Best Of

Update:

Anyone still think we're not going to make it?

I think we're going to make it.

The required rate to reach milestone 63 has plummeted to about 2.7 million per hour, which sounds like a lot but that's lower than at any time during this banquet or last year's banquet. As I write this (the charts are a bit behind as I created them a little while ago) we just crossed 7.2 billion. 20 million to go. At the current slightly slower rate I'm seeing milestone 63 should now be reached sometime around 2:30am Pacific, a couple hours later than originally projected.

And here's my current best case / worst case scenario:

Worst case scenario gets us to just past milestone 63 at 7.257 billion and the best case scenario gets us right up to milestone 64 at 7.297 billion. Whatever milestone 64 is, we will at least see it. As far as I'm concerned it could be a gold crystal: my opinion of the banquet will not be changed by it whatever it is. If there is a last minute blitz in the overnight hours we have a long-shot chance at milestone 64, in which case we might see milestone 65 ten seconds before the event ends. Get your cameras ready.

I'd say good luck to us, but milestone 63 is a done deal at this point, and once again I won't be awake to see it. For those that have been following along, thanks for that. I will have a bit more to say about the event after it ends: some additional analysis regarding scoring, spending, and depending on when they come out an analysis of the rank reward cut offs. It might take some time as I still have a bunch of homework to do.

The community was debating this very topic 3 and a half years ago lol

I look at the game.

I don't actually have any special access to the game's data: to the extent I've ever been shown anything in the game data, it would be covered by NDA and I wouldn't be able to share. For Road to the Crypt or the banquet events, I just periodically check the game and record the data, and then use spreadsheets to analyze it. My spreadsheets are ugly, but that's because they match my mental state of the analysis which is equally messy, but its just how I work.

Since you're asking and since I have nothing better to do right now (I've opened about all the crystals I'm going to and are basically out of units and battleships and I've almost maxed the milestones in the current arena set already) I'm going to bore everyone with some analysis technical discussion absolutely no one is interested in right now, but as there's about four hours left to hit milestone 63, we have the time.

Here's where we are now. You might notice there was a big glitch in our speed: we dropped to 8.5 million points per hour, then jumped up to 17 million points per hour. What happened in the game to cause that to happen?

Nothing. Those two data points are relatively close together. I happened to collect banquet data at about 9:10pm Pacific, 9:35pm, and 10:15pm. That's too close together, because the game doesn't update the leaderboard in real time. It tends to update on timescales of about 10 to 15 minutes. So I normally try to collect data about an hour apart or later.

Consider what would happen if the game happened to update every 15 minutes including on the hour, and I happened to collect data at 10:14pm and 10:16 pm. At 10:14pm there would be a certain amount of points. At 10:15pm the game would update and the score would increase. At 10:16pm I would check and see that increase that was posted at 10:15pm

To me, those points came in between 10:!4pm and 10:16p, about two minutes apart. If I'm calculating scoring rate by just looking at the amount of points scored in each interval divided by the duration of the interval, this instantaneous rate will be sky high. But then if I looked at 10:20pm the score would be the same, and I would see zero points scored per unit time. In other words, if the game is updating at a certain frequency and I sample at a frequency that is similar but not exactly the same, I will see wild swings in the rate of scoring. These are basically harmonic interference patterns in my data. That's what that wild spike up and down are. I try to avoid that in general, but this time I forgot when I last looked and ended up with too many samples in a short interval.

This sort of thing comes up all the time in data analysis. Any time you have data sampled at a certain rate (like the game's score board) and then you sample that data at a different rate, you have to account for these kinds of effects causing artifacts in your analysis that are don't actually exist in reality. I've seen professionals chase their tails for a very long time looking for root causes of problems that turned out to be harmonic hallucinations.

Processed data can lie, but as it turns out raw unprocessed data can also lie.

Okay, fine. You all can stop snoring now. At the moment we're about 45 million points away from milestone 63. At our current actual average scoring rate that's about 4 hours out. If and when we reach it, Kabam will unveil another milestone. I'm sure they will be thinking about what will anger the player base the most, and sticking that in there. It might be a PHC, it might be a first appearance 5* Groot, it might be 50k solo shards. The possibilities are endless.

It is possible we will reach milestone 63, unveil milestone 64, and then reach milestone 64 all while I am asleep. It depends on how much we slow down in the overnight hours. And I don't know what the event extension will mean for scoring, One more day is one more day.

Yeah but those missed payments will cost you an arm and a leg.

Just a nuke?

He shuts down Evade, Unstoppable and Miss. That's a pretty rare combination of utility!

What's Sparky's?

Evade special attacks, taunt and occasional power drain.

The reason people say he can 'handle a lot' is because he can ramp up without hitting the opponent, and wins fights before anything bad happens. He's a classic example of damage = utility.

Personally I prefer Stealthy, but I can certainly see why people rate Sparky higher.

Nah. As much as I've enjoyed this Banquet, I also find it a bit stressful to manically gather units wherever you can to open more crystals to get more rewards. Especially since it is the holidays and you might actually have other plans than just arena grinding. I'm very content with where I'm at right now and I wouldn't want it to keep going past tomorrow.

For sure there will be a slow down until the 23rd when the Unit event pops but yeah, as a global community, we're doing very well. I have to say though, Kabam hit this years banquet right on the nose, after years drop rate issues and communication gaps, I think they found the formula.

I stopped spending last year and found myself spending on this event. Good Stuff

Probably has more to do with the fact that ranked rewards has r5 gems. Whales wouldn’t want to r5, using the r5a, a Cosmic champion and then get a r5 Cosmic Gem. In that case, they’d rather wait what the r5 gems are and then r5 using the generic materials

Rng and probability.

All the gems i have opened.

5 mystic, 6 tech, 1 science 1 mutant. Ggs. I need 5 skill ags at this point.

Can't wait for the whining when/if we get more milestones.

By the way, we're about to pass the 26th milestone since people started declaring the event "over" due to the milestone rewards.