Thread:Marvelus/@comment-27496405-20190519011243/@comment-26838855-20190527223619

Introduction
OK, so I broke down the 2018 problem with all the evidence we have now and how Avengers: Endgame impacts The Punisher: Season 2's effect on this problem.

I have also provided some weightings to the evidence. They are not concrete, this is like Cloak & Dagger just providing guidance - more than you would have without. I have also in doing this pinpointed the various things in Agents of S.H.I.E.L.D. that make the gap since Principia feel small, and have broken things down into soft evidence (not based on actual evidence but rather a strong feeling or implication) and hard evidences (dates and numbers etc.).

Firstly, All the Comforts of Home, The Real Deal, and Principia are firmly December 2017 for five reasons: Even if The Devil Complex to The End were to be moved right to June 2018, it works better to have a 6-month jump (even though that's hard) and keep those first three episodes in December 2017 than move them to Spring 2018. So moving those episodes from December 2017 is not in question.
 * The Real Deal has a "12-06-2017" date.
 * Principia has a "12-17-2017" date.
 * The agents are said to have only been gone for "months".
 * "Tidings of comfort and joy" line implying Christmas-time.
 * All the Comforts of Home refers to Thor: Ragnarok ' s events, implying them to be very recent.

Agents of S.H.I.E.L.D. Soft Evidence for Early 2018

 * In The Real Deal, the team discover that Coulson is dying. For them, it has been only a few days since Coulson made the deal with Ghost Rider, yet it has become quite bad. They do not know when he will die, but considering it took just a few days to get this bad, it would be strange for it to take up to 7 further months to kill him.
 * How could this be explained away if it's mid-2018?: The progression may be extremely variable, and there may also have just been an initial quick burst from the deal with Ghost Rider making things fairly bad, and the rate after that is slow. Fitz and Simmons might also have been able to help just a tiny bit, though from their dialogue, it seems unlikely.
 * How close to mid-2018 can the Snap realistically be?: I can't see him surviving more than 3 months before his heart stops in The End (they then resuscitate him for a little bit longer). If he survived 3 months after the team find out on December 6, 2017, so about March 6, 2018, then the Snap could've been no later really than the Thursday closest, March 8, 2018.
 * How strong is the evidence?: 4/10. How convincing is the explanation?: 3.5/10. Overall: 4 - (3.5 ÷ 10) × 4 = 2.6.
 * In Principia, Fitz says in December 2017, "The, uh, the gravitonium device that we used to seal the rift is barely strong enough. It's duct tape on the Hoover Dam. It's not gonna hold." Coulson asks, "How much time before the dam starts leaking?", and he says, "Well, uh, it's hard to calculate that precisely, but, based on the increased frequency of transit, electromagnetic waves, I'd say that our days were numbered." While he clearly doesn't know exactly how long they have, the implication is absolutely no more than a couple of weeks.
 * How could this be explained away if it's mid-2018?: It's unpredictable, and maybe Fitz managed a few things between Principia and The Devil Complex to buy them more time, adding more and more "duct tape" to that "Hoover Dam".
 * How close to mid-2018 can the Snap realistically be?: I can't see it lasting more than a month after Fitz's comment on December 17, 2017, so around January 17, 2018. The Devil Complex is 2 weeks before the Snap, so January 31, 2018, then the Snap could've been no later really than the Thursday closest, February 1, 2018.
 * How strong is the evidence?: 5/10. How convincing is the explanation?: 4/10. Overall: 5 - (4 ÷ 10) × 5 = 3.
 * In The Devil Complex, Simmons says she would like to have a honeymoon with Fitz, implying they recently married. In The Honeymoon, Simmons jokes about this being their honeymoon, implying they recently got married.
 * How could this be explained away if it's mid-2018?: Fitz and Simmons have been confined to the Lighthouse and occupied with the rift problem. Some people delay their honeymoons because they have to.
 * How close to mid-2018 can the Snap realistically be?: I can't see it being up to 4 months after their December 6, 2017 wedding, so then The Honeymoon would be no later than April 6, 2018, and the Snap is 11 days later so around April 17, 2018, so the Snap could've been no later really than the Thursday closest, April 19, 2018.
 * How strong is the evidence?: 1/10. How convincing is the explanation?: 9/10. Overall: 1 - (9 ÷ 10) × 1 = 0.1.
 * In The Devil Complex, Mack is testing Yo-Yo's newly made robotic arms. They got the arms from Hale's mechs in Principia. Considering how fast they usually work, it really shouldn't take them very long to make the arms.
 * How could this be explained away if it's mid-2018?: Fitz and Simmons are likely devoting most of their time to the more urgent and severe problem of closing the rift. Mack is also helping, but they are also limited by their situation in terms of resources, and maybe wanted to be especially cautious to make sure the arms didn't have any underlying evil programming.
 * How close to mid-2018 can the Snap realistically be?: I can't see it being up to 6 weeks after they got the arms on December 18, 2017, so then The Devil Complex would be no later than January 29, 2018, and the Snap is 14 days later so around February 12, 2018, so the Snap could've been no later really than the Thursday closest, February 15, 2018.
 * How strong is the evidence?: 3/10. How convincing is the explanation?: 4.5/10. Overall: 3 - (4.5 ÷ 10) × 3 = 1.65.
 * In The Devil Complex, Mack says to Yo-Yo, "Your arms aren't fully healed yet. What you need is time." Yo-Yo: "My arms may not work, but my eyes do. At least let me sit in Control, keep an eye out for anomalies." Mack: "Maybe, but you have to allow yourself to heal. We'll handle the fighting for now." As well as this, Yo-Yo is still in her hospital-esque bed for a lot of the episode. This implies that Yo-Yo is still healing and still needs to be taken care of, which shouldn't be for more than a couple of months absolute maximum.
 * How could this be explained away if it's mid-2018?: Mack's just being overprotective, Simmons is being overly cautious, Yo-Yo's also struggling psychologically, and they also need to keep working with her to program the arms and don't want her to get going again until the arms are ready.
 * How close to mid-2018 can the Snap realistically be?: Yo-Yo should really be out of her bed by 2 months on after December 4, 2017, so then The Devil Complex would be no later than February 4, 2018, and the Snap is 14 days later so around February 18, 2018, so the Snap could've been no later really than the Thursday closest, February 15, 2018.
 * How strong is the evidence?: 5/10. How convincing is the explanation?: 3.5/10. Overall: 5 - (3.5 ÷ 10) × 5 = 3.25.
 * In The Devil Complex, Deke's still a bit dazed about having worked out that Fitz and Simmons are his grandparents.
 * How could this be explained away if it's mid-2018?: Deke might have just been mulling that over for a while.
 * How close to mid-2018 can the Snap realistically be?: Deke should really be less dazed after a month or so after his discovery on December 18, 2017, so then The Devil Complex would be no later than January 18, 2018, and the Snap is 14 days later so around February 1, 2018, so the Snap could've been no later really than the Thursday closest, February 1, 2018.
 * How strong is the evidence?: 1.5/10. How convincing is the explanation?: 5/10. Overall: 1.5 - (5 ÷ 10) × 1.5 = 0.75.
 * In The Devil Complex, Fitz says, "Well, I've temporarily plugged the dam, but it could burst at any minute, releasing more of these anomalies into the base." The phrasing implies the temporary plugging was recent.
 * How could this be explained away if it's mid-2018?: Fitz has temporarily plugged the dam again more recently but it's a losing battle, which is also aware why now it's "any minute" when it wasn't so urgent before.
 * How close to mid-2018 can the Snap realistically be?: The implication from the tone of the dialogue is that it's no more than a month since the patching on December 6, 2017, so then The Devil Complex would be no later than January 6, 2018, and the Snap is 14 days later so around January 20, 2018, so the Snap could've been no later really than the Thursday closest, January 18, 2018.
 * How strong is the evidence?: 2.5/10. How convincing is the explanation?: 5.5/10. Overall: 2.5 - (5.5 ÷ 10) × 2.5 = 1.125.
 * In The Devil Complex, Coulson says to Piper, "Good first stop on your redemption tour," so this is her first chance at redeeming herself, which shouldn't have taken potentially up to 6 months.
 * How could this be explained away if it's mid-2018?: It's fine, not much seems to have happened so there's no particular chance it would appear she could have had - plus it probably took them a little while to begin to trust her again anyway.
 * How close to mid-2018 can the Snap realistically be?: This could really be anywhere up to 6 months after her betrayal on December 4, 2017, so then The Devil Complex would be no later than June 4, 2018, and the Snap is 14 days later so around June 18, 2018, so the Snap could've been no later really than the Thursday closest, June 21, 2018.
 * How strong is the evidence?: 0.5/10. How convincing is the explanation?: 8.5/10. Overall: 0.5 - (8.5 ÷ 10) × 0.5 = 0.075.
 * In The Devil Complex, Daisy asks Fitz, "Are you okay? Have you even slept?" Fitz responds, "During what possible period of time, Daisy?" This seems to imply that he hasn't been sleeping much in the days since they first plugged the rift, desperately trying to find a way to resolve it. But we know because of Principia and because of Yo-Yo's arm stumps that it's been at least a few weeks since The Real Deal at any rate so he can't literally be referring to having not slept since they got back, and has to really just be referring to recently, or even since he last spoke to her.
 * How could this be explained away if it's mid-2018?: He's just referring to recently when he's been frantic.
 * How close to mid-2018 can the Snap realistically be?: This could be up to a month after the closure on December 6, 2017, so then The Devil Complex would be no later than January 6, 2018, and the Snap is 14 days later so around January 20, 2018, so the Snap could've been no later really than the Thursday closest, January 18, 2018.
 * How strong is the evidence?: 1/10. How convincing is the explanation?: 8/10. Overall: 1 - (8 ÷ 10) × 1 = 0.2.

How close to mid-2018 can the Snap realistically be?: ([March 8, 2018] × 2.6 + [February 1, 2018] × 3 + [April 19, 2018] × 0.1 + [February 15, 2018] × 1.65 + [February 15, 2018] × 3.25 + [February 1, 2018] × 0.75 + [January 18, 2018] × 1.125 + [June 21, 2018] × 0.075 + [January 18, 2018] × 0.2) ÷ (2.6 + 3 + 0.1 + 1.65 + 3.25 + 0.75 + 1.125 + 0.075 + 0.2) = February 13.49020, 2018.

How strong is the evidence?: 2.6 + 3 + 0.1 + 1.65 + 3.25 + 0.75 + 1.125 + 0.075 + 0.2 = 12.75.

Agents of S.H.I.E.L.D. Hard Evidence for Early 2018

 * In The Devil Complex, snow is shown on the ground, suggesting it is still winter.
 * How could this be explained away if it's mid-2018?: They're just fairly high north. Even so, you could be hitting summer which is tricky.
 * How close to mid-2018 can the Snap realistically be?: This could be up to the end of March really, so then The Devil Complex would be no later than March 31, 2018, and the Snap is 14 days later so around April 14, 2018, so the Snap could've been no later really than the Thursday closest, April 12, 2018.
 * How strong is the evidence?: 3/10. How convincing is the explanation?: 4.5/10. Overall: 3 - (4.5 ÷ 10) × 3 = 1.65.
 * In Rise and Shine, Coulson says to Hale, "Listen, my team wasn't hiding all these months. We traveled to the future." From this, it can be taken that it has only been "months" since May 12, 2017, when the agents were sent to the future. While it has to now be 2018, it should not be later than early 2018. As well as this, this piece of dialogue suggests that the amount of time that his team have been back in the present is negligible, because he is generalising the whole time since May 12, 2017 as the time that they were not around, but in the future. It should at least be less than a month so their time back doesn't consist of a "month" that could be added to the "months", for the "months" to only encompass their times away. If they'd been back more than a month he should say "all those months".
 * How could this be explained away if it's mid-2018?: While it's not the end of the world for Coulson to call 13 months "all these months", the other part you can't really justify. You can just assume that Coulson misspoke and since they were gone slightly longer than they've been back, he can generalise the time as the majority being when they were away, but that's poor and doesn't really make sense.
 * How close to mid-2018 can the Snap realistically be?: This could be up to just under a month since they returned, so then it would be no later than January 3, 2018, and the Snap is 13 days later so around January 16, 2018, so the Snap could've been no later really than the Thursday before, January 11, 2018.
 * How strong is the evidence?: 7/10. How convincing is the explanation?: 2/10. Overall: 7 - (2 ÷ 10) × 7 = 5.6.
 * In Missing Pieces, Enoch says he was going to protect Fitz for the next "73 years, 261 days". The team are in "Earth Year 2-0-9-1" in Season 5, Part 1. Plus that's supported by the 2017 + "74 years later". If they left the future as early as possible, December 31, 2091, then since Fitz wakes 4 days prior, the latest he can possibly have woken is December 27, 2091. Even if Fitz were to wake as late as possible, the latest this scene could be is then April 10, 2018. Except the scene is implied to a) be after The End, just due to the way these stories are told, and b) in the new timeline, so after The End, because it can't really have happened in the original timeline, which it would have to if it was pre-timeline split.
 * How could this be explained away if it's mid-2018?: I've been making notes about the exact details of what happened after the ship was split in two to try and work out a way that this could have also happened in the original timeline and the divergence point only occurred a while after the ship is split, at the time of The End. So far, it's not impossible.
 * How close to mid-2018 can the Snap realistically be?: The timeline parting, the same day as the Snap, would have to be no later than April 10, 2018, so the Snap could've been no later really than the Thursday before, April 5, 2018.
 * How strong is the evidence?: 8.5/10. How convincing is the explanation?: 4/10. Overall: 8.5 - (4 ÷ 10) × 8.5 = 5.1.
 * In Fear and Loathing on the Planet of Kitson, Enoch says he was going to orbit Jupiter with Fitz for the next "72 years, 312 days". The team are in "Earth Year 2-0-9-1" in Season 5, Part 1. Plus that's supported by the 2017 + "74 years later". If they left the future as early as possible, December 31, 2091, then since Fitz wakes 4 days prior, the latest he can possibly have woken is December 27, 2091. So the latest Season 6, Episode 3 could possibly be, if Fitz were to wake on the latest possible date of December 27, 2091, is February 18, 2019. But the show is 13 months on from Season 5 (Coulson lived longer than expected after the Season 5 finale so lived closer to the "weeks" Simmons mentioned, plus it's mentioned that Fitz died more than a year ago), so Season 5 would have to finish in January 2018.
 * How could this be explained away if it's mid-2018?: You could argue for it being not Earth years but somewhere else, but the universal translator should be translating to Earth years and it's so close to Earth years. You could argue the destination date changed in the new timeline, but they don't know the timeline has changed nor that the agents won't be there in the future, they think they are - the date shouldn't change. You could argue that Season 5, Episodes 1-4 span months into 2092 so Fitz wakes in 2092 but that's pretty much impossible. This is really reaching. It also doesn't account for the Episode 1 261 days line - sure, that can be separately dealt with as I've mentioned, with its own stretch of the imagination working out how it could fit into the stem timeline for the first couple of months after it, but that's also a stretch and it seems odd to not have the Episode 1 and Episode 3 scenes 314 days apart when they couple like that. Honestly, there's no obvious explanation.
 * How close to mid-2018 can the Snap realistically be?: This is the 1-year anniversary of Coulson's death, so Coulson would've had to have died by February 18, 2018, and he died after at least 1.5 weeks, so the Snap would be no later than February 7, 2018, so the Snap could've been no later really than the Thursday before, February 1, 2018.
 * How strong is the evidence?: 9/10. How convincing is the explanation?: 1/10. Overall: 9 - (1 ÷ 10) × 9 = 8.1.

How close to mid-2018 can the Snap realistically be?: ([April 12, 2018] × 1.65 + [January 11, 2018] × 5.6 + [April 5, 2018] × 5.1 + [February 1, 2018] × 8.1) ÷ (1.65 + 5.6 + 5.1 + 8.1) = February 16.60880, 2018.

How strong is the evidence?: 1.65 + 5.6 + 5.1 + 8.1 = 20.45.

Film Soft Evidence for Early 2018

 * When Thanos attacks the Statesman, it is seemingly reasonably soon after Thor: Ragnarok. This implication comes from the fact that when the mid-credits scene of Thor: Ragnarok begins, minutes before Avengers: Infinity War, Loki is asking Thor if he really thinks it is a good idea to go back to Earth - suggesting they only recently decided to head to Earth. Thor's hair is almost identical in Avengers: Infinity War to how it was cut in Thor: Ragnarok, right down to the poorly cut strips on the side of his head, suggesting there has been little time for his hair to grow. It is unclear how many resources are on-board the ship, and how well the Asgardians would be able to live on the ship, and for how long. But overall, from all these things, the shorter the gap, the better.
 * How could this be explained away if it's mid-2018?: With Loki, it could be argued that they are simply discussing it again as they get closer to Earth. With the hair, it could be argued that with Asgardians ageing slower, or otherwise, Asgardian hair could grow slower than humans' (though by 5 years later he has long hair again so maybe not that much slower). With resources, it is possible that they had resources on-board from Sakaar, or that they stopped off for resources along the way. They're 22 jump points from Asgard and we know from Captain Marvel that jump points aren't just easy to get to - in that case they were a day away from the next one. It could take a while to travel between them. The 22 jump points supports that some time has passed.
 * How close to mid-2018 can the Snap realistically be?: No more than 3 months should really pass from November 21, 2017, so no later really than February 21, 2018, the Snap is the next day so shouldn't really be later than February 22, 2018, the Snap could've been no later really than the Thursday closest, February 22, 2018.
 * How strong is the evidence?: 6/10. How convincing is the explanation?: 4/10. Overall: 6 - (4 ÷ 10) × 6 = 3.6.

How close to mid-2018 can the Snap realistically be?: February 22, 2018.

How strong is the evidence?: 1.65 + 5.6 + 5.1 + 8.1 = 3.6.

Film Hard Evidence Supporting Mid-2018

 * Ant-Man and the Wasp has Scott finishing his 2-year house arrest, which should end on June 20, 2018.
 * How could this be explained away if it's early 2018?: Scott got some sort of good behaviour 3-month reduction.
 * How close to mid-2018 can the Snap realistically be?: The Snap is 8 days later, so June 28, 2018.
 * How strong is the evidence?: 9.5/10. How convincing is the explanation?: 2/10. Overall: 9.5 - (2 ÷ 10) × 9.5 = 7.6.
 * Ant-Man and the Wasp includes a "4/30/2018" laptop date.
 * How could this be explained away if it's early 2018?: Misprogrammed laptop date.
 * How close to mid-2018 can the Snap realistically be?: The house arrest would end on May 2/3, 2018, then the Snap would be May 10, 2018.
 * How strong is the evidence?: 5.5/10. How convincing is the explanation?: 4/10. Overall: 5.5 - (4 ÷ 10) × 5.5 = 3.3.
 * Other time references both in the films and from external sources: 6 years since The Avengers, 4 years since Guardians of the Galaxy Vol. 2, 2 years since Captain America: Civil War, work better with mid-2018 than early 2018, but early 2018 is still possible. Overall, add about 3.5 because it's not about the strength of the evidence but just how much it leans one way.
 * Overall: 3.5.

How close to mid-2018 can the Snap realistically be?: ([June 28, 2018] × 7.6 + [May 10, 2018] × 3.3) ÷ (7.6 + 3.3) = June 13.16514, 2018.

How strong is the evidence?: 7.6 + 3.3 + 3.5 = 14.4.

The Punisher Hard Evidence Supporting Mid-2018

 * The main events of The Punisher: Season 2 take place April 27, 2018-May 17, 2018. There is no way it's just 3 months after the Snap, it should really be before.
 * How could this be explained away if it's early 2018?: The world has just moved on somewhat, very quickly. Highly unlikely.
 * How close to mid-2018 can the Snap realistically be?: Thursday right after the main events, May 17, 2018.
 * How strong is the evidence?: 10/10. How convincing is the explanation?: 0.5/10. Overall: 10 - (0.5 ÷ 10) × 10 = 9.5.
 * The main events of The Punisher: Season 2 take place April 27, 2018-May 17, 2018. If it were after the Snap, you would have to believe Frank, Billy, Curtis, Karen, Matt (referenced), Mahoney, Rafi, Marion, and Turk all survived, which there's a 1 in 512 chance of.
 * How could this be explained away if it's early 2018?: All these characters survived the Snap.
 * How close to mid-2018 can the Snap realistically be?: Thursday right after the main events, May 17, 2018.
 * How strong is the evidence?: 8.5/10. How convincing is the explanation?: 0.5/10. Overall: 8.5 - (0.5 ÷ 10) × 10 = 8.
 * The main events of The Punisher: Season 2 take place April 27, 2018-May 17, 2018. Jeph Loeb indirectly says they take place before the Snap.
 * How could this be explained away if it's early 2018?: The comment is not directly about The Punisher: Season 2. As well as this, as evidenced by his Agents of S.H.I.E.L.D.: Season 6 comments, it's all he really could say. But if it's early 2018, again, you encounter the problems mentioned above, so there's still all those problems so you're not excusing much.
 * How close to mid-2018 can the Snap realistically be?: Thursday right after the main events, May 17, 2018.
 * How strong is the evidence?: 7/10. How convincing is the explanation?: 0.5/10. Overall: 7 - (0.5 ÷ 10) × 10 = 6.65.

How close to mid-2018 can the Snap realistically be?: May 17, 2018.

How strong is the evidence?: 9.5 + 8 + 6.65 = 24.15.

Conclusion
OK, so, the thing about this is that it's not about finding an in between date. Sometimes you try to find a compromise between different bits of evidence to best satisfy things. Sometimes you have to weigh evidence and go one way or another, like Cloak & Dagger, and like this.

I'm about to calculate from the above evidence which way things are currently leaning. Whatever the result is, will have a look at which side it is closer to to see which way makes more sense.

Now, the reason this is an either-way thing is either you completely accept certain bits of evidence or completely reject them. You don't just have Scott's house arrest end 2 weeks before the 2 years is up due to compromise, you either go the full 2 years or you drop that. You don't go 1 week before the end of The Punisher: Season 2 due to compromise, you either go after The Punisher: Season 2 or quite a while before to allow any sort of plausibility in it not being addressed. That sort of thing.

The weights here are not set in stone and more just guidance to how things stack up (after all, this isn't like my Phase One blog or upcoming full project with detailed weightings and leniency etc.) like Cloak & Dagger: Season 1. Except with Cloak & Dagger: Season 1 it was sufficient to show the mostly clear option, and it doesn't seem like this will be so clear.

OK, so there's a few options:
 * January 11, 2018 - As early as possible to maximise the time before The Punisher: Season 2. Ignores Ant-Man and the Wasp.
 * March 29, 2018 - If you decide Scott got his final 6 months halved. However, this is just impossibly close to The Punisher: Season 2. This option just no longer exists.
 * May 17, 2018 - As soon as possible after The Punisher: Season 2. Somewhat aligns with the April 30, 2018 Ant-Man and the Wasp date if you assume 2 weeks between the main events of Ant-Man and the Wasp and Avengers: Infinity War, which is annoying because it mixes up the order with Agents of S.H.I.E.L.D. only to potentially be put back how it was and also because it makes Natasha's line in Infinity War more difficult to swallow, but acceptable on its own.
 * June 28, 2018 - The 2 years of Scott's sentence.

So there are three options here. However, the setup is not a case of "This evidence would ideally place this here and this here" and mathematically calculating the best placement. It's a case of just pulling either way between the two sides - earlier or later. This is why I have included the "How close to [the other side] can the Snap realistically be?", adding in that factor of how much give each side has. And from this, naturally, the result will fall somewhere in the middle - either extreme is already closer to the middle than January 11th and June 28th. It is thus hard to just go, "Well, the result is closest to May 17th so May 17th," because it will be, and that was not the point of this. The point of this was to consider weights and gives of either side then find the point between either end that the compromise would fall at if you were to compromise, and then round to the side it is closest to if appropriate. Therefore I will first treat it as simply the two options of January 11th and June 28th, and then see after I have the result how May 17th might factor in.

The other problem is that each possible point does not have equal precedent. With Cloak & Dagger, it was simpler: You either have February 2015, February 2016, February 2017, or February 2018, because the February is set in stone and we're just weighing up each year. However, here the January 11th does not actually have a specific provided date it is fitting, it is simply the "earliest possible" date. May 17th does not really have a specific provided date it is fitting, though is linked with an actual date - the "4/30/2018". And the June 28th does actually have a specific precedent, as June 20th is specifically from the end of the 2-year sentence and June 28th is the latest that Avengers: Infinity War can end due to Peter's school so they're squeezed in in a more concrete placement. If you go with a result with more specific precedence in the evidence, you are more concretely satisfying a chunk of evidence. Whatever you do you have to throw out some evidence, but if you went for May 17th, for example, you are throwing out either side to some degree, whereas if you go for June 28th you are throwing out the evidence for earlier even more (but some of it is either "it works" or "it doesn't" and if you were to go for May 17th you'd also be throwing it out anyway, so the difference in satisfaction is not all proportional to the distance from the ideal answer) but fully fitting the evidence for later, providing a more satisfying and decisive answer.

Therefore perhaps the June 28th should therefore have a larger catchment area than January 11th - that it's not just rounding either way, but June 28th has a "stronger pull", like a stronger magnet. When you factor in the May 17th as well, the catchment areas of what result should round to what become very complicated. Imagine a case where it's January 11th, June 21st, or June 28th, and there's much more precedent for June 28th than June 21st. If the result were say June 20th, it's not only closest to June 21st but it's also on the other side of June 21st than June 28th is, so you might not even consider that June 28th would come into play. And yet if you're as far as June 20th, it makes more sense to just push it those extra 7 days to fit the one with extra precedent. So you can't even divide the gaps between each possible date in a weighted manner because it's possible the reach of June 28th could extend to even before May 17th, and therefore the reach for May 17th is a short window that actually ends before May 17th. It's all very complicated, and I'm not sure how to account for this "magnet factor" yet.

So, for now, let's just first address it as two sides, January 11th and June 28th, and see what that yields, then just take a specific look at that result and what I feel about it, and then bring May 17th into the picture and have a look at how I feel about that.

So, with that said:

([February 13.49020, 2018] × 12.75 + [February 16.60880, 2018] × 20.45 + [February 22, 2018] × 3.6 + [June 13.16514, 2018] × 14.4 + [May 17, 2018] × 24.15) ÷ (12.75 + 20.45 + 3.6 + 14.4 + 24.15) = April 8.26381, 2018.

Again, this actual date is not important and of course not the one to go with (this is exactly what I was discussing - if you go with April 8th you completely mess up The Punisher and are satisfying neither camp). It's about what it represents: the middle of the tug of war rope between either side, earliest and latest, is 52% the distance from earliest to latest (slightly on latest's side). This is 87.26381 days from January 11th, 80.73619 days from June 28th, so it slightly leans in that direction (plus that direction has more precedence so even if it were 83:85 you might still go in the 85 direction).

But now let's factor May 17th in. As mentioned, the catchment for rounding to May 17th is not necessarily March 15th-June 7th, or whatever similar. The later boundary is not necessarily even after May 17th, as, as mentioned, there is an argument to be made that May 16th might even be worth rounding to June 28th, depending on how much you put into this idea of precedent for each date and, "If it's this far, it might as well go all the way."

However, April 8th? It is hard to argue that once you have introduced the May 17th date, its catchment region ends as early as pre-April 8th/June 28th's catchment region extends so far earlier than May 17th that it reaches April 8th. So it would overall seem that the way the tug of war is going, May 17th is the most comfortable point at the moment.

But this is not necessarily permanent. This is not a clear-cut case with an outstanding big majority result.

As more evidence is added, it could tip things more towards June 28th, or could just make the struggle even more difficult. Certainly for me at least, if the calculation is hitting May 1st or later as the result, you can round that all the way to the June 28th answer. Possibly earlier. All because of that idea of just going all the way to at least satisfy one side more fully rather than in taking a compromise (which May 17th is in some regards), not satisfying either side because while the compromises for the pieces of evidence which are about the distance from the ideal total the same inconvenience either way, if you sit at May 17th you throw out the pieces of evidence which are make or break, either "it is firmly this way" or "it's not", out on both sides, when settling for either side allows you to keep one set at least.

My ultimate conclusion at the moment is this: We still can't definitively place the dates for these events, but we can take some things from this analysis:
 * When you're looking at the evidence as two sides pulling, the evidence only just pulls towards June 28th, but when you consider it as three possible placements, the evidence is definitely far enough towards May 17th at least that we can comfortably for now say it is "May 17th-June 28th, post-The Punisher: Season 2", and that we don't have the option of March anymore (previously saying "March-June"), nor is it likely that evidence will amount to pull things right back to the beginning of the year. This gives us a narrower window to work with, as well as allowing us to more comfortably have the 2018 page structured with The Punisher: Season 2, then these events, then The Punisher: Season 2's epilogue. We might also rename "Spring/Early Summer" to "May/June", again more specific, so progress.
 * With this near-cementing of The Punisher: Season 2 being pre-Infinity War, we can also structure watch orders more conclusively.
 * With this narrowed field of possible Infinity War placements, we can also more narrowly place Agents of S.H.I.E.L.D.: Season 6.
 * When it comes to calculations, such as to do with character ages in Infinity War or whatever, we can use the May 17th date for now. Because those calculations are not about rounding a date one way or another but just getting as close as possible, and May 17th functions for now as, at least, a circa date usable for approximations.

I personally don't find May 17th satisfying for the reasons explained - it can be argued that you might as well go all the way to June 28th. So again, I am not saying this is the placement. But it's a step in the direction towards a placement.

Let's hope Agents of S.H.I.E.L.D. doesn't continue to contradict this with lines like that Enoch one.

And with the inconclusiveness of this analysis, one might ask, "What if we never receive another piece of evidence?" Well, my answer to that is that I would put in a lot of work to undergo a more specific and accurate complicated process to actually make a final decision. But since we have reason to believe that more evidence has a good chance of coming in the near future, I think it is worth holding off on a conclusive decision for now. The results of this analysis were not fruitless, because we can still make some edits to the 2018 page making things more specific, just not all the way.

Also, it's good to collate all the evidence.

Footnote
Out of interest, what would've happened if I'd finished this analysis a few days ago before that Enoch line?:

How close to mid-2018 can the Snap realistically be?: ([April 12, 2018] × 1.65 + [January 11, 2018] × 5.6 + [April 5, 2018] × 5.1) ÷ (1.65 + 5.6 + 5.1) = February 26.84615, 2018 instead of February 16.60880, 2018.

How strong is the evidence?: 1.65 + 5.6 + 5.1 = 12.35 instead of 20.45.

([February 13.49020, 2018] × 12.75 + [February 26.84615, 2018] × 12.35 + [February 22, 2018] × 3.6 + [June 13.16514, 2018] × 14.4 + [May 17, 2018] × 24.15) ÷ (12.75 + 12.35 + 3.6 + 14.4 + 24.15) = April 16.24503, 2018. A little bit more decisively in that later direction. That line is certainly annoying because it's a step backwards when it felt like things were finally heading in one direction.