Sim results for human errors?

Anything mage specific not covered by any of the other subforums, like raid instance guides for mages.
Katsumi
Posts: 202
Joined: Fri Nov 25, 2016 11:21 am

Sim results for human errors?

Unread postby Katsumi Thu Dec 29, 2016 1:28 pm

Hi everyone,

After I spent some time unsuccessfully looking for this, I'm going ahead with a new thread. If this turns out to be redundant as a result of my lack of knowledge, please do not hesitate to lock it.

Inspired by Maywtf's thread on finding "general guidelines" for gear composition I was wondering if the same approach could be taken for general player mechanics. What I found so far points to personal in-depth log analysis as a tool to get better, but the plethora of "help me with my dps" threads here and in other places suggests that this is clearly not working for the majority of players. To put things into perspective, there seems to be a gap of knowledge between the superb guides that get people started and the hardcore theorycrafting that is being done on the very other end of the spectrum.

My perception is that improving after a certain point gets very difficult (who would have thought), but not because of a player's mechanical inability or a lack of theoretical information on what one should do to achieve optimal results. It seems to be a problem to many that identifying the most impactful mistakes (which are pontentially different for everyone) fails because a) the learning player is not aware of them (otherwise he would not make them) and b) the teaching player is not able to explain what she is doing better ("unconscious competence"). To make things worse, on the teaching side very few people are willing to sacrifice their time to help someone else get better while on the learning side many people don't have to time at hand that it takes to find out everything on their own.

The main questions would be the following:

1) What are the most common mistakes that players make?

Is it messing up the rotation, time not spent casting, interrupts due to movement, pulling cds at the wrong time etc.? Do we have stats on this and if not, how do we get there?

2) How impactful are these mistakes?

Do we know how much of a dps loss each of these potential mistakes are? If not, how do we calculate it?

3) Consequently, which mistakes should I focus on avoiding first?


Let's say I am using the Wriggling Sinew trinket. It's a common mistake to simply forget to activate the trinket, which reduces its real dps value outside of sims drastically. If we knew our own error rate we'd be able to take this into account and compare it accurately to another trinket that activates automatically (a result of interest for the theorycrafter). However, we'd also know how important it would be for the player to stop doing that (a result of interest for the casual player). Another example would be to place a Rune of Power right before a boss casts a gound effect on you. Not only do I lose the casting time of Rune of Power as dps, I also don't get the buff if I have to move out of it. While that seems to be a massive dps loss, I have no idea how much it actually is.

The idea of this thread would be to end up with a tier list of mistakes, basically a data-driven document that serves as a roadmap for players who want to improve. It would provide evidence for what's important to focus on and what is not (don't even start thinking about stat weights if you still keep messing up your burst phases) and help people advance faster. Any ideas and support in this effort would be greatly appreciated.

Thank you for your time,

Katsu.
Last edited by Katsumi on Thu Dec 29, 2016 3:32 pm, edited 1 time in total.
Katsumi
Posts: 202
Joined: Fri Nov 25, 2016 11:21 am

Re: Sim results for human errors?

Unread postby Katsumi Thu Dec 29, 2016 1:29 pm

Reserved for a summary of the answers to the questions raised above.
kesMow
Posts: 34
Joined: Tue Dec 13, 2016 12:29 pm

Re: Sim results for human errors?

Unread postby kesMow Thu Dec 29, 2016 3:06 pm

Awesome idea, love this!

I haven't seen anything like this before and don't know where you can find all theese information you're looking for...
I think you should just wait for people giving tips and/or telling mistakes they've done...

For the sinew example, the first thing is to not forget using it of course :)
Another thing interesting about it, is to use it during your combustion so it is a guaranteed crit, and it will do twice damage. At the beginning, I was using it with a macro aligned with combustion but it didn't always crit cause sometimes the 10th proc was outside of the combustion and it didn't crit (even with 60% crit chance :( ).
The good approach is to delay a bit the use of the trinket 2 or 3 spell before you use combustion so it will always be a crit.

I will probably post other examples or stuff in here later on, don't have enough time atm... :)
Katsumi
Posts: 202
Joined: Fri Nov 25, 2016 11:21 am

Re: Sim results for human errors?

Unread postby Katsumi Thu Dec 29, 2016 3:52 pm

Thank you kesMow for your quick reply!

That might be a good idea to continue - list all mistakes we can think of, sort them in categories and then think about how we can estimate the impact of each of them. Maybe we could use the following form to make the contributions easier to harmonize:

Short description:
Forgetting to use a trinket when it's off cooldown.
Long description:
Not using a trinket when it could be used is a direct dps loss induced by inattention. The impact depends on the type of the trinket and increases with the usefulness of the effect of the trinket upon activation. It is most severe for trinkets with high burst that can be misplaced, such as Wriggling Sinew, but still significant for trinkets with effects that need to be syncronized with burst phases and timed correctly. Different trinkets will have different dps error margins.
Potential impact assessment:
Evaluate uploaded combat logs to assess trinket usage efficiency across player percentiles to find out how common these errors are across the player base. Compare the difference between good and bad players, when RNG, ilvl and overall skill are taken into account.
Potential remedy:
Easily recognizable weakauras (something that jumps in your face when you need to use your trinket).

and another one:

Short description:
Forgetting to use potions.
Long description:
Using potions at the right time as well as consistenly can make a difference. Since their cost is no longer a real issue, forgetting to use them in a dps loss that can no longer be justified.
Potential impact assessment:
Sim dps with and without the use of potions. For the economically constrained player, calculate costs in gold spent on potions per hour.
Potential remedy:
Easily recognizable weakauras that trigger when potions can be used. Combining the use of potions with the use of on-use trinkets in a macro.
Last edited by Katsumi on Thu Dec 29, 2016 5:31 pm, edited 2 times in total.
User avatar
Maywtf
Posts: 35
Joined: Tue Nov 22, 2016 1:10 pm

Re: Sim results for human errors?

Unread postby Maywtf Thu Dec 29, 2016 4:50 pm

Inspired by Maywtf's thread on finding "general guidelines" for gear composition I was wondering
I am absolutely flattered.

Let me add something on my own. I have worked with many mages in the past, so these are the most common things I tell them.

Short description:
Lack of encounter-oriented knowledge.
Long description:
For all of the raid encounters, you should gather information about their skills, mechanics and phases. This knowledge will prepare you better for the fight. It helps with cooldowns' timings and positioning. There is nothing worse than having to move when you just activated all of your cooldowns. Choose appropriate talents for the encounter. For example, you take Unstable Magic on Ursoc, but on Elerethe/Il'gynoth Living Bomb is usually a better choice.
Potential impact assessment:
The best impact assessment in this situation would be logging the fights on before/after you adjust yourself to the mechanics present.
Potential remedy:
Check out the videos on youtube. There are well-described guides on the Internet that will guide you through. Ask other mages.


Short description:
Lack of preparation.
Long description:
For the best results possible, you should basically maximise what you already have. Enchants/Gems/Flask/Food/Potions summed together actually make quite a big difference.
Potential impact assessment:
Flask: 1300 main stats
Food: 275 secondary stats
Enchants: 400 secondary stats from rings, 200 main stats for cloak, additional damage out of neck enchant
Gem: 150 secondary stats per gem (one 200 main stat gem)
Potential remedy:
Always try to enchant and gem your gear. Try to get food and flasks for your raids. You don't have to be a millionaire, you can always pick up Alchemy/Cooking and make your own stuff.
Katsumi
Posts: 202
Joined: Fri Nov 25, 2016 11:21 am

Re: Sim results for human errors?

Unread postby Katsumi Thu Dec 29, 2016 5:29 pm

This may be trivial and more an the human side of things, but I think it should be mentioned:

Short description:
Lack of mental agility and/or competitive attitude.
Long description:
Competitive dps is hard work, requires extreme amounts of attention and drains mental stamina. However, some days you come home from work and just want to chill, or are just tired for other reasons. Many people play after long shifts of work or when the family is finally asleep at night, others mainly play to relax and not to compete. There's absolutely nothing wrong with that, but it might explain why you're winning the dps race on some days, but not on others with similar RNG.
Potential impact assessment:
Compare results when you're mentally fresh vs. when you're not. Compare results when you're giving it all you got vs. when you're half-assing it because you'll down the boss anyway. Getting numbers on these differences from many people would lead to a general idea about the impact, but personal bandwidths could vary widely.
Potential remedy:
Be aware of your physical and mental state and choose your goals in the game accordingly. Use warcraftlogs not to feel bad about where you rank currently, but to track your personal progress, play more consistently and improve continuously.
Camjee
Posts: 13
Joined: Mon Jun 22, 2015 3:25 pm

Re: Sim results for human errors?

Unread postby Camjee Sat Dec 31, 2016 5:53 am

I sometimes start to hardcast pyro mid combustion because of the "spell que" stuff they added, which maybe loses me a pyro in combustion.

If anyone know how to turn it off please do tell :D
Syana
Posts: 94
Joined: Fri Oct 07, 2016 11:34 am

Re: Sim results for human errors?

Unread postby Syana Mon Jan 02, 2017 11:36 pm

Short description:
Forgetting to use a trinket when it's off cooldown.
This is more akin to something like, 'Improper use of trinkets'. Just to avoid a situation where someone reads it and decides macroing their trinket to their filler ability is a good idea.

Because as you mention yourself certain trinkets have different requirements for optimal damage output. Simply using a trinket off cooldown will more than likely lead to a dps loss during most encounters.

Return to “General Discussion”

Who is online

Users browsing this forum: No registered users and 5 guests