The Team Energy Restore Skill is underrated

[graceful narwhal]

The TLDR is that a team wide energy restore mon is effectively a team wide help % bonus. Each point restored per day on average is worth, conservatively about 0.6% team wide help speed increase.

Warning. this contains extremely fast and loose math. the math around this needs to be fast and loose for reasons you'll see.

To start, I'll just say, I'm a hypocrite. I do not currently use energy restore mons. After doing the math for myself I am probably going to start doing this though.

To understand why we need to look at help rates.

the base expected rate assuming you sleep once per day, and sleep for 8.5 hours looks like the above chart.
it starts at 2.5x and slowly linearly goes down to 1x.
the trapezoid has an area of 28.875 help hours
then you have a rectangle with area 7.5 help hours
this leads to 36.375 help hours per day
which is a base help rate of 1.51. That 1.51 is. how much you are helping on average as compared with the rates shown for the mon.

A skill activation that adds energy to your mons adds a sliver that looks like this.
The value of this sliver is based on when it occurs. if it happens early on, you get a full 16.5 hours of the extra energy boost. if it happens when ur already drained you only get a number of minutes equal to the amount of energy regained.

This makes it hard to come up with a true value for energy restoring mons, because it matters a LOT how frequently they are doing it. If they use it a lot, especially during the morning, you're way more likely to get value out of them.

I'd go so far as to say an energy restoring mon scales quadratically with how frequently it activates. A 10% increase in activation rate due to a skill++ nature is probably worth 20% or more, the higher the energy your mons when it activates, the more the mons benefit. The more energy you have for future activations (and the more likely it is to activate again) Everything about this compounds

At nearly full energy, a skill trigger that fills you up to full is worth an extra 1.5/1.51% help speed per percent restored. or basically 1% extra help speed. this means ideally your 5% energy regain skill trigger can be worth up to 5% help speed boost. and thats per trigger

At empty energy, it only boosts you by 1.5% for 10 minutes per point. so for 5 point restored its worth 50 minutes * 7.5%. This means the SAME skill trigger that during the day is worth an extra 5% energy is only worth an extra 0.26% help speed AT NIGHT. Note that 10 points restored is worth 100 minutes * 15%. So doubling the skill effect quadruples how much help bonus you get at night. 10% energy is worth ~ 1% extra help speed AT NIGHT. But night gets vastly overshadowed by day so we can mostly ignore this effect.

The value of the restore energy skill trigger is based on the amount of energy you currently have. and it linearly goes down throughout the day (when ur asleep ur presumably not activating the skill trigger so we can ignore this).

I would say as a super rough estimate on average for non nappers (nappers its even better), restore energy is worth an extra 0.6% help speed PER energy point PER time it happens. This is a conservative estimate and doesn't account for the fact that help events are more likely to happen in the morning, it doesn't account for the compounding nature of help events. It IGNORES overcharge. It IGNORES napping. It only accounts for the fact that you're going to sleep for 8.5 hours so you're going to sleep when your energy is at, at lowest, 6%, and at its highest, 100% in the morning. In practice the value is likely closer to 0.8% but I am being conservative because even conservatively this skill is AMAZING and underrated. 20 team energy restored over a day is equivalent to a TEAM WIDE 12% help speed boost.

That's all. questions and feedback are appreciated.

The Team Energy Restore Skill is underrated

[rare yarrow]

It's not just team energy, all energy skills have been massively underrated and will continue to be because they don't provide instant gratification. Thanks for doing the math!

[graceful narwhal]

This is true. However the team energy one restores the most energy overall so it's in my opinion the best one. The one that restores randomly to one mon is also pretty good since I believe it's like 18 energy or something at level 1

The self energy restore is less good because it's only like 8 energy.

[carmine scroll]

are you in the mathcord? would you care to check my work in the forums section there? i happened to make calculations somewhat similar to yours earlier today

[graceful narwhal]

dont ask me to check your work my math is terrible and a lot of this is very rough approximations

[carmine scroll]

haha alright well i think what ive found has similar implications to your approximations

[graceful narwhal]

well I can tell your math isn't exactly right because this combination should, I believe lead to a non 0 value for k. or at least for what you're trying to calculate.

the area of the shape in red that I drew and the effectiveness you're trying to determine has a width of c+b and a height of c

the effect/power of the skill should thus be exactly proportional to (c+b) * c
and overcharging is possible up to 150% so b shouldn't be capped at 1-c

still I think ur math is very close to right

and I like the way you framed it because thinking of it in terms of c and b allowed me to realize this extremely simple formula (c+b)*c

and the numbers happen to work out such that when c+b = 1, you get about 1%*c help speed bonus over the course of a day

because of this coincidence you could more or less say that your daily help speed bonus gained from any given skill activation is equal to (c+b)*c%

[carmine scroll]

oh yeah the lower bound for b should be c and the upper bound is 1-c because my graph is super scuffed and doesnt account for ranges that are too low or too high LOL then the numbers either get wacky or undefined

but yes u make a good point with these observations, perhaps things can be much simpler in calculations

[frozen atlas]

do we have concrete data that shows how much more efficient a Pokemon is in relation to their current energy level?

[gloomy mango]

I did similar calculation and found that 4 berries 1 support is slightly better than 5 berries without self energy restore. Definitely 5 berries with self energy restore would be the best but the most strongest berry mons do not have the skill.

[jaunty isle]

what’s the % when the 1x starts

[jaunty isle]

so dotrio gang in cyan beach 🗿🗿🗿🗿🗿

[graceful narwhal]

I believe that still caps out at 100 unless they changed it in the latest patch

That one was mostly good for just getting your energy back with less sleep needed which tbf is still great if ur playing normally and amazing to get extra back from naps

[hazy mauve]

if you are interested, here is a calculator i build for energy4everyone calcs:
only edit the yellow cells.
can make yourself a copy to have it private.
The linked file is set to "everyone can edit"
https://docs.google.com/spreadsheets/d/1fd3UUaEwg1Ys5DFETJsnlfRuGJosRrt9RoGDuIxYtz0/edit?usp=sharing

looks like this

[graceful narwhal]

id recommend making it locked so people have to make copies however this is awesome

[graceful narwhal]

If you want to make this super accurate, add 1600 to all the productivity values as that's the base productivity at 0 energy relative to the formulas you're using (2/3rds of the boost gained from 100%

and you can super simplify the energy regen formula because energy level is frozen while asleep so it only needs this syntax

min(150;IF(P9>$B6;P14;P14-6)

also I'd add =max(0; ...) at the start to all your formulas

still i love this. id pin it if I could

[hazy mauve]

Ill check later thanks

[hazy mauve]

Added this and a lot more. Have a look at v3

Recovery bonus subskill support now

Fixed recovery to be the steap wall its every morning. Added the first hour of next day.

No sneaky snacking supported yet

[hazy mauve]

V4 some more improvements

Max 100 energy from long sleeps

[golden badger]

you think e4e twice a day is average ?

[hazy mauve]

Skill pokemon do procc it more than once normally

[golden badger]

OK that make sense

About like eevee

[hazy mauve]

Yes

[golden badger]

I wonder if the correct thing for estimation purpose is just take that effect and split it over 16 hours.
Maybe with a triangle distribution because it's more likely to trigger early day.

[hazy mauve]

In the calculator you can adjust everything

[graceful narwhal]

This chart rocks
Thanks

[graceful narwhal]

Is sneaky snacking confirmed to be at 250%?

I'm curious about if there's anything about it in the math server

[hazy mauve]

Ill have to reverify it…

[graceful narwhal]

Well is it at least approximated at 250%? I'm sure exact figures are probably impossible to determine

But I wonder if anyone's done any research on it

[hazy mauve]

Yes in #1146751257244090428 message

[graceful narwhal]

If I'm understanding your chart it's around 225% of what you'd expect from normal helps if you had 0% energy and all normal helps resulted in berries? And this is regardless of energy level? Is that all correct??

[hazy mauve]

Later i checked with my other tests and it matched 250%. But thats why i have to reverify too

[graceful narwhal]

Is the rest of what I said correct?

[hazy mauve]

I assume yes

[graceful narwhal]

Oh god that's insanely good. I guess it's a strong strat to forget about energy altogether and just get 5 berry mons and never check them

[hazy mauve]

Some people do that lready xD

[graceful narwhal]

How's that working for you this week? Where are you at total score wise? Just for a point of comparison

[hazy mauve]

I dont anymore. Need ingredients too xD. And my kids like to collect berries and watch them being eaten

[graceful narwhal]

Ah fair enough

I still have that kid desire to tap too.

[hazy mauve]

Only here at cyan. Little snacking when i come back to app late, no planned snacking strat

[hazy mauve]

@mossy viper if you are interested to share a energy calculator, have a look into the google sheet in mentioned message. Its a link with edit perms now. But if people make a copy its probably better. Any feedback also welcome.

[hazy mauve]

Got the perfect one

[serene orbit]

Are people still doing this? Seems kinda busted to not check and run berry instead of energy if that is the case

[hazy mauve]

Doing what? Testing?

[serene orbit]

Yeah I was curious if it really gave the strength equivalent of rolling berry every time. I found this Reddit post which seems to suggest that opening the game more often increases the amount of strength. https://reddit.com/r/PokemonSleep/s/hEWjb87k9P

[serene orbit]

Just seems like it would be crazy strong lol

[hazy mauve]

yes ADHS strat is strong, if you have the time and patience. combine it with high energy, from an E4E and pillows and it rolls flows like butter.

[cobalt pine]

This is from 2 months ago, didn't they patch this?

[cobalt pine]

Or do they mean checking the app without tapping the pokemon? I'm a little confused

I'm actually very confused

About what exactly is the sequence of actions being taken

Tapping the pokemon but not collecting the drops?

[serene orbit]

Lol I think he’s opening the phone but not tapping the pokes

[hazy mauve]

not collecting. but checking every 10 or so minutes, and you force the snacking screen to show up, when you go into a menu and out to main screen again. then it shows if a help was colected.

this screen will then show you a value, that is x2 helps. but the time that passed, was only the time to generate one help. you always get double this way.
If you get the screen after 2 helps time, you will see a 3x helps snacking screen, and so on...

[hazy mauve]

I did test with my slowbro at above 100 energy. And i also did not get 250% bonus. I got 221%. Timed when the next help should come along and recorded my screen sround that time 3 times. Works like a charm with my slowking. Can calc it to precision of 1-2 seconds with 2.21 multiplier.

[graceful narwhal]

Interesting. I was going off of the mathcord numbers but either those are outdated or they were wrong to begin with

[livid ember]

@twilit coral

[twilit coral]

yeah i remember reading that they patched this pretty quickly, like one week after

[cobalt pine]

I think that may be why I can only get it to work on my older devices

[twilit coral]

woah, how? by running an older version?

[cobalt pine]

Maybe they are slower? Or differently optimized?

[twilit coral]

the app usually forces us to update though right

[cobalt pine]

Should I grav the specs and submit them somewhere?

It could be something hardware based

[twilit coral]

btw has anyone confirmed that over 100% energy gives the same helps/h as 100% energy?

[hazy mauve]

Im on it

[hazy mauve]

seems to be not the case. at 105 and 135 the same time recorded

[twilit coral]

ok yeah

i think i also have data proving that >100% gives basically the same rate as 100% energy

i plotted the helps per h against the energy of my sylveon at the time of sleep

all these are datapoints from sleep sessions

if we include the sessions where the pokemons weren't sleeping, i did some math to figure out the equivalent mean energy (which usually is just the midpoint of the start and end) and here's the results

either way, the R^2 is basically 0 and there's zero gradient

most of the outliers here are due to recorded sessions lasting less than 3h