How to make tanks more fun to play (devs please read)

[LineageRazor]

On what grounds do you think the bottleneck should be healer? If the server is 20% tank, 21% healer, and the rest is DPS, then we would still more often than not run into a tank shortage. The gap doesn't need to be large for it to be the main shortage. You don't trust a chart, which is based on actual data SE had on their servers, yet them you throw out nothing but opinions as if it is more factual. I'm more inclined to believe a chart made up of actual data instead of your assumptions any day of the week.

On the grounds of THIS CHART. I'm basing my assumptions on the chart, which I clearly stated is NOT a correct representation of the actual population of jobs that people play on a regular basis. The CHART says DD 51%, Tank 23%, Healer 20% - hence, healers are the bottleneck BY THIS CHART. This is clearly not the case in-game, where healers regularly have to wait for queues, and tanks rarely do.

As for opinions, it's not hard to see from experience that yes, tanks get into queues instantly, and healers have to wait a short while, as I play both WHM and PLD. DD's do enough complaining on the forums to make it clear that the queues are worst of all for them. So, yes, this chart is clearly not usable as proof that there's no queue problem. All very tidy empirical data, accumulated through first-hand experience and second-hand reports.

[Tiggy]

On the grounds of THIS CHART. I'm basing my assumptions on the chart, which I clearly stated is NOT a correct representation of the actual population of jobs that people play on a regular basis. The CHART says DD 51%, Tank 23%, Healer 20% - hence, healers are the bottleneck BY THIS CHART. This is clearly not the case in-game, where healers regularly have to wait for queues, and tanks rarely do.

As for opinions, it's not hard to see from experience that yes, tanks get into queues instantly, and healers have to wait a short while, as I play both WHM and PLD. DD's do enough complaining on the forums to make it clear that the queues are worst of all for them. So, yes, this chart is clearly not usable as proof that there's no queue problem. All very tidy empirical data, accumulated through first-hand experience and second-hand reports.

That number is also many months old. It's not really useful to diagnose today's experiences.