How to sim ..for a healer

Classes Priest
23

Due to the nature of healing in general, it is effectively impossible to sim healers. Unlike DPS, whose goal is to simply deal as much damage to the boss as possible, healers instead aim to keep players alive. This means sometimes playing in sub-optimal ways in order to save a life. Equally, healing is a zero-sum game, in that if one person does exceptional HPS, the HPS of their co-healers will suffer as a result. These compounding factors, along with the randomness injected into the system by players interacting with their characters, makes simulating healers next to impossible.

(Source: https://www.icy-veins.com/wow/holy-priest-pve-healing-simulations)

I guess the guide writer at Icy Veins is wrong, and so is every other guide writer who never even mention healer simming.

3 Likes
24

The arthor of the source that states napkin maths. Lists himself as the arthor of the icy veins guide if I remember correctly. He also invites everyone to the priest discord. Atm I dont have anything against him, he looks to be trying to prove things the right way and accepted the limits of his own position. Well the last time I checked anyway. He called it napkin maths so idoits cant use his work to attack ppl. A guide is a guide, its not a proof.

So why is this very same source, also programing a sim application for healing… So he can prove his arguments. Please prove that its effectively impossible to sim healers.

Stop attacking people beause you cant prove your arguments. Ask Mr Robot is not wrong because your arguments needs it to be.

25

Please lets not repeat this back and forth again, its in way to many threads already and serves no purpose.
Just state your position and leave it at that

26

Heal sims are kinda flawed at the moment, you can do it but there isn’t really any solid ones.

DPS sims are more popular than healing sims because DPS and healing are very different. DPS sims are for doing as much DPS as possible, since more DPS is good in almost every scenario. Healing has a different job though, which is to prevent people from dying which does not coincide with playing optimally or most efficiently because sometimes you will have to go against both of those concepts to save a player in danger.

Also some specs are extremely flawed on such tools from a playstyle angle, like Disc priest. If I run a askmrrobot sim of a raid encounter on my character I do way less HPS on the sim than ingame, because it simulates a completely incorrect playstyle that doesn’t work with the spec’s strengths. How can I use it to evaluate stats when it won’t even play the class properly?

1 Like
27

I have simmed a log using ask mr robot mathematics and I got an accurate result. This is with my own sim. The boss sims fine. The default sim is not a sim of one boss. Its multiple different healing scenarios. Thus you get a gear build that is best overall. If you want to sim bosses, then use the tools the website provides. You can create your own playstyle with the rotation Script and sim an ingame like boss with another boss Script. Both you need to create.

Simc does not sim bosses as well. It is wrong for many bosses too but everyone gets why. Even if you add, adds to the dummy by adding more dummies. You still dont get a simulated completely correct playstyle or encounter. Even if you call it patchwerk. You get nowhere near. Yet no one runs down simc at all. It serves its purpose. It helps you gear.

Given that the arguments against me are based on a sim, far less advanced (we are talking in development, its a spreadsheet) than Ask Mr Robot. One could state there is a clear double standard going on here. The only way to beat this in development sim is with prue perfection.

Or our source is right but you cant sim healing. Yet the source uses a sim and is developing one.

This type of arguments only happen when there is no substance to a position. They are smoke screens to hide behind when you dont have a point or real position. If one is right its easy to provide the evidence, if you worked it out and proved it. You dont have to attack other sims or sources. You would have all the proof you need at hand.

You dont need napkin maths or other non sense.

28

The biggest problem I have with askmrrobot’s disc sim is the raid healing rotation, it is absolute garbage. You will not get an accurate sim unless you program your own rotation in, which makes this tool pretty poor until you get someone who knows how to play the class to program an actual raid healing rotation in.

I ran another sim just now:
https://www.askmrrobot.com/wow/simulator/report/39d8d7d76d41466d808e4bcc01b3263c

Disc playstyle in raids is identical in almost every healing scenario, there is only 1 thing I need to look at to see if the spec is being played even remotely properly which is the spirit shell ramp. I selected the raid healing script (apparently based on Vectis which should be fine), with raid healing rotation. The HPS was again lower than it should be and when I looked at the rotation settings for the script and the log…

  • Fight was over 5 minutes long but SS was only cast 3 times(??)
  • Only checks for 5 atonements before using PWR in a 20 man raid
  • mindbender is cast BEFORE double radiance, reducing its uptime during spirit shell
  • mind blast was not casted once, not just during spirit shell, just never casted it’s not in the rotation programming
  • trinket also not paired with spirit shell despite perfect synergy and matched cooldown

No wonder I only do 6k HPS according to the sim, I’m not using my cooldowns correctly, one of the most important spells isn’t being casted at all. To fix this I’d need to program my own rotation? For the vast majority of players playing disc, the vague stat priority based off napkin maths you find on icy-veins is more reliable than askmrrobot is right now.

1 Like
29

You write your own, you get a default template as a base. The whole rotation and boss Scripts can be written by you. Dont tell me you use the defaults and call it a day. Thats not the point of the sim. It also does not prove the sim is flawed. It just shows the default Scripts suck. Every class/spec sims fine for me for logs. You get close enough to call it accurate. There is little wrong with the mathematics the sim uses as well. You can get a lot wrong and still get within 3% of the log value. That is accurate btw. I get a lot closer with napkin maths.

The difference here is I accept the limits of my information. Every sim is imperfect in some way. Its not me that has to prove anything. If Ask Mr Robot can get accurate results with its mathematics which it can (its the ingame tool tips for gods sake). Then thats enough for doubt and thats all I need. The counter evidence against me is a [working hypothesis] which is not yet able to be used as evidence or even released for scrutiny. This means I have the completely stronger position. Attacking ask mr robot is pointless. You need to still prove your own position first. Given this is FC is the only spec for holy priests, that would be a lot of evidence.

No one can say I am right or wrong. Thats my whole argument. Also Ask Mr Robot updated the holy priest Scripts. This is all about holy, so disc is off topic. Its a red herring argument.

Both the maths for holy and disc work in my sim. I get the warcraft event log results, close enough to be accurate.

Even if I ignore most of the buffs, I can still get within 3% of the warcraft event log result. You dont have to be super accurate.

30

Sorry Daddy Aellas, I think I deserve a spanking now :heart_eyes:

31

You’re describing a different use case for the sim than I was talking about. Considering Ask Mr. Robot provides class guides (which I decided to read: turns out is also extremely flawed because it’s based off the default scripts), it’s clearly marketing itself as a tool where you can put your character in and it will tell you what to equip and what spells to cast. I’m concerned primarily about players wanting to learn how to play and gear their class (people like OP). They’re going to run the default scripts, the addons that pull data from AMR will most likely also use the default scripts too. As a result they will get the wrong stat weights based off an incorrect playstyle, while also producing much less throughput than they are capable of.

All you did here was assume OP was playing Holy based off the armory, since they did not specify this is not relevant. Let’s assume you’re correct and OP mains holy; disc priests might still read this, there is no point dismissing that, there are lurkers on forums like this.

32

Again, Ask mr robot is not flawed because of what scripts you use. The underlying mathematics is sound. You are ment to create a script for your own builds. Thats why the tool exists. You are given the tools to write what you want. You can write your own boss and rotation scripts. The scripts themselves cant make the sim flawed because only the underlying mathematics can do that. If the sim was flawed it would not matter what script you run. The result would be garbage. Yet the evidence for FC is napkin mathematics, a working hypothesis. With such a weak position its no wonder that you have to attack every other sim. You dont want anyone talking about the joke that is your own evidence. With such weak evidence all I have to do is create weak doubt.

Given that the holy priest scripts have been updated and there is no issue with the default ones. I am golden. You can cherry pick sim issues all you like, it wont change the weak source I have been given.

The main true issue with the legos as pointed out by Swol.

“I’ve been having trouble with the healing legendary items in the optimizer because they are mostly relatively weak. We’re running into lots of margin of error issues because they don’t affect the total healing output a lot, especially for Discipline.”

This is why a very accurate sim is a likely requirement and you cant base anything on hasty generalisations. You have to use the most robust sim available.

This excludes napkin maths. Mine or anyone else.

O look all my posts are being reported, what a surprise.

33

Ignoring the conversation above, if you want help with gear choices use
https://questionablyepic.com/live/.
The creator is active in all (healing) class communities and by now it should work for all classes. You can enter your gear (exported via simc) and add statweights / logs if you want very narrow advice (e.g. for a specific fight). In general the results are pretty reliable and bugs get fixed very quickly.

2 Likes
34

Its alpha level, mostly useless because there is little implemented (thats why its alpha). Also because its simc, “HPS: Coming Soon”. See Legendaries at a Glance.

35

What do you mean it is simc? I know for a fact it doesnt use anything from simc except the export feature from in game addon to get all your items/chosen traits to be able to make proper calculations. How do you know it is useless? I dont defend the site itself, but I dont get how you can dismiss it with any argument that isnt applicable to AskMrRobot aswell?

1 Like
36

He’s adamant that HPS is important for some reason, and HPS simming is desirable and useful. :woman_shrugging:

1 Like
38

It’s not. Your HPS in pretty much every case, doesn’t matter. It certainly does not matter when deciding which item to upgrade.

And the answer to that has been given in the first two posts: you don’t.

Nope. Because you can’t reasonably sim neither a raid nor a key. When your precious sim ends up with 10k hps, while in a +17, most people average around 3-5k HPS, and don’t even peak anywhere near 10k hps, not even on Pride, that pretty much confirms that the sim is irrelevant.

When the sim ends up simming spells that you never use in your spec, leading to a stat weight distribution that is actively detrimental to your performance, simming is not only useless, but harmful.

Because DPS is important. Even for a healer. Go sim for DPS, that is useful. Simming for HPS is not. There’s not a single encounter in the game where you want to gear for HPS. Not one.

39

My sim works with logs. That means it 100% sims raids or mythic+. Doing so is the most easy part of siming. Simming a log is easy. Doing what Ask Mr Robot has done is harder by may factors. Its your complete lack of understanding that makes you think your arguments are worth the time of day.

Siming a log is simming a boss. With the db files I use, I can also sim the boss. Create the phases for the boss fight and even copy the damage output of the boss for each phase. Model the danage each player in the raid recieves. I can spawn the right creatures as adds. I can use the movement and rotation of real players in the raid. Model the overhealing for each healing spell I cast. Then I can change the gear and rotation of one healer. Talents and convant abilities.

Change my lego from DI to FC. Then get more or less the same result for the output. This should be expected, well if you think about it. You don’t get a valid comparison. In Raids the healing available is that left over by the other healers. So you have the same maximum amount of hps you can do. If both gear builds have enough output then about the same hps will be recorded as the output. This if you heal only when required. Attempting to heal more becomes overhealing.

This is the same with mythic+, there is only so much to heal. If both sets of gear and legos have enough output. Then both will perform the same and trying to heal more is just overheal. Thus, screaming its not a real raid or mythic+. Just illustrates the total lack of understanding being shown in this thread.

With the sim you have to be smarter. Make sure there can be a comparison. Siming raids bosses and mythic+ logs is not illuminating. Its also not a requirement for a sim to do so.

Anyway the only way to have a sim like a log is to completely sim all the healers. You cant have just a holy priest spreadsheet. You have to sim the HP’s of each player and have each healer complete to heal when damage happens. Thus, you can get the overheal and how well you complete with other healers.

The main issues you can run into, is that you get high output in your spreadsheet but cant complete well with other healers. So the logs show you below other healers. Even if you have high output, that build is not good for a competitive healing enviroment. You cant determine that is the case without a sim, that has a raid group of other healers or having thousands of logs results. As the healing output is competitive, other healers can get there first.

You are not showing a maximum healing output for a build that would be shown by that of a spreadsheet. Its the result of different players and the different decisions they make. Thus you need a log to sim a log perfectly. Thus you sim something you already have the answer for. Thats why simming logs is not too illuminating.

If you take a log and create a sim from it, the results will vary but you will get a more realistic result for the boss simmed. The rotation scripts for each healer will keep things consistent.

You are still competing against other healers, there is still a maximum amount of damage that can be healed. The output will be realisitc and accurate. You can run the sim thousands of times and get the most common result. Present that as an answer to the end user. The result is only valid for the scenario you created, that of the log or boss encounter.

Ask Mr Robot is the more correct way of doing things IMO. The output is against a range of possible scenarios, thus you can judge the result as a more general representation of performance.

HPS is the healers main output, arguments that it should be ignored are self serving non sense. Its like understanding we need to breath o2 but then arguing that we need to ignore that and focus on other things when we create a life support system. Ultimately leave o2 out of the life support system equation.

Hello,

This is an automated message from World of Warcraft Forums to let you know that your post was removed.

https://eu.forums.blizzard.com/en/wow/t/how-to-sim-for-a-healer/275039/37

Your post was flagged as inappropriate: the community thinks it is offensive, abusive, or a violation of our code of conduct.

This post was flagged by the community and a staff member opted to remove it.

Please review our community guidelines for details.

Constant now.

40

This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.