Friday, March 02, 2012

Good Players are Good Players

In a comment to a previous post, Spinks makes a comment that I think is illustrative of the divide between me and a lot of readers:
You seem to be thinking a lot lately about being able to measure how good someone is at PvE. But it's not going to work when you have one person who is amazing at stuff that involves interacting with the environment/ interrupts etc but can't get the pinpoint timing that you'd need to max dps output, or vice versa. Or someone who is a decent player but gets very very very stressed if they are asked to perform a raid-critical task.
Let's say that a good DPS player has four tasks:
  1. Maximize DPS output
  2. Movement
  3. Interrupts
  4. Utility Stuff
Further, let's say that you can measure the individuals performance on a scale of 1 to 10, with 10 being best.

If I am reading Spinks correctly, she believes that performance of these tasks are independent. That performance in one task is not indicative of performance in a different task. That you could have a player who is a 10 at maximizing DPS output, but a 4 at interrupting.

In my experience, this is not the case. Performance of the tasks is correlated. Scores for an individual will cluster around the same point.  A really good player might have stats of 10,9,9,8. An average player might be 6,7,7,5.

Now, performance is not absolute and innate. Things require practice. If you never interrupt, then you'll probably require some attempts to get the hang of it.  You can move all your scores up by learning new techniques and practicing more.

But in my experience, good players are good players along all dimensions. If they show skill in one task, they can pull their performance in all the other tasks to the same level.


  1. Sure, good players will try to find ways to improve their performance in all dimensions and so those will be correlated. However the relative difficulty of tasks varies quite a bit between people.

    As an example, take pre-nerf Atramedes. There was a significant danger of people dying to raid damage if the gong interrupt wasn't hit within one second. If you have a 300-400ms latency that can become an impossible task, no matter how "good" the players you have are; sometimes the light in the Pacific isn't in harmony. That same can be said for all of the areas you listed, dependent on design, of course. It's not just latency either, but a whole slew of things that break the relationship between the tasks.

  2. I agree with you, for the most part. It seems silly to suggest that someone capable of ridiculous DPS numbers - something that requires incredible feats of consistent timing - would somehow be bad at interrupting for reasons other than lack of practice.

    That being said... just like athletes, people choke. I knew great DPS with horrible, horrible tunnel vision. Beating your own record on the meters is a competition with yourself, and no one will really notice if you failed to squeeze in that extra Steady Shot; nailing all 21 interrupts on General Vezax can be an altogether different sort of pressure, especially when a failure is so obviously televised.

    But in general? Yes, good players are good. And, arguably, a good DPS that cracks under pressure isn't a good player depending on whether a steady hand is a part of the definition.

  3. You say it's not correlated, but I have played with people who did good dps but were dreadful at interrupts, or who were amazing at stuff like interrupts, debuffs, utility but never performed as well on dps meters as other players in identical gear/ specs on straight dps fights.

    I'm not theorising, I'm saying what I've seen :)

    Yes, an amazing player will be amazing at everything. But I'm not talking about rock star players.

  4. Azuriel: Yes, Vezax was precisely the fight I had in mind.

  5. My experience matches Rohan's: the good/great players I met were able to perform all tasks above average, and I've never seen people who excel at one and suck completely at another. Another mark of a good player is that he doesn't care if he gets an assignment where he will not shine (for example by topping the DPS meters): being part of a raid team means that you care more for the outcome of the fight than your personal stats.

  6. I don't think the performance is not correlated but I think the correlation is weaker than you say. I'm going to disregard players with performance anxiety etc. who might do much better/worse in all tasks based on whether their tasks are critical because I believe them to be outliers, my opinion is following:

    The strongest correlation is between player's performance in a task and the effort they make in the task - there's even a saying "genius is 99% hard work and 1% talent". (I agree that good DPS does not require something one would call a genius but I believe this saying applies not only to being a genius.) Hence, a player who puts effort to be good in a particular task is (with rare exceptions).

    I agree that most people do not distinguish between the tasks when trying but identify themselves with goals like "I'm going to be really good" or "I will do my best but if it doesn't work, then it doesn't work" or "I'm just going to have a jolly good old time" but some of them might and do only put their effort in some of them.

  7. I respectfully disagree. I've seen way too many top DPSr's not stand on the chains during Jin'Do the "Pug Breaker".

    Maybe this is the case for HM DPS folks who HAVE to min max in all areas or else they don't advance. These folks are really a different class of player.

  8. What you seem to be defining is skill. Skill is composed of multiple components that different players excel in and those different components are challenged by different raid encounters. Personally I broke it down to perception of a goal, knowledge, execution and readjustments.

    The trend you are seeing between being able to move and high dps is there because you have to know how you are suppose to move and how you are suppose to maintain a high level of damage. Both of these factor into knowledge. However there is significantly more stress on execution of various raid mechanics that require very little knowledge of what to do. The challenge remains doing them.

    I wrote a post on this a little while ago.

  9. I respectfully disagree as well. Some people, especially healer/tank mains, do great at all of it except producing maximum DPS.

  10. I don't fully agree with Rohan's assertion, but I would question how many of these "top DPS" are only "top DPS" because they're overgeared and aren't moving when they should.

  11. Good leadership can make some of the mediocre players heroes. Not because they are carried, but because of good leadership which immerses their best play. Bad leadership can make a group (e.g. a PuG or failing guild) of awesome geared and individually skilled players wipe on the easiest bosses and makes it "just not work out".

    I've found UI can help very much improving your ability to interrupt. Interrupting requires positioning, timing, and awareness. Tho depending on fight mechanic this task is an easier task than knowing your rotation and executing it correct, but there have been mechanics where a 1 sec cast had to be interrupted. Performing acceptable DPS plus performing interrupt plus following the tactic is also known as multitasking. I've seen people on interrupt duty perform awesome DPS who forget to miss one important interrupt. Missing one important interrupt is not acceptable.

    Another important aspect is the learning curve and time spend. There are some very good semi hardcore raiding guilds (2-3 times a week ~3 hrs) which are 8/8 HC.

    I agree with Kelindia; all of the above can be summed up in word: skill. Some people are skilled in healing, other in learning tactics quickly. Some people are good in dance (Alysrazor was a joke to them), others are better in theorycrafting (they are well known for min-maxing on the EJ forums). While you can say someone who does not perform as say SV hunter also would not perform as fire mage + interrupt duty I would not say this is a law. Perhaps more a rule of thumb, and if the SV hunter is following the tactics he might be worth to invest in because perhaps he can improve on his DPS thus becoming an OK player. I mean, if you do really need a SV hunter and the fellow fits in your team without an alternative he may be worth your time. That kind of decisionmaking is also part of leadership.

  12. I also disagree with this argument, but only in the sense that a "good" player likely has even ratings in all, but an "average" or "bad" player is clearly unbalanced in these ratings (assuming these ratings are the values of a "good" player). The amount of DPS a good player and a player who tunnel visions and ignores fight mechanics can be really close, so you clearly can't say that the good player's is a 9 while the bad player's is a 6.

    I've played with a few players who were like this over the years.

  13. I'm not sure I agree - at least based on the anecdotal evidence of myself and my raid team.

    We have one dps player who almost always tops the charts, but he does tend to be really quite slack about switching targets, causing us more wipes than I'm entirely happy with on HC Blobby, our current progress fight. Another player is excellent at understanding the fights, and will often come up with ideas that I haven't thought of, and consistently puts out reasonable numbers, but is rarely above 4th on the dps charts (in 10s, so bottom half).

    I'm a pretty good healer. Not amazingly awesome, but competent, usually heaing the right people, making decent and thoughtful use of cooldowns &c, but no matter what dps class I try, I can never put out more than mediocre numbers - and it's not for lack of understanding - I can theorycraft my little cotton socks off, but as soon as I have to pay attention to mechanics I lose focus on my rotation/priorities, and my dps falls off a cliff. Healing, being far more reactive from the start, doesn't seem to give me the same problems.

  14. I disagree with this statement, mainly because I see people specialize on certain things all the time in WoW. That doesn't mean that they're awesome in all areas, but it doesn't mean that they're not, either.

    Some people are great at DPS and/or interrupts, but suck at tactics. Other people get the tactics, but don't have the physical skills to do more than middle of the road DPS.

    I believe the problem is in syntax: what you call 'good', others call 'great'.

  15. I expect you're experiencing some form of selection bias. In an aristocracy guild, most of your players have already selected themselves based on being good players.

    Those of use further down the food chain have a far different experience. We run into players who are fantastic at cranking out DPS, but suck at movement. We run into players who are great tanks or healers, but are atrocious when switched into DPS roles. We run into people who are great at interrupts, but only because their entire focus is on the enemy's cast bar; the cost is, they become terrible at any other task that requires situational awareness, such as "getting out of the fire."

    So yes, at the cutting edge of the game, "good" players are "good" at all the various possible tasks. But that's only because those who are only good at one facet of gameplay have already been weeded out. So you almost never =see= those players.

  16. Rohan, what do you make of players who excel in one role but go to pieces in a different role? I consider myself a good (though not great) healer, but no amount of practice has served to make me a more than passable DPS player? Movement, utility, interrupts, cooldown use... all come pretty naturally and I can fine tune them with study and practice. Fine tuning a dps rotation seems to be beyond me.

    I guess I consider it part of being a skilled player to know when to specialize and be excellent rather than diversify and be mediocre?

  17. I agree mostly that good players are good overall as well as specifically within the 4 parameters you listed. However, cranking out good DPS eventually becomes muscle memory.

    A really good player can top DPS meters almost on autopilot.

    Movement (getting out of fire, stacking when necessary, etc) is also pretty easy.

    Utility can sometimes be forgotten (oh crap, my Kings wore off 2 minutes ago...let me fix that!) but that doesn't make a player good or bad, simply forgetful.

    Interrupts are the most iffy thing you listed. Like a couple others have said, I've raided with people that just KILL the DPS meters but couldn't interrupt to save their lives. Conversely, I know a couple people who aren't super great at DPSing but would probably score a 9 or 10 in all three of the other categories.

    I think, at least when you are talking about only DPS, Movement, Utility and Interrupts...the skills are most certainly independant of each other. If you're really good at three of the four, I think you have the right to call yourself a good player. Obviously, if you're really good at all four, you're probably a great player.

    Blanket statements are never a good thing because there are always exceptions to the rules.

  18. I think you overestimate the degree of correlation. Certainly, I think excellent players should be able to perform all tasks well (unless they really have a blind spot for some particular thing) but I don't believe that someone who is excellent at most tasks will necessarily be very good at all. For example, I can easily imagine somebody who is great at DPS, interrupts and easy movement, but only average at kiting.

  19. I think it's also important to factor in priorities.

    Someone's performance in DPS compared to interrupts/dancing may not necessarily be about skill.

    I don't raid anymore. But in my day, I knew MANY high DPSers who consciously chose not to interrupt and blamed it on poor skill, because they are dedicated wholly and solely to pure DPS and do not want to sacrifice even one GCD.

    By the same token, there were several self-sacrificing team players who would pick up the interrupt/'jogging around to hit adds' slack for others (because every moment melee DPS spends running is usually a moment not doing DPS), because they wanted the team to succeed. Performing more than their expected share of these DPS-reducers did not make them bad at DPS.