all the shuffles.

Dominik Chłobowski
When you(?) came into the IRC channel, you listed other methods your mentor
suggested you mix riffling with. One of those included pile shuffling. I'll
point out that if your mentor thinks riffle-shuffling on its own is not
enough, adding a pile shuffle into the mix also isn't, as it doesn't affect
the randomness of the deck and you could still know the position of a card.

Yes that was definitely me! But if I claimed that my mentor told me to pile count then I simply misspoke. He never said that, but there was another L1 judge present there who also thought you had to use multiple methods and included pile shuffling as one of the things I could do additionally.

If it's just about the top and bottom card then pile counting can help, depending on what order you pick the piles back up.

Riffle shuffling is sufficient as long as the top and bottom cards do indeed change (and you shuffled enough times of course). Alternate methods of shuffling do help ensure that this doesn't happen, but are not required.