Susan

http://onlinemariogames.net

Bosses that enrage or change tactics (added damage) at low health are not uncommon, and the “end of the fight” can sometimes be a higher pressure situation than the start of the fight. Being able to down the last bit faster is often an advantage — and definitely when pushing new content.

An example of some spreadsheet hunters: look if they have the Kill Shot glyph. Most useless non-PvP glyph, but many hunter’s take it to take them higher on the dps. The sheet told them it does …

In reality it is just a show-off glyph they can take out of their pocket when the raid-team already got the boss down to 20% (or wiped X times trying). The thing is the team will also get it down then without the speardsheet-showoff hunter shaving of a few secs of the fight and pushing him upwards on the recount-logs. He’d better glyphed on something that helps the raid also during the first 80%.

(And yes, there are those one or two multi-stage bosses where this is different. Sure, but did you honestly glyph on purpose for that boss or are you just trying to find excuses? )

Raises a question / point which I’ve been at some contention with you over for a while, but haven’t mentioned, which is your definition of “Theorycraft” – and to extension, what part of the puzzle it addresses.

On the whole I agree with your statements — Spreadsheets are not the whole picture & that experimentation is an important part of answering the question “How do I get the best out of my character?”

My contention is that empirical data collection is not the same concept as theorycrafting. In my mind I divide the two data sources — just like the division between “Experimental Physics” and “Theoretical Physics” — To *me* Theorycrafting is the developing of the models aimed at developing a simulation of the game mechanics.

Using these models to predict in game behaviour and help make decisions on gear or glyphs is ok, but hinges on the complexity of the simulation process used .. To date spreadsheets have very simple simulations — about the only fight even close to the simulation is Patchwerk.

Answering the questions that involve more complex situations, require more complex simulators. OR the development of empirical experiments (in practice both).

When you present your “scientific” exploration of topics, I see it as a step beyond the definition of “Theorycraft” because you take it to the operational experimental stage. As oppose to developing experiments to test the theoretical models.

To me, you’re addressing the questions of “what is better” rather then “extending theorycraft” (i.e. improving the models)

Regardless … Anyone who claims the models in the spreadsheets are 100% right, is crazy, and anyone who believes the simulations the spreadsheets run are representative of real fights is likewise loony. Spreadsheets offer an interesting perspective – a good point to produce hypothesis for empirical testing too – but are not the be all and end all of knowledge.

Paws

But I’ve found too many discrepancies between either my own experience and the spreadsheet, or Frost’s rigorous testing and the spreadsheet, to trust it for anything beyond gear decisions.

So agreed on all counts. For those that are disagreeing on the grounds that the spreadsheet is just a tool, guide, etc. I doubt Frost would disagree with you. That is the intended purpose, and the spreadsheet does a great job being useful. The problem is the people that DO use it as their only source of decision-making. And many such people exist.