Question about sub-event: Prodigal Sun - China

Spoit

Phantron wrote:

Spoit wrote:

Your tolerance for scaling is a lot higher than most people's.

Personally, BP is at 'normal' levels and already caused a couple wipes, even almost wiping me during a try with cmags/lazythor

BP is more of a quantity over quality. Some of these games he will pick up 12 black right away and wipe the floor with you, but most games he won't because the AI doesn't care about prioritizing colors. I've beaten with all kinds of random guys I sent out to die on purpose (ironically, I had to lose those fights on purpose even when I could've easily won).

I found Hawkeye to be much harder despite being lower level because you have to get through all the goons first before you can even hurt him. BP has a lot of HP, but it's still less than Hawkeye + 2 goons put together. Hulk can usually get you an easy win at the cost of his life against those ultra high level guys too.

Hawkeye is just another countdown tile though, so if you're bringing a cmags or OBW or whatever to already control them, it doesn't matter so much. Also, he's paired with much weaker troops, that you might consider eliminating before hawkeye anyway, or can let half of their countdowns go off (the yellows, or the relatively weak pistol shots) without it being a big deal. Whereas there are multiple nodes with BP where you actually have to worry about eliminating the countdowns, like the rocket trooper or sniper, which puts you at more of a disadvantage denying black, compared to you almost assuredly having a use for the red and blue you deny hawkeye

Dragon_Nexus

Phantron wrote:

Dragon_Nexus wrote:

Phantron wrote:

Dragon_Nexus wrote:

So I haven't played this game since about this time yesrurday...unless PvP counts, in which case it was 3am. It's been more than 12 hours, and the nodes will give me about 100-200 points. Meanwhile the guy in first place has 4000.

So err...did they turn rubberbanding off completely? Because I'm not going to sit here and grind out 60 nodes per sub event to get up to that guy.

There is no rubberbanding for LA/Wakanda.

Oh. So clearly I picked a bad day to go to work, huh? Great idea putting the round without any rubber banding during a business day so I'd end up losing out massively.
Dude in first place had 4600 when I came on, and I had 800. And the best node was for 270 and the enemies were level 160. What, I'm meant to grind all these nodes to gain 3800 points? Not going to happen. This has really put me off playing this event now...

What do you think happens to the scaling of the guy with 4600 points?

Scaling is a resource and if the time is right, you should use it. I figured this out pretty early and then worked out the math and decided it'd be crazy talk to try to keep up with the guys with 4600, so I simply didn't do the nodes that I could have. Hopefully the guys who ruin their scaling will fall behind on the next round, but if not, kudos for them for making a bold gamble. History suggests that such gamble won't pay off anyway.

Okay, so what is the actual strategy then? Do a sweep of the nodes then wait 12 hours, or what?
The goal posts move with every event.

Unknown

anyone noticed how China had no asian bad guys, but suddenly they're everywhere in washington and L.A.?

Unknown

Dragon_Nexus wrote:

Okay, so what is the actual strategy then? Do a sweep of the nodes then wait 12 hours, or what?
The goal posts move with every event.

Doing more missions is always better but it comes at the cost of time + scaling. The trick is to know when you got to sacrifice your scaling for a lead, or to catch up. The existence of rubberband or not changes how much of the resource 'scaling' you must commit to get a lead. When rubberband is high, you don't have to sacrifice much scaling to catch up (but then, neither is your opponent for taking that lead in the first place). When rubberband is low, you have to commit way more scaling to do the same (but again, so does your opponent).

PvE events function a lot like cycling. The guy trying to break away from the peloton (the masses) expands a lot of energy when trying to break away, and if you screwed up all you do is wasting your energy. But if you have big enough of a lead then the masses give up and then you can cruise rest of the way. At the top positions you'd see the leader just break away with a 3K lead and that's basically a challenge of 'better burn your scaling too if you want to catch me!' And since most people don't want to risk that, the leader can then cruise to an easy finish, even though his scaling is likely in a worse place than his challengers, but it doesn't matter when all your challengers gave up because they see the leader is 3K ahead. Now what if the challenger called the leader's bluff? Well in that case you're fighting some pretty crazy guys and there's no clean way to beat fellow crazies.

Dragon_Nexus

Phantron wrote:

When rubberband is low, you have to commit way more scaling to do the same (but again, so does your opponent).

Okay, what does this actually mean. You talk like you're meant to manipulate the scaling effect but that's where you lose me.

Spoit

Dragon_Nexus wrote:

Phantron wrote:

When rubberband is low, you have to commit way more scaling to do the same (but again, so does your opponent).

Okay, what does this actually mean. You talk like you're meant to manipulate the scaling effect but that's where you lose me.

The more nodes you finish, the more your personal scaling goes up. However unless you grind a ridiculous amount, community scaling will be way more significant than personal scaling, which effects everyone

Unknown

Dragon_Nexus wrote:

Phantron wrote:

When rubberband is low, you have to commit way more scaling to do the same (but again, so does your opponent).

Okay, what does this actually mean. You talk like you're meant to manipulate the scaling effect but that's where you lose me.

When the event ends, the winner is the person with the highest point, not the person with the highest points + lowest scaling combined.

Let's use this event for example and assume the no sudden influx of point enter the system. Right now it looks like 2000 points is a very solid lead. There are always enough easy missions so far that prevent you from unable to do any missions even if they all hit 395 (and 395 goons really aren't that hard). It seems downright unlikely to make up 2000 points in any given round. So, if you lead by 2000 points going into the last round, it doesn't matter how bad your scaling is.

Also, having an early lead discourages competition. Your competitors do not know what your scaling is. Even if you've very nasty scaling, your competitor only sees your score, and based on the prevalent attitude on the board it seems most just assumes a guy with a significant lead is just super awesome/lucky/whatever. And if your competitor gives up, that makes your terrible scaling suddenly not so bad.

Now, you can't possibly predict how everyone will react, though people tend to play in a way that capitalizes on this anyway. If you're in any borderline placement area (say, 20th versus 21st that determines 2 3* or 1 3*), whoever is #20 is obviously going to think "I better build my lead more" if he sees #21 kind of close, and then whoever is the new #20 will think the same thing, and if they all did that, by the time #21 comes back he sees whoever is #20 having a very large lead and think 'oh no I can't possibly catch up', without realizing the 5 or guy ahead of him almost certainly all grinded extra to ensure they didn't fall to 21 so they're likely to be hosed on scaling.

Spoit

I'm not sure if I really buy that line of thinking. Even once you do have a good lead, don't you still have to do regular clears of the now level 300+ nodes in order to keep it, otherwise you lose it?

Then again, I don't even bother looking at the top 2 scores or whatever, just the #20 cutoff

Dragon_Nexus

Phantron wrote:

Dragon_Nexus wrote:

Phantron wrote:

When rubberband is low, you have to commit way more scaling to do the same (but again, so does your opponent).

Okay, what does this actually mean. You talk like you're meant to manipulate the scaling effect but that's where you lose me.

When the event ends, the winner is the person with the highest point, not the person with the highest points + lowest scaling combined.

Let's use this event for example and assume the no sudden influx of point enter the system. Right now it looks like 2000 points is a very solid lead. There are always enough easy missions so far that prevent you from unable to do any missions even if they all hit 395 (and 395 goons really aren't that hard). It seems downright unlikely to make up 2000 points in any given round. So, if you lead by 2000 points going into the last round, it doesn't matter how bad your scaling is.

Also, having an early lead discourages competition. Your competitors do not know what your scaling is. Even if you've very nasty scaling, your competitor only sees your score, and based on the prevalent attitude on the board it seems most just assumes a guy with a significant lead is just super awesome/lucky/whatever. And if your competitor gives up, that makes your terrible scaling suddenly not so bad.

Now, you can't possibly predict how everyone will react, though people tend to play in a way that capitalizes on this anyway. If you're in any borderline placement area (say, 20th versus 21st that determines 2 3* or 1 3*), whoever is #20 is obviously going to think "I better build my lead more" if he sees #21 kind of close, and then whoever is the new #20 will think the same thing, and if they all did that, by the time #21 comes back he sees whoever is #20 having a very large lead and think 'oh no I can't possibly catch up', without realizing the 5 or guy ahead of him almost certainly all grinded extra to ensure they didn't fall to 21 so they're likely to be hosed on scaling.

I understand the psychology to it, but I don't know how to translate that into progress.
Back in the day, you grinded (ground?) the nodes until they dropped to 1 point. Then you ground (grinded?) the 1 points over and over to squeak out ahead.
Then they added rubber banding so you could grind a little bit and catch up later if you put in the work, but overall you still had to keep at it to keep the points up for the main event rewards.

Now...I don't know what to do. I don't know what's expected. The way to progress is completely obscured. Unless they actually want me to play with every refresh, something I can't do because I have a job, I'll have no chance. Even then the scaling will make it difficult to win the nodes anyway. Meanwhile you seem to suggest letting everyone else screw their scaling up while I chip away a little at a time...something that will cripple my main event score.

Unknown

Spoit wrote:

I'm not sure if I really buy that line of thinking. Even once you do have a good lead, don't you still have to do regular clears of the now level 300+ nodes in order to keep it, otherwise you lose it?

Then again, I don't even bother looking at the top 2 scores or whatever, just the #20 cutoff

People will still do their normal clear, but having a big lead discourages people from doing something extra, which is something you don't want them to do when your have a lead with unfavorable scaling.

Let's say all events have completely equal points, and we have exactly the same scaling, so I jump out to an early 3000 point lead and took a corresponding scaling hit. But, as long as you're not aware of this, to you it may look like I'm just some crazy guy with crazy points, so you'd never think about doing extra to make that up. You'd do your normal clear, sure, but I can keep up with that too even with my bad scaling. As long as you're not hitting each node 4 times, I can probably manage with say a twice per node clear even if they're level 395s. I sure won't be able to manage that 4 times per node, which is why I want to convince you to not chase. But this requires me to start off with a big lead. If I'm only 300 points ahead of you, you're far more likely to chase than 3000 points, even if the scaling difference is unchanged (that is, in this case I'd have scaling comparable to yours and should be able to fend off your chase). It's better to have bad scaling but not being chased, than to have good scaling that is constantly being chased. And if someone will chase you while 3000 behind, there was probably nothing you could've done to convince that guy to give up anyway.

Of course since the difficulty/value of point changes, you can't say for sure whether this strategy will pay off every single time, but I'm seeing fairly wide separation in the reward tiers. I'm ahead by 3000 points over #3 in my main bracket, and I don't see why the #3 would try to chase because it's likely to be futile, and it's even better if he doesn't chase at all, since then I don't have to do anything beyond a normal clear.

Spoit

And that's the crux of my argument. Just because you are willing, much less able, to do 2 clears at 395, doesn't mean that the argument holds for people who are less crazy than you.

And more importantly, doesn't really apply (and thus is not very helpful advice) to people who aren't going for top 2

Unknown

Dragon_Nexus wrote:

I understand the psychology to it, but I don't know how to translate that into progress.
Back in the day, you grinded (ground?) the nodes until they dropped to 1 point. Then you ground (grinded?) the 1 points over and over to squeak out ahead.
Then they added rubber banding so you could grind a little bit and catch up later if you put in the work, but overall you still had to keep at it to keep the points up for the main event rewards.

Now...I don't know what to do. I don't know what's expected. The way to progress is completely obscured. Unless they actually want me to play with every refresh, something I can't do because I have a job, I'll have no chance. Even then the scaling will make it difficult to win the nodes anyway. Meanwhile you seem to suggest letting everyone else screw their scaling up while I chip away a little at a time...something that will cripple my main event score.

I didn't grind nodes to 1 either, but I certainly grinded a lot more than other guys in my main bracket did except for the guy who's trading #1 spot with me.

You only have to do better than the guys who are likely to compete. This also applies to all rewards tiers. A 3000 point lead might be enough to discourage #3 from trying for #2, and maybe 1500 is enough to discourage #21 from trying for #20. You also have to have a feel for when people are slacking. It's pretty clear that people mostly wimped out of Wakanda, so if you grinded harder than normal on Wakanda you'll walk out with a very big lead. Now, depending on how the event unfolds it may or may not matter, but so far most events do reward building an early lead. You want to grind when you perceive there is a weakness amongst your competitors. If the missions are all easy, even grinding them down to 1 won't scare anyone. Right now Hell's Kitchen mission sure looks very scary, in fact so much that it's not clear to me if you can even grind these at all, so I'm pretty sure whoever does well in Hell's Kitchen bracket is going to walk away with a huge lead in the main bracket. Now I might guess wrong and it turns out everyone can grind Hell's Kitchen to 1, but you can usually get a pretty good feel of how the community is doing based on the complaining on the forums, not to mention just looking at the nodes themselves. The combination you face in Hell's Kitchen is often quite nasty and this is where you should make your move (if you can make such a move at all, that is).

Unknown

Spoit wrote:

And that's the crux of my argument. Just because you are willing, much less able, to do 2 clears at 395, doesn't mean that the argument holds for people who are less crazy than you.

And more importantly, doesn't really apply (and thus is not very helpful advice) to people who aren't going for top 2

I can't see the points past #10 but here's what my main bracket currently looks like:

39.8K
38.4K (me)
36.6K
36.2K
34.6K
34.5K
34.5K
34.5K
34.0K
33.8K

I think it's pretty safe to say there are roughly 3 grouping now (38K+, 36K, 34K). Assuming no drastic changes, the people in the 34K range sure aren't going to challenge the 38K guys. The guys with 38K should (I hope) be able to fend off the 36K guys and likewise for 36K versus 34K. I think the only real uncertainty in this bracket is who is going to finish #5 (it'd be whoever is on top of the 34K group).

Now as you move down the difference in point will probably be less, but then we're talking about less capable players, so while at the top 10 we're roughly separated by 2K tiers that look fairly impossible to make up, down at say #20 maybe people are separated by 1000 but that should also be just as hard to make up. No I don't think people necessarily think this much, but it's really easy to see how even a small lead in this event can be very hard to overcome as shown by all the complains about not enough points to catch up, so people definitely recognize when they'll have a hard time catching up.