Seems to me like you’re probably not going to be able to square this circle because the data that you have access to is incomplete, or rather, not as comprehensive as the data Blizzard has. Also, without knowing Blizzard’s exactly methodology for how they came to their conclusions, you’re going to have a real hard time replicating their results. I’m sure that’s immensely frustrating for you, but I get the feeling that what you’re trying to do is a little like trying to finish an incomplete puzzle in the dark: it’s hard to see, and even if you get everything in the right place, you’re still missing pieces.
I really don’t have any beef with you, Micro, but maybe it’s time to just… Chalk this one up as a loss and tackle the next thing, ya know? It’s real quiet over at the WD forums if you wanted to come and help me kick up a fuss, see if we can’t get the poor WD some more love and attention
Looking at this graph, at the top end WD and DH performs similarly: similar GR being cleared with similar paragon and duration.
However, seems like DHs data in the 1st 1000 starts at 117 onwards while WDs 1st 1000 starts at 107.
Note that in my post here, I am not commenting on anything about WD and DH. My observations are as noted above. However, some ‘geniuses’ who like to average GR values as a gauge of balance like to twist observations by averaging data points across GRs. As I stated 10000000 times, this approach is wrong. As you can see in the table in my post (given at the end), I averaged GRs of all the clears and stated that this way of data evaluation is not right. I am mentioning that averaging many GRs across each other is wrong and it will give wrong observations (you can see this explanation below the table).
So Micro has been using that table to say that I said the WDs are 7 GR weaker. Lol. WD performance is similar to DH, however more players are preferring DH because it may be better for pushing overall. But I prefer not to comment on other classes. And see this happens, some people will bend over the data in any way to support their agenda.
As somebody who looked at these leaderboards and their meaning, there is no 1 fixed value of difference among classes. The difference varies sometimes slightly sometimes by a lot depending on the compared GR level. If one has to put a number to this difference, he has to consider the mean curve. It is the most logical approach and perfect balance will not be 100% possible ever. Some builds will simply perform better at higher GR while some will excel at lower GR.
I’d argue it’s dead easy to make a fuss, it’s the whole ‘effecting real and tangible change’ that’s considerably harder, but point taken
I don’t think anyone is saying that Blizzard is infallible, just that they have access to more data points than you or I have, so maybe - just maybe - they’re seeing something we’re not?
Anyway, I want it noted for posterity that I tried. Promise you’ll be nice to us WDs when we’re smashing solo 150s off-season?
No, just no. You are wrong there is no “maybe”. At this point you just look delusional. Accept that you were wrong, you don’t even need to do it publicly, and do so quickly. If not for the sanity of everyone else on the forums then do it for your own mental health. Smashing your head into a wall will eventually cause permanent damage.
However, for a model 5K player, Blizzard’s table showed in non-seasons that average WD would clear GR 130 and DH GR 125. Your analysis shows that at the top end, their performance is similar.
I may very well be wrong. You believe Blizzard is accurate. They may be absolutely correct. Unfortunately, there is no way for anyone outside of Blizzard who can make a determination of “facts” given the transparency of the data and analysis methods.
What I can say is using Blizzard’s reported data, it seems counterintuitive that an average 5K paragon crusader would clear GR 136 in seasons and GR 138 non-seasons.
This would suggest either their is an inherent issue in their data scaling, severe limitations in a dataset, or something else such as maturation kinetics of the GR clears. Blizzard said that both tables used data derived from similar timepoints.
Ha! You are making it sound like the Blizzard math people are sitting around manually stimulating their “Kinet’s” by hand!
I think you should rephrase that sentience a bit!
I was trying to get at the idea for the non-season leaderboard people already had some gear on hand (obviously not the AoV set at the start of the new era of course) to push/get a leaderboard spot versus early season where it would be much more difficult to project/upscale the data. A reality check, though, would be that there 5K projection for seasons had to be similar to above non-seasons.
You don’t have access to all the data. You only have access to leaderboards data.
You can’t just analyze based on leaderboards data.
So, what’s the point in keep arguing whether you’re right or wrong? Your analysis was never good from the beginning, because you don’t have access to all the data. Blizzard clearly told you so, are they wrong about this?
It doesn’t matter how good you are with numbers, if you don’t have good/complete data to begin with, your analysis is just garbage. You may very well be just shut the f up.
I do not know if I am right or wrong. Neither does anyone else who does not work at Blizzard.
For clarity, I am not arguing that I am right or wrong, I am simply saying that there are some peculiarities in the presented data that seem counterintuitive. Other posters speaking in absolutes “MicroRNA was wrong” also do not know the merits of that statement given what you just wrote.
Yes and Blizzard also mentioned that the leaderboard data overestimates the “average” player. Therefore, you can see how their numbers compare.
Complete data is not needed to analyze relative build strength. What is easily available in bulk for the leaderboard are GR tier, paragon level, clear speed. Manually, you can get the legendary gem levels, augments, and builds for the leaderboard clears in game.
I have actually done this in the past and posted on it. One oddity at that time was I previously assumed that the legendary gem levels for a “meta class” would be higher than a non-meta class (excluding capped gems and pain enhancer). It turns out that wasn’t the case for the “average” players I analyzed at the top end of the leaderboard.
Moreover complete data actually is problematic. You can think of it this way: Informative, high quality data informs, bad/irrelevant data muddies the waters.
For example, there could be a high paragon players that actually avoid the leaderboard to avoid “detection”. This would make “complete data” for classes that are easily botted skewed.
You really like to cherry pick words right? How about this: You don’t have access to all the relevant data for analysis. Therefore your analysis is garbage.
When dealing with human behavior, it is not possible to have all the relevant data.
For example, one of the most important pieces of information to compare class strengths is what percent of each class’s players try to push with that class versus playing casually without testing their limits.
This is one of the arguments about why looking at top clears can give insights. These players are pushing a builds limits. Their data may not reflect “the average” player but is much better to use to estimate build potential.
Folks, again: I urge you to STOP REPLYING TO TROLLS.
It’s evident by certain posting trends that some folk will reply ad nauseam even when they are proved wrong, shut down through official channels, and otherwise dunked on. There is no point in replying to these folk because, contrary to their claims, there is no discussion to be had–only endless argument, which is equally pointless, because the Fun Police have (for now at least) been officially shut down.
Please: Don’t feed trolls. Don’t reply to them. Don’t give them attention. Ignore, ignore, ignore.
No worries, my friend. I’ve been there plenty of times. It’s very easy to get sucked into “discussions” with trolls because they’ll stick to their guns even when the guns have been taken away, smelted, and made into candlesticks.
Really, the bottom line is this: Blizzard has made an official statement on balance.
They’re doing it the way they’re doing it.
Not everyone is going to like it, and they certainly don’t have to like it, but it is what it is.
The matter of balance is, for now, closed. The Fun Police have been silenced with an official statement, and a few of them still want to make a fuss. As long as we don’t reply to them and feed them, they will, eventually, go away. And we’ll all be happier for it.
It is not always about having the most complete data but the most informative data.
To illustrate this point, there are cancers of unknown tissue origin. To identify tissue origin, biomedical researchers have used global gene expression profiling to interrogate the expression of ~20,000 or so genes. In terms of specificity and sensitivity, a global analysis works rather poorly in comparison to looking at the most informative genes. The problem with a global analysis is that it has too much “useless” information that makes a predictive classifiers much more noisy and unreliable.
Looks like microtroll got pwned hard. Now he is trying hard to play everything down.
That is him after you guys do that.
