Short cryptographic ciphers like Zodiac's Z13 (13 characters) can produce thousands of valid name solutions when using transposition and substitution techniques, making it difficult to determine which solution is correct without additional contextual evidence; the apparent uniqueness of a solution often depends on the analyst's prior assumptions and biases rather than cryptographic certainty.
Install our extension to search inside any video instantly.
Zodiac and Black Dahlia SOLVED! By Sherlock or Charlatan? (Part One)Added:
Alex Baber has just solved two of the world's most famous cold cases, the Black Dalia killing and the Zodiac murders, and he's got the codereing and cryptographers to prove it. At least according to recent news reports. If true, this is huge news. But didn't the media outlets say the Zodiac case was solved before? FAL Zerawi cracked Zodiac ciphers and found Lawrence Kane's name.
But wait, the case breakers already cracked it. Zodiac was Gary Francis Post. Or wasn't Zodiac the biological father of Gary Stewart? But the news reports said Dennis Kaufman's stepfather was Zodiac. Does that mean Zodiac wasn't Deborah Perez's father? It seems like everyone was Zodiac. Well, let's give this new story the benefit of the doubt.
Alex Baber is an amateur sleuth operating behind an entity called Cold Case Consultants of America.
He describes himself as an autodidact polymath, variously diagnosed as an empath with perceptionism, his term for neurodeiversity, and a high IQ.
>> They put me in this special leadership class uh tested my IQ. I tested 161 at the age of 14, 165 at the age of 18, 168 in my 20s. In short, he regards himself as a self-taught genius with mental superpowers and deep knowledge on a wide variety of subjects. And for some reason, he reads backwards.
>> I I read from right to left. And most of the time, I read my books from the end to the beginning. Babber believes he's identified a man who was responsible for both the murder of Elizabeth Short, also known as the Black Dalia, and the crimes of the Zodiac Killer. The man is Marvin Margolus, who was once Elizabeth Short's roommate. Cleared by police in the initial investigation, he still turns up as a suspect in the Black Dalia sleuth community. Apparently, Baber's discovery was compelling enough to convince two former LAPD homicide detectives that he cracked both infamous cases. Former LAPD homicide detective Rick Jackson says, "In my opinion, these are solved cases.
There are too many links with both.
There's overwhelming circumstantial evidence.
>> In terms of your level of confidence, one to 100. Where are you?
>> I'm at 100%.
>> Jackson previously worked with Mity Roberts, an LAPD detective who was very familiar with the Black Dalia case.
>> Former homicide detective Mitsy Roberts was in charge of all things Dalia at the LAPD for 17 years.
>> She also believes Alex Bab has cracked both cases. I feel like when you get a coincidence after a coincidence after a coincidence that leads to the same conclusion that it's no longer coincidence, it's a fact. That seems what's happening in this case for sure.
>> Author Michael Connelly also got involved. He's a best-selling crime novelist whose books have been adapted to movies and TV shows.
Connelly packaged up Alex Bab's story as a podcast called Killer in the Code.
Alex Baber says he identified Marvin Margolus by working on Zodiac's 13 character cipher, one of the remaining unsolved cryptograms from the killer. He claims he used AI to generate millions of names and narrow them down to one.
>> You looked at 71 million names.
>> What I did was I was able to eliminate 94% of 71 million right off the top just because they weren't real people. And so then you were down to >> and then it went from there to a little over 7 uh,000 and then from there it went down to 184 and then to 35 then to 14 and then to one.
>> He claims the final name from his process was Marvin Merrill which turned out to be an alias for Marvin Margolus.
His investigation produced more of what he believes are links between Marvin Margolus and both the Black Dalia and Zodiac Killer cases. I'm going to set those aside for later in the video for now because according to Michael Conny's podcast, >> I happen to know very little about ciphers and cryptography. But breaking the Z13 code is the foundation on which this entire investigation stands.
>> Okay, it all hangs on the Z13 code. But wait, Bab doesn't even call it evidence.
>> The Z13 solution was never viewed as a piece of evidence. It was always viewed as a investigative tool.
>> Huh. So the result is good enough for an investigation but not for evidence. Then what value does it have exactly? It must have a lot of value because Baber and Connelly got at least three people with cryptography backgrounds to endorse his codereing work.
>> I had a brother who worked for the NSA and I knew he could get me somebody really good to either tear this apart or confirm it. It was like three guys with about a 100red years of experience at NSA came back and said this is legit. He did it. They called it monumental what he did.
>> Former NSA employees Ed Giorgio, Patrick Henry, and Rich Wiznookski reviewed Bab's solution, found that it withstands scrutiny, and developed details to support it. Does Alex Baber's Z13 solution really withstand scrutiny? In the next part of this video, let's scrutinize his code breaking together.
Then, we'll examine some of his evidence that isn't related to cryptography. And finally, we'll get to what might be the most peculiar part of this story. Alex Baber himself.
The Daily Mail posted a video of Bab demonstrating his Z13 decipherment procedure. Here's how it works. First, let's point out which cipher symbols are regular letters of the alphabet. Then the rest which are various symbols, the zodiac crosshairs symbol, circled eight symbols, and this one that looks like an upside down T. We're going to take the original cipher and turn it into two rows in a grid. The first six characters go into the first row. Then move symbols from the end to make the second row like this.
But now we're missing a symbol in the first row. Baber fixes this by putting a null or blank there. It's just a placeholder. It doesn't end up standing for anything. His grid has an appealing symmetry here. Some columns only have normal letters and others only have special symbols. Now we've got a nice 2x7 grid. Let's number the columns from 1 to 7. Next, Bab tried different ways to rearrange the columns. He's only moving the columns around because it preserves that nice symmetry where the letters are together and symbols are together. These different rearrangements are called permutations. There are 5,040 different permutations of these columns.
Ber says AI told him that of those arrangements or permutations there was only one key that came back with a probable name. His Z13 key has two parts. First, we need the permutation of columns. This is basically the sequence of column numbers. Then we need the substitution table which tells us what letters to replace the symbols with. The column permutation Babber found was 3 4 5 7 2 1 6. So let's start with the original grid where the column permutation is 1 2 3 4 5 6 7. Now we move the columns into their new places which gives us 3 4 5 7 2 1 and six. To finish it off, Baber uses this substitution table. Now we can replace the symbols with plain text letters. The N's become M's. Zodiac crosshairs symbol becomes A. Circled eights become Rs. E is V. A's are eyes. K is N. upside down T is E and M's are L's. Ignoring the placeholder, now we can read the name Marvin Merrill.
Marvin Merrill is an alias that was used by Marvin Margolus later in his life.
There's evidence of this on ancestry.com. His social security record shows both names and the California death index shows Marvin Merrill. Now we ask the big question. Is this the solution to Z13? Did Zodiac, who also happened to kill Elizabeth Short, give up his actual alias in the cipher this way?
>> You're that sure?
>> Without sounding cocky or arrogant, I have to say yes. I am undoubtedly sure that this is the right guy.
>> But you first have to accept all of his assumptions. Let's rewind a little bit and look at the original Z13 cipher.
Instead of adding more steps, let's treat it as a simple substitution. So, all we're going to do is replace symbols with their corresponding letters. It's a super simple premise. Even under this super simple premise, we can find a lot of names that fit.
We can even find a lot of phrases and nicknames that fit.
Even though Z13 already has many solutions, Bab's column permutations multiply them by about 5,000.
This happens because we aren't just assigning letters to eight different symbols anymore. We're also scrambling the columns in over 5,000 different ways. Ver also made a huge assumption that Zodiac put his real name in the Z13 cipher. It seems reasonable to assume this, right? After all, he did say, "My name is" right before the cipher. Zodiac wouldn't lie to us, right? Well, except when he bragged about his diabolical death machine, which nobody ever found.
Or when he was going to shoot school kids in buses, but never did. Or when he wrote that he had 37 victims, but offered no proof to back him up, unlike his earlier crimes. Wait, didn't he offer his identity before? In one of the Z408's letters, he wrote, "In this cipher is my identity." And a few days later, he sent a letter that said, "By the way, are the police having a good time with the code? If not, tell them to cheer up. When they do crack it, they will have me." But when the cipher was cracked, Zodiac admitted, "I will not give you my name because you will try to slow down or stop my collecting of slaves for my afterlife."
Maybe it's just me, but I don't think we can trust this lying loser Zodiac. Why should we assume he'd just turn himself in with a puzzle? On the other hand, maybe he gave in to the taunts from American Cryptogram Association President Dr. Marsh, who challenged Zodiac to put his name in a cipher.
Maybe Zodiac felt his name was safe, hiding behind an unusual cipher scheme.
We simply don't know what Zodiac did.
The cipher message could be something completely different from what Bab expects like what? Let's go through some possibilities.
One, it contains the name of a real person and it's documented somewhere. If so, then you can exclude names that don't show up in those records. This is Alex Bab's assumption as well as many other people's. Two, the cipher has the name of a real person, but it's not documented anywhere. Public records aren't 100% accurate or might be hard to get. For all we know, Zodiac was a nomad and lived off the grid, or he was in and out of mental institutions whose records are harder to get. Three, it contains the name of a real person, but Zodiac misspelled it, either by accident or on purpose, maybe to throw us off. He made tons of spelling errors elsewhere. If you want to find the name, you'd have to guess how he might misspell it. Four, the cipher contains an alias or a nickname, then there's a good chance it wouldn't even show up in public records.
This could be another way Zodiac screws with us with his dumb games. Five, the cipher contains a real name, but it's not his name. He could be making a joke or framing someone. Zodiac would get a chuckle if authorities cracked the cipher and paid a visit to some innocent person. Six. The cipher says who he's not. Remember how Z340 says that wasn't me on the TV show?
>> My head aches. I'm so sick.
I'm going to kill them. Take care of yourself, Sam.
>> What if Z13 really says, "My name is not Sam from TV." Seven. What if he just gave another clue instead of giving his name directly? Imagine this scenario.
Zodiac hid his name somewhere in the previous cipher Z340. Then he sent Z13 which says, "By the way, have you cracked the last cipher I sent you? My name is mentioned in it." Zodiac could be screwing with us again, sending us on another wild goose chase, like for his death machine. Eight. The cipher could say something completely useless, like his first two ciphers turned out to be, boasting about his crimes and elusiveness.
My name is Never to be seen. Or Dr. Garlic's solution from episode 23.
Doctor, eat a torpedo.
Nine. There's no decryption. The cipher is just decoration. Random symbols.
Unintentional deception by Zodiac. I don't think this is likely since we had real messages in his first ciphers, but we don't know for sure.
The point of all this is Baber has not excluded these possibilities. He's assuming Z13 has to decode to a real name. If he narrowed down a list of people, it doesn't really mean anything because he hasn't ruled out these other options. He can only say he's simply not interested in those. It boils down to personal preference and not something that we can fully test.
Baber's method makes it easy to find solutions for each of those options since Z13 is so short and he added the transposition step. I do think Zodiac might have included transposition because we know he used it in his 340 cipher, but it makes so many more solutions possible and we can't rule them all out. Let's explore some of them.
There are lots of ways to transpose the cipher. about 130 million ways. In fact, why did Babber pick this grid-based one?
I think it's because he liked the symmetry that comes out of it since each column has only one type of character, a letter or a symbol. Or maybe it just happened to work out that way when he found the name. Just to review, here's how Bab's method finds Marvin Merrill's name. Starting with his grid with columns numbered, we arrange them into this permutation.
then apply his substitution key and we can read Marvin Merrill.
Can we find other names with his original grid? Of course, with different permutations and substitution keys, we can find these, James Isabel, Joseph Johnson, Robert Bennett, Michael Laam, and Charles Hirs.
So far, I've found close to 80,000 different names just using Baber's steps. How can he be 100% sure Marvin Merrill is irrefutably the only correct one from so many? Many of the names are real people, and I'm sure many aren't.
But does it matter? Like I said earlier, Zodiac may have used an undocumented name or misspelled name or a fake name or alias. We simply don't know. You can't dismiss a name just because you can't find any information about it. And names aren't the only solutions. Let's look at some other interesting decodings that come out of Bab's method. Did Zodiac give himself another nickname?
Dreaded Ripper or Eric the Killer? This one's cryptic.
My name is literally dead. Or a little more specific, my name is literally Gary. What about my name is Fred and I'm a man? There have been a few Zodiac suspects named Fred. Another one like that is, "My name is Nathan. I am a man." What about, "My name is also a male name." Maybe he was being obvious as a joke. How about another suspect? My name is Deadeyed Allen. Maybe that refers to suspect Arthur Lee Allen. Or maybe he's just saying, "My name is an abomination."
Here's a good one. Imagine Zodiac hid his name somewhere in the Z340 cipher.
Then he sent the Z13 letter which says, "By the way, have you cracked the last cipher I sent you? My name is mentioned in it." Makes sense, right? Keep the wild goose chase going. Another one that works as a little clue is my name is written within. That could mean somewhere within the Z13 letter, right?
Maybe in the bomb diagram it came with.
How about my name is encoded inside? Is he saying who he isn't? My name is never Jeffrey. I think this solution says it best. My name is known by nobody.
Well, except for Alex Babber. Is it fair to reject all of these alternate solutions that use his technique just because they don't say what he wants them to say?
But NSA cryptography experts who analyzed his solution endorsed it.
Rich Wiznookski, Patrick Henry, >> this is Patrick Henry, I should say, another cryptographer uh who has background at GCHQ and uh uh at NSA, I think.
>> And Ed Giorgio.
>> Ed Gorgio, who over his 30-year career at the National Security Agency is the only person to have ever served as both the chief code maker and codereaker for the NSA.
What are they seeing that I'm not? Their public report can be read here in an interactive notebook showing their analysis and experiments.
They mentioned a way they rejected names. Since survivors describe the killer as white, reject surnames for which less than 10% of those having it are identified as white. Is this reasonable? What if the real Zodiac happened to be in that 10%.
Also, again, there's no guarantee he used his own real name or even a name at all. They say his method is consistent with historical cryptography texts and Zodiac would have had access to Helen Gain's book on cryp analysis. The chapter describes a similar scheme. The writing in or the taking out of a text is said to be done by straight horizontals or by reversed horizontals backward or by alternate or alternating horizontals written alternately in both directions. The alternating horizontals in both directions matches what Bab did to set up his grid.
The book also describes a column-based transposition which is also similar to what Bab did. Gaines shows how a keyword can be used to specify the column ordering of these grid permutations.
Let me explain how Helen Gaines example works. In the example, we're encrypting this message. Let us hear from you at once concerning jewels. We're going to use columnar transposition. So, we need a keyword and a grid. Our keyword is Halifax, a sevenletter word. So, in this scheme, that means the grid has seven columns. We write our plain text into this grid, left to right, top to bottom.
There are some leftover spots at the end. Sometimes we ignore those, but in this example, we're going to use filler to even out the block, similar to what Baber did with his grid.
Next, we need to assign the numbers 1 through 7 under this sevenlet keyword.
Let's look at the letters in alphabetical order. A comes first in the alphabet, so we assign it number one.
Then there's another A, so let's just assign it number two. The next letter in alphabetical order is F, number three.
Then comes H number four. I is next number five. Then L number six. Finally X number seven. To make the cipher text, we read along the columns one at a time and write out the letters. Start with number one. Write those letters into the cipher text.
Then number two. Write those out into the cipher text.
Then number three.
Then four. Then five. Six. and finally seven. The resulting cipher text looks very cryptic. To decode the cryptic text, we have to reverse the process. We know the keyword Halifax and that it's seven letters long. So, we know the grid has seven columns and our keyword goes on top. But how many rows? That's just the total cipher text length, 42 / 7, which is six. We repeat the numbering process which is marking the letters of the keyword in alphabetic order. Now we start writing the cipher text back into the columns. The first six letters go into the column marked one.
Then the next six letters go into the column marked two.
Then these go under three and so on until the grid is filled up.
Now we can read the message again.
Let's go back to Bab's decryption of Z13. Once he writes Z13 into the grid, he rearranges the columns 1 through 7 into a new permutation 3 4 5 7 2 1 6.
Then he applies a substitution key and finds Marvin Merrill's name. The NSA cryptographers figured out they can make this specific permutation using a single important keyword, Elizabeth. as in the name of the Black Dalia victim, Elizabeth Short.
>> It was a stunning discovery. Marvin Merrell had used the name of the Black Dalia as the keyword to a Zodiac cipher.
It was another irrefutable connection between the cases.
Irrefutable might be putting it too strongly. Let's take a closer look.
Elizabeth is nine letters long, but we only have seven columns in Bab's grid.
So, how do we apply the keyword? The NSA folks took some liberties here.
Elizabeth has two E's, so they decided to simply ignore the first E. Now we're down to eight letters. Since Z would be the last letter in alphabetical order, they just drop it. Since they already have seven other letters, the resulting modified keyword is liab. Bab's column order is 3457216.
How can we get those numbers from the keyword liab? First, let's write 1 through 7 under the keyword. Then, let's put those letters in alphabetical order and keep those numbers with them.
That gives us a b h i l t and bab's numbers 3457216.
By using the keyword, you don't need to memorize these numbers. Instead, you can just memorize the keyword and put the letters into alphabetical order.
Bab's NSA code experts are super impressed by this keyword at an aha moment. Definitely, that's something that really lends weight to this solution being correct. It's not just an anagram of the cipher text that you then apply a substitution to and get a random name. It's a very structured reordering using the kinds of permutations that we use in standard reference works. Hey, a keyword actually generates the permutation. And what is that keyword?
That keyword is Elizabeth. Oh my god, we got great prior evidence in favor of that.
>> We saw earlier we can find thousands of names just from Bab's grid approach. Do other keywords besides liabth give the same permutation numbers? Absolutely.
It's even easier if you allow dropping up to two letters like they've done with Elizabeth.
First, here are some where you don't have to drop anything. T-shirts without any changes gives the exact same sequence of numbers.
It also works for one girl and old girl on a girl. Spinoso to be fun on a desk. That might reference the creepy desk poem in the Sherry Joe Bates case. Side eye reminds me of the Halloween card. If we start dropping letters like they did, we can find thousands more. Robinson works if you drop an O. Pendleton with E and O left out. Could that be a misspelled reference to Marine Corps base Camp Pendleton? Zodiac did say he would disguise his killings as robberies.
Are these keywords less valid than Elizabeth? I asked Georgio and Henry why they thought the Zodiac would use a keyword that if discovered could connect him to another case. I think he was trying to leave clues and he was trying to make it sort of tantalizing but hard and so that when somebody did finally come up with the solution, they would be able to see how it was constructed and what the keyword was and it wasn't just something made at random.
>> Couldn't some or all of the keywords I found be considered potential clues?
especially if Zodiac is someone whose name has never come up before. These are all just a small sampling of the possible keywords. Elizabeth is just one of many. They ignore the others because it doesn't fit the conclusions they've already made. And keep in mind that every one of these grid solutions has its own big bubble of possible keywords.
The Z13 solution was never viewed as a piece of evidence. It was always viewed as a investigative tool.
I think they are using the tool as part of confirmation bias. They found ways to support their suspect while ignoring all the other possibilities. It's like testing a new medicine, assuming that it's safe and ignoring any tests that show it's hurting people. Hold on a second before we move on from the keyword. We have three scenarios to consider, assuming the box permutation is right. One, Zodiac made a box permutation and used a keyword such as Elizabeth to determine the column ordering. This is what Babers's team proposes. Two, Zodiac made a box permutation and just picked a column ordering completely at random. He left it up to us to figure out the ordering.
Three, Zodiac made a box permutation and the number itself is the keyword. For example, what if 3457216 is just a phone number? While Babers's team focuses on one specific solution, it is highly unlikely they adequately ruled out the tens of thousands of other names and keyword combinations that can be generated from this cipher. Vber needs to share his exact methodology to show he didn't just ignore these alternatives to fit his predetermined suspect. It's entirely possible he was already biased towards finding Marvin Margolus, especially considering back in 2021 he was already clipping Margolus related articles from newspapers. In fact, he was interested in Marvin Margolus as far back as 2019, but I'll get into that later in this video. It's a doozy.
Another claim in the report is that there is some structure to Bab's substitution key and that it might point to possible Zodiac victim Sher Joe Bates. It has to do with a classic way of building the substitution table. When we make a substitution cipher, we need a plain text alphabet and a scrambled up cipher text alphabet. Then we can encode our message by looking in the key for the cipher letter for each plain text letter. But you have to know this key somehow to make or break the cipher.
It's hard to remember this cipher alphabet because it's random. An easier way is to use a keyword. Let's start over. Here are the plain text letters.
We'll pick a keyword. How about my name, David? Write the keyword at the beginning of the cipher letters, but we leave out the duplicate D because we don't want any letters to repeat. Now we have four of the substitutions. To get the rest, we write out all the letters we didn't use in alphabetical order.
That gives us a full substitution table with a mixed up alphabet, and we only have to remember the keyword David to encrypt and decrypt messages. Ed Gorgio and his friends tried to find a keyword that fits Bab's substitution key. Let's start over with the full plain text alphabet and empty cipher text alphabet.
Here's my representation of Bab's substitution key, which produces the name Marvin Merrill from his cipher grid. We can fill out the full key by putting the normal looking cipher letters in their right places. A goes with plain text I, E with plain text V, K with N, M with L, and N with M. But what do we do with these symbols? Where do they go? The code breakers take some wild swings with this. First, since the crosshair symbol is kind of like zodiac's signature, they replace it with Z.
Then Z goes with plain text A. Then they say, "This symbol looks like an arrow pointing down. The closest letter we have to a downward arrow is V." So they replace the symbol with a V, and V goes with plain text E. Next, the circled eight symbol. Babber claims Margolus owned a wartime bayonet and similar bayonets bear the same circle 8 symbol.
They decided to replace the circle date symbol with B because it vaguely resembles the symbol or could stand for the word bayonet.
The rest of the cipher text alphabet is empty since the corresponding plain text letters don't appear in the name Marvin Merrill. Bab's code breakers found multiple keywords that come close to generating Bab's substitution table. One of them is the word tightened, but they started it at the end instead of the beginning. Then they wrote the rest of the cipher text letters backwards. When they compared that substitution table to Babers, they found six matches out of eight symbols.
Another keyword they tested was Sherry Joe, the name of possible Zodiac victim Sher Joe Bates. But they made one small change. They wrote the keyword at the end again, but kept it in a normal reading order. Then wrote the rest of the letters backwards like before.
Bingo. It also leads to six matches.
That keyword seems significant since it relates to Sher Joe Bates. But by their own admission, a handful of keywords have the same exact effect of matching six assignments. I found a lot of others, too. So, what can we even conclude from this? that Zodiac used a keyword like Sherry Joe to set up a substitution table, but it only works for six out of the eight symbols in Bab's key. The code breakers severely underestimate how many keywords can be used to reconstruct Bab's Z13 key. First of all, half of those six matches are guesses where they replaced these symbols with normal letters. Crosshairs with Z, this one with V, circled eight with B. But those aren't set in stone, right? They can't read Zodiac's mind.
Let me add my guesses, too. The crosshairs could be T or O. In the Z48 key, the same symbol stood for D. And in Z340, the symbol decoded to A. The circled 8 could be an O or E, which stands for 8. If you squint at it, it kind of looks like an H or an X or even an S. The upside down T could simply be T. Z408 had an upside down T which decoded to R or this very similar Ishaped or hookshaped symbol with the same curved bottom which decoded to O and Z340 had similar symbols upside down T which decoded to L and these hookshaped symbols which stood for T.
Isn't it fair to test all of those guesses too? They only tested one possible set of guesses but at least 180 of them are possible.
They also didn't test all the possible schemes for using the keywords to make cipher text in the substitution table.
Let's go through all the different choices. We can write the keyword at the beginning both forward and reverse. We can write the keyword at the end both forward and reverse. And we can write the remaining alphabet both forward and reverse.
That's eight different ways.
But the codereers report only includes these two.
But wait, there's more. You're not limited to making the cipher text alphabet using a keyword. You can do the same thing to the plain text alphabet.
So all those options apply there, too, leading us to 2880 combinations and even more because of schemes I haven't included, but Bab's team only tested two of them. As expected, we can find lots of keywords that generate most of Bab's key. Simple words like forest, bodies, buried, attracted, ciphered.
Plenty of other names work besides Sher Joe such as Diane, Jennifer, Jodie, Barbara, Grace, Deborah, Jamie, Michelle, and some Zodiac related phrases at Tahoe, attacked her, forced her, chased Jensen, decoded it, die right, and these all relate to Officer Richard Ratatic shot in a parked car. Zodiac tried to take credit for it.
This just shows Babers's Z13 solution isn't conclusive, even with the NSA help. Their grid system, arbitrary choices, deletions, and incomplete keys create thousands of valid possibilities.
With that many options, it's far too easy to cherrypick a partial keyword like Sherry Joe or Liabth just to fit a pet theory. It's also possible Sherry Joe Bates wasn't a Zodiac victim. Plus, their claim that a real solution needs a structured keyword ignores the fact that Zodiac's solved ciphers don't even have them. The code breakers also missed something else big. Other reasonable grid layouts. Let's look at Bab's grid again. He wrote half of Z13 in the first row, then the second half in reverse in the second row, then used this placeholder to make it even. But that's not the only way to do it. Another way is to put it at the bottom like this.
The symmetry is still there. We can also make a grid where we don't need a filler. If we just remove this middle circled date symbol, then we can write the rows and have a nice symmetric grid with no gaps.
Another possibility discussed in the past is that all the circle date symbols are nulls. If we remove them, we can still make a nice symmetric grid like this. I think it's also fair to write the rows in a different order. Bottom row first, then top row.
That gives us eight total variations of Bab's grid. What would make any of them unfair to use other than personal preferences? We simply don't know which system Zodiac preferred.
Of course, we can find so many names from these other grids too.
John Conan, James Salamone, Richard Isaiah, Charles Hayes, William Allen, David Rider, George Cook, and not just names. Here are some phrases that make sense.
My name is really hard, a call to action.
Seen in the post, Lee dear Slayer, and my personal favorite, my name is Horseshit Rex.
If we accept the codereaker's conclusion that Marvin Merrill was the only correct answer, then we have to simply ignore all these other possibilities that they didn't even test.
A Swedish man named Thomas Hefner found something that Ed Giorgio thinks solidifies their case even more. It is certainly getting solid by the date. And now we have additional confirmation from Thomas Hefner. his his initial observation is yet another confirmation that we have it correct. The bottom line is that when you look at the combination of two cipher messages, one name, one keyword Elizabeth, one encryption method, and the potential keyword Sherry Joe, the evidence went from likely to very likely to very very likely to almost beyond a shadow of a doubt likely. They're talking about the unexplained final 18 characters of gibberish from Z408, Zodiac's first solved cipher known as Z18. Is it just junk or is there another secret message?
Hefner says if you remove all the E's, you get 13 characters like Z13. If you misspell Marvin with an L like this, then it has the same frequency signature as those 13 characters.
That means you can rearrange them, apply a substitution key, and get Marvin Merrill. Then Hefner goes into a tortured analysis that forces patterns to appear between his Z18 substitution key and Bab's Z13 substitution key. In other words, he found a substitution that kind of works and then tries to find interesting patterns to hold up as proof that it's right. But the problem is millions of names can be extracted from Z18. Did Thomas Hefner test them to show that interesting patterns don't show up for those too? No, he didn't. I don't really want to do it for him either. Honestly, this is the most exhausting aspect of cryp analysis in this case.
>> I'm tired, boss.
>> Testing ideas is tedious work. Too many people simply don't bother. Hefner just finds the patterns and calls them consistent, but never calculates the chances or even tests his hypothesis.
Alex Baber didn't stop at Z13. He also has a solution to Z32, a cipher that's even worse than Z13 to verify because it's much less restrictive. These are the only repeating symbols. So, all you have to do is come up with something 32 letters long where these pairs of letters match.
Zodiac said zero is to be set to magnetic north. In his little list letter, he included a crosshairs with a thick zero. And also the hint, the Mount Diablo code concerns radians and inches along the radiance. Baber's decryption reads, "Estimate four radian and 3 in."
The Daily Mail explained how he used that plus the other clues to locate Elizabeth Short's burial site. When enlarged, placed over the peak of Mount Diablo, and turned clockwise to 16.9°, the magnetic north coordinate for 1970, as Zodiac had instructed, the darkened zero positions directly over Mountain View Cemetery in Piedmont, California.
It looks like he estimated the angle as 4.2 radians, which is about 240°, and the length as 2.62 in, which is roughly similar to the measurements in his solution.
Babber admits that personal bias drove the solution.
>> We went into it with a from a perspective of a bias, meaning that okay, if we know that and we know the information surrounding the phone call to the Oakland PD on the date and with the birthday correlation with Belly in the hometown, could the Z32 somehow be connected to Elizabeth? That came at it from a biased point of view.
What he's really admitting there is that he ignored all the other possible solutions because they don't connect to Elizabeth Short. And there are vastly more solutions for Z32 than there are for Z13.
Mathematically, fitting a message to Z32 is about 12 times easier than Z13.
Since verifying Z13 solutions is already virtually impossible, Z32 is 12 times worse. With so many degrees of freedom, we're stuck dressing up solutions with embellishments like Baber's location method and picking favorites instead of proving anything.
Z32 deserves its own video, which I hope to do someday.
Okay, let's summarize what I think we're looking at. According to Alex Baber and his gang of codereakers, Zodiac encrypted the name Marvin Merrill by creating a substitution key based on the possible keyword Sherry Joe, which only makes 75% of Bab's key, and did a grid permutation using the keyword Elizabeth, except you have to remove the Z and the first E. And Z18, if you squint at it hard enough, corroborates this because you can force Marvin Merrill's name to fit. and Z32 can be made to point to Elizabeth Short's grave. Meanwhile, every step of this process produces thousands of other names, phrases, and keywords, some of which they say they've tested and eliminated, and others they haven't considered. Bab thinks using two steps for Z13 actually makes the appearance of Marvin Merrill's name more remarkable. So, what's the odds that somebody like me would have to hunt down a prime suspect in another famous case, say, Elizabeth Short, that has a guy named Marvin Margolus, who happens to have an alias, this Marvin Murrell, that happens to fit the the letter count for the Z13, which happens to come out when you apply not one but two layers of cryptography like the Z340.
I pointed out earlier that including Bab's second layer of transposition actually makes it more likely for extra names to appear. In fact, it makes 5,000 times as many names appear.
I'm not saying Zodiac didn't use two layers. The sky's is the limit since he used two layers in Z340. But having two layers doesn't improve your solution's chances of being right. And the second layer could be a different transposition than Bab's grid. The other bad assumption Baber makes is that if Z13 did reveal a name, we'd recognize it as a suspect in some other case. Why?
Couldn't it be some other unknown person that isn't on anyone's radar?
Joseph D'Angelo, the Golden State Killer, was active between 1974 and 1986 and wasn't identified and arrested until 44 years later. His name never turned up before as a suspect. And again, there's no guarantee Zodiac even encoded his real name to begin with.
Bab told me that renowned cryptographer Bruce Schneijider confirmed his solution, but pointed out that Z13's shortness leaves it open to other possibilities. When I asked Bruce what he thought, he said he told Baber about the same flaws that I found, and that's why Schneider isn't in any of Bab's press materials like the other cryptographers.
Bab calling this a confirmation is misleading.
I'm surprised that those other cryptographers have endorsed this solution and aren't concerned with the serious problems I've pointed out.
>> Giorgio told me that Baber belongs on the Mount Rushmore of Zodiac breakers.
He added that his team's conclusions withstand the popular internet claim that a cipher as short as 13 digits is unbreakable. So, a short cipher about which nothing is known often cannot be broken. But the fact that such a cipher sometimes cannot be broken is not the same as saying it can never be broken, especially when real world constraints are involved. And for some ciphers, under sufficient conditions, a solution can nearly always be found.
I mean, if you set up the real world constraints to favor the name Marvin Merrill above all the other thousands or millions of names and phrases that work with Z13, then of course your solution will be the right one. They said this in their analysis. Knowing what we now have as a plausible solution, let's examine how surprising this solution is when we take some of the assumptions it suggests we should have used all along.
Incredible. They're using circular reasoning to justify the solution by coming up with assumptions about it after the fact.
This is like if Bab shot an arrow at a blank wall and the NSA crypt analysts came and drew a target around the arrow and declared bullseye and then ignored all the other attempts.
But maybe my analysis is flawed. What did I get wrong? What am I missing?
Please let me know. Leave comments or send me an email.
I reached out to some other cryptography experts to get their opinions, including a retired NSA cryp analyst I know.
Unlike his former colleagues, he doesn't think highly of Baber's solution and says their endorsement paints us in a bad light and that if he was somehow associated with Bab's solution, it would stain his reputation.
Bill Brier used to provide professional codereing services to the US government.
He noted that one of Conny's legitimate complaints is that some people are shooting down the entire case without looking at the evidence. Brier has carefully listened to Conny's podcast and has studied Giorgio's analysis. He believes that Connelly is dismissing the legitimate concerns of outside experts by labeling all critics as amateurs or naysayers.
Brier wrote, "They made an unsuccessful attempt to produce a Sher Joe keyed substitution alphabet using a highly subjective assignment of symbols to letters and then they somewhat clunkily employed the keyword Elizabeth and a disposable null in a columnar transposition. Other methods can produce other suspects names. Ultimately, the claimed solution depends on Marvin Merrill being the Zodiac Killer. Baber's evidence and proposed solution are entirely unconvincing in my opinion.
Bill also said, "Even when a specific method can be shown to yield only one suspect's name, there are plenty of combinations of other systems and keys that also yield only one suspect's name.
People have sent me solutions to Zodiac's short ciphers for many years.
There's a good chance there are examples in there of systems and keys that yield only one suspect's name." Another codereaker told me, "Baber's solution and other similar ones are like magic tricks where the decipherment steps disguise the flaws of the solution, and you even have to decode the decipherment steps to figure out what's wrong with them." He also said, "Everyone should wait for the FBI to verify the solution because if that's missing, you have nothing."
Prolific cryptologology author Klaus Schme said Baber's solution is just one of many possible solutions, but there is not enough data to test which key is correct. He affirms that Elizabeth is just one of many possible transposition keywords and that Sher Joe is one of many possible substitution keywords. He concludes that Baber's solution is indistinguishable from coincidence.
Computer scientist and cryp analyst George Lazri, known for leading a team that cracked over 50 encrypted letters from Mary, Queen of Scots, is very bearish on Z13. He said he thinks Zodiac took the cipher's secrets with him to the grave and that there is no value to any decryption due to its short length.
Satoshi Tomo is a patent engineer with an astrophysics degree who has also written many papers about historical cryptography and manages cryptologology site Crypana. He was also on the team that cracked the Mary Queen of Scots letters. When he took a look at Bab's solution and the former NSA cryptographers report, he agreed with them that Z13 could be analyzed by the techniques that they used. He also agreed with their assessment that the Z13 solution wasn't conclusive, but they had pointed out a chain of incredibly strong coincidences. However, he says it's just one of many possible solutions and wonders how many other candidates might be out there. He also suggests someone could test Baber's method by trying to find Ed Giorgio's name or some other relevant name using the same scheme.
I also reached out to a former FBI codereaker. He says the solution looks highly contrived, relying heavily on confirmation bias to force a predetermined outcome rather than following good crypt analysis. The methodology is riddled with arbitrary manipulation, such as randomly using a null character and an incomplete transposition keyword and baselessly assuming Z13 is a simple substitution cipher despite Zodiac's history with homophonic substitution. He also believes they completely ignore the mathematical limits of unicity distance.
The extremely short cipher text inherently allows for hundreds or thousands of valid name matches, a flaw confirmed by his own database testing.
He also criticized their use of an incomplete demographic data set since the Social Security Administration data is limited. He concludes that their analysis reads as if they accepted the conclusion and worked backwards to find things to support it while ignoring whatever doesn't fit.
I invite any other cryptography experts to chime in, leave comments, or send me an email.
Bab doesn't just hold up his three NSA codereers. He told me he had support from a significant number of cryptographers now totaling in the double digits. Who are they? Where are they? What do they see that confirms Baber's solution that I can't see? I heard him boost the number quite a bit.
>> We got 50 plus individuals that are experts that have come to us since our release of the notebook, meaning the the Python notebook showing the decryption methodology.
Ed Gorgio said, "The probability that anything else is correct is orders of magnitude smaller. It is the greatest sleuth story ever told.
>> If you were on a jury, you would convict uh Margolus of uh both cases. I suppose >> absolutely.
>> I just don't see how anyone could say that in the face of the endless possible decryptions, permutations, and keywords.
The only way to make that conclusion is to already be convinced that Marvin Margolus is both Zodiac and the killer of Elizabeth Short and that somehow all the other solutions don't have the specific qualities you're looking for.
So when people out there assert, hey, you can make this decrypt to any name you want. They are correct. But can you make it decrypt to something that has all these other attributes like we had prior knowledge that this was a suspect?
Like we had prior knowledge of the encryption system, like we had prior knowledge that this was this person called Elizabeth. You just can't put all those three things together and come up with a believable answer. You have to take all three as a package.
What about we had prior knowledge Zodiac said he wouldn't give his name or that he'd be more likely to play games with us instead of being direct or that we wouldn't recognize his name because it never came up in investigations.
There's another interesting possibility.
What if we caught Zodiac and he did turn out to be Marvin Merrill, but he didn't actually put his name in the Z-13 cipher? Just because we found a name in it doesn't necessarily prove it's there on purpose. That's just how much doubt there is caused by the thousands of alternate names that fit.
>> You have to figure out what the last solution is and if that solution is in fact probable and can be sustained. mean that we can go out and find evidence around this individual that identifies him as being the perpetrator of the crime, right?
>> We know the BTK killer was Dennis Raider. He put out his own short cipher, too. Did he hide his name there? I bet if we poked at it long enough, we could make it cough up a version of Dennis Raider's name, but it wouldn't be right because we already know the real solution.
Paraphrasing Ronald Coos, if you torture the data long enough, it will confess
Related Videos
VALORANT's Latest 'Exclusive' Tier Bundle is Rough...
KangaValorant
17K views•2026-05-28
Flight Attendant Mocks Poor Looking Black Woman — Mid Air Announcement Exposes Her Real Power
SkyboundStories-b4r
184 views•2026-05-28
I FIXED My Friend’s Blown Turbo RX-8… Then Sold It
Cameron-RX8
134 views•2026-05-28
NewsWatch 12 at 5: Top Stories
NewsWatch12
1K views•2026-05-28
Simon Jordan & Danny Murphy deliver PREDICTIONS for Arsenal's Champions League FINAL with PSG
talkSPORTArsenal
6K views•2026-05-28
Botting is OUT OF CONTROL in Classic WoW (Again)...
SolheimGaming
108 views•2026-05-28
The "AI Job Apocalypse" is CANCELLED!
WesRoth
9K views•2026-05-28
STREET FIGHTER 6 - INGRID Story Walkthrough @ 4K 60ᶠᵖˢ ✔
RajmanGamingHD
12K views•2026-05-28
