[POE] Building Better Builds: Forum Posts, Part i

I've become interested in making my Path of Exile builds better.

One way to do this is to see how other players are creating their builds. Since this is a pretty big topic, I'll first focus on passive skill tree choices by players.  I'm particularly interested in these because many players post guides to their builds - including passive skill trees - on the forums. (Data time!)

I grabbed the 600 most recently updated threads in each class forum and extracted build links on the first page of each of these threads. Because PoE's passive skill tree is constantly changing, I removed builds that are no longer valid with the current skilltree or were less than 80 points. Because sometimes authors post multiple versions of a build, I selected only one build from each thread (the build closest to 100 points). This left me with 914 valid builds.

	Duelist	Marauder	Ranger	Scion	Shadow	Templar	Witch
valid builds	101.0	106.0	144.0	123.0	166.0	112.0	162.0
all builds	707.0	540.0	657.0	732.0	633.0	591.0	615.0
average posts	68.8	79.3	47.7	68.2	49.1	58.8	80.6
average views	25071.7	33006.9	17866.2	18892.1	14617.9	15087.8	32962.1

Curiously, Templar has the least build options to chose from, combined with relatively few posts / views for each thread. I'm curious whether this will lead the developers to consider rebuilding the templar starting area.

Defensive specialization: 
Whether in a hardcore or softcore league, one of the biggest challenges in Path of Exile is simply staying alive. For this section, I'll focus on player choices to get life, energy shield, armor, and evasion. When evaluating life- or energy- based builds, I omitted any build with less than 50 of that stat (because, well, death).

A few observations:

• Life-based builds tend to get around 200 life, with marauder builds getting slightly more life than the other classes. I'm actually surprised at how many builds get less than 200 life, which I considered an absolute necessity. 
• Rangers and shadows tend to focus on evasion, while duelists and marauders focus on armor. 
• Of all the classes, only witch has multiple builds getting a 'reasonable' amount of energy shield for CI / low-life. 

There is, of course, a potential problem with this data. The wide range of life / energy shield values suggests some of these builds simply aren't viable. For example, a large number of builds from new or inexperienced players might tend to undervalue defensive stats. To correct for this, I weighted the data by the number of posts in the thread (on the assumption that high-quality builds are more likely to get attention).

While weighting the data did remove a large number of the low-defense builds, the basic story is unchanged: players tend to get at least 200 life, energy-shield based builds are uncommon, and strength-focused classes get armor while dex-focused classes get evasion. It is notable that a few high-attention builds (like cyclone marauder) tend to dominate the data.

Weapon specialization:
Turning to offense, here I looked at whether a build grabbed a damage specialization node (+damage with a weapon type / spell).

A few features that stand out (to me):

• Wow, caster options are popular (even omitting elemental damage builds).
• Ranger is overwhelmingly the most popular choice for bow builds. The lack of a bow-templar makes me consider running one.
• One-handed builds are more popular than two-handed builds for all classes.
• Swords are slightly more popular than axes, despite the focus on soultaker in the current meta, though this is potentially due to buzzsaw builds. 
• Specializing into wands / staves are remarkably uncommon options (2.6% and 1.4% of builds, respectively). Staves make sense - other than Pillar builds, not many characters go for a staff (they're also devilishly hard to roll). But wands? I seem to remember everyone trying wander early in the league.

Thoughts:
I'm frankly a bit surprised at how few defenses many builds grab, particularly given the common refrain that 200% life is a bare minimum for PoE. I'm equally surprised at how many builds choose to specialize in a weapon, given how that limits purchasing options at endgame.

There are plenty of places I'm considering taking the analysis. I'm definitely going to look at keystones. I'm also interested in some general descriptive-stats for the passive tree (shortest paths, nearest clusters) and how this influences the build meta.

So many fun data adventures to be had.

[GW2] Professional Voting

Today, Guild Wars 2 extended their Collaborative Development Initiative (CDI) to ask players to post their feedback on the ranger profession.  (Note: 'professions' are what most MMOs would classes).

Ranger was chosen in this thread, where the developers asked players to list 3 professions (in order of priority) they would like to see discussed first.  But isn't choosing a profession based on posting in a thread open for 24 hours on the forum problematic? Particularly active forum users would be likely to overwhelm the thread. And its reasonable to expect that posters would be biased toward their own class, instead of looking out for game balance as a whole.

So how reasonable was the thread as a method of getting feedback?

The results:

Ranger was the most mentioned profession, followed by elementalist and then necromancer to round out the top three. Rangers are particularly weak in GW2's pve, and one playstyle (bearbow) is consistently viewed with scorn.

Breaking down the votes by order of priority indicates that not only was ranger the overall top choice, it was most frequently listed as priority #1 by voters. Most of the ordering between professions remains the same.

But are these votes contaminated by bias?

Prior Posting Activity

One objection to this poll might be that not only are forums generally unrepresentative of player opinions, but frequent forum posters might be likely to overwhelm any thread. Hardcore and casual players might be expected to disagree on which professions need the most attention.

While generally valid objections, they don't apply particularly well to this poll.

For nearly 20% of voters, voting in this thread was their first post on the forums. For others, posting is a pretty frequent activity. Voters had an average 205 posts before the thread (with a maximum of over 4600 posts!). 

Curiously, however, prior posting activity doesn't change the results. Both first-time and prior posters overwhelmingly agree on their preference order, favoring rangers, elementalists and necromancers. (Though mesmers do a little less well among first-time posters).

Profession(al) Bias

Another way to look at the votes is to examine what other forums voters are active on. Players might be expected to be biased toward voting for their own profession, which could be reflected by their posting in profession-specific subforums.

Here, I took a look at the past 50 forum posts of each voter (excluding first-time posters), and recorded whether they'd posted in one of the profession-specific subforums.  I compared this against their listed vote preferences.

In general, the votes are only slightly biased toward forums voters are active in.  For example, 30% of those who ranked elementalist as priority #2 had posted in the elementalist subforums, compared to only 15% of all voters. But this pattern doesn't hold generally. Those who picked engineer, mesmer, thief or warrior were no more likely to be active in those professions' subforums. So, in general, forum posting activity is at best weakly related to vote choice.

The exception - as ever - is rangers.  While only 5% of voters had posted in the ranger forums previously, 45% of those who listed ranger as their top pick had posted in those forums. Among top choices, ranger is the only profession that shows such a strong pattern. However, even excluding these voters, ranger would still be the top pick.

Thoughts

Clearly, a 24-hour poll isn't the best way to generate clear feedback on which profession requires the most attention.  In this case, however, the ranger profession is so overwhelmingly in need of attention that the effect of these (otherwise valid) concerns is overwhelmed.

Fingers crossed for some pretty excellent fixes.

Disclaimer: I have a ranger at 80, but I only use it to loot jumping puzzle chests since returning to GW2.

The Great Server Migration

Thought I'd share a visualization of how the SWTOR server migrations have shaped realm populations. Here I'm just taking a look at realm status (1 for light, 2 for medium) over time, by region and type. Short story: origin servers are ghost towns, destination servers are hopping (but that much was clear already).*

Slightly cooler story: even before the migrations, as far back as April, destination servers had higher populations than the other servers in their zone / type. There is one exception. Until just days before the migration, The Swiftsure averaged a higher population than The Bastion (which ultimately became the destination server for West Coast PvP servers). Makes me wonder why Bastion got destination status. Rounding errors between status and true population numbers? Server hardware?

I also went ahead and re-collected data on posting volume on the PvP forums. Activity - in both posts and views - is definitely down from earlier in the year. Curiously, while 1.2 saw a remarkable increase in activity on the pvp forums, the 1.3 release had almost no effect. It looks like a focus on the server migration drew posters away from the forums around the roll-out of that feature.**



*The gap in data from the start of April comes from my failure to backup a week's worth of data and then getting a drive failure.
**The white band around May 5th is a day of data I'm missing.

Voice of the Player: the PVP Forums

Mos Ila: Honorable Mention, Hives of Scum and Villainy

The PVP forums may be the most wretched hive of scum and villainy in the galaxy, but there are always a few holdout posts that I find helpful, or even inspiring.

What about the din, though?  Can something be learned from the countless threads of complaints, anecdotes turned into truths, and general hatefest?

I ran a program to download the approximately 32,000 threads posted to the PVP forums, taking the first page only.

Most threads are only active for a day before they no longer get replies (75%).  After a week of activity, only 5% of threads remain active. The median post gets around 5 replies and 240 views. There are a handful of notable outliers - including one post with around 330k views (a dev thread from Jan 18 that gave an update to Ilum's pvp).

Just from this activity, its immediately clear that 1.2 brought a dramatic increase in forum post volume. With the general renewed excitement over SWTOR with 1.2, this is a good indication that the volume of posts matches interest in the game. Posting is something that complements play time, not opposes it.

Class Warfare: calling down the nerfbat
What about the ever present cries for nerfs on the forums? Is the rise-to-power of sent/marauder reflected in forum posts? The stunlocking dominance of ops?

For this, I'm not going to do anything super fancy, just a simple search through the first 10 posts in each thread for mentions of each class (accounting for things like spelling mistakes, or clear references to a class via a popular spec, like 'concealment' matching to operative, or 'galactic overlord of awesomeness' to sorcerer.) I'll also combine the Empire and Republic advanced class mirrors, and reference them by their Empire names.

Looking at the percentage of threads that mention an advanced class, I'm actually astounded just how well that worked. The fights between operatives and sorcs, followed by the overwhelming numbers of sorcerers, until the rise of marauders with 1.2... its all reflected in the volume of mentions of each class on the PVP forums. In contrast: powertechs, snipers, and juggs are all but invisible.  Assassins have made a respectable rise, one that I'd expect to continue, given their impressive on-demand cooldowns and burst.

From this, I'd say forum chatter is a reasonable way to capture community awareness of issues.

Topical Applications
Next up: I sampled some of the other issues that players raised for discussion on the forums, looking mostly at pvp zones, exploits, and some quality of life issues. Again, for each category, I'm looking at word counts, accounting for things like misspellings and including related words.

The first plot nicely captures the rising attention given to Ilum, and its eventual irrelevance. Alderaan - the least innovative of the warzones - also gets the least attention.

Regarding bugs and exploits, mentions of bots and hacks are relatively infrequent - and steadily decreasing over time. Hopefully this is a reflection of the developers getting a better handle on the situation. Lag and bugs, however, remain persistent issues.

I'm glad to see that despite all the attention given to 'unsubbing' threads... they're rare. Like, people spend more time talking about powertechs rare. Premades and stunning, though - they're frequent issues.

But forum goers are weird!
For ages, developers have noted that players who post to the forums are 'different' and don't really represent MMO players in general. I think that makes intuitive sense.

But that only goes so far. Again I'll look at class balance. But this time, I'll divide posters into four groups, based on the number of posts each forum-goer made in my sample.

Post Groups	1 post	2-10 posts	11-100 posts	101 or more
Individuals	10,802	13,009	3,220	218
Posts Made	10,802	51,010	88,336	40,286
% of posts	5.7%	26.8%	43.9%	21.2%

Even though individuals who made 101+ posts are less than 1% of the population, they make over 20% of the posts in my data.

But those who post often are no different from those who post infrequently. Whether looking at those who make one post, or more than 101, the trend is still there. Each group is currently discussing marauders most frequently, was talking about sorcs before 1.2, and largely ignores the powertechs. It is definitely the case that the trend is smaller in the one-post group, but it is still there. 

Class balance concerns are relatively even across each group.

Forums: Windows into MMOs.

I'm pretty impressed with just how rich a source of information the forums can be (particularly given that this took... all of a couple hours to run, once I'd downloaded the data). 

But this was just what struck me while playing with the data. If you've got a pvp-forums related question, hit it up in the comments (or email) and I'll see what I can find.

That's no thread...

This week's feedback request is definitely a poll in disguise:

"What planet in the Star Wars universe do you want to see in Star Wars: The Old Republic that is not currently in the game?" (source)

Hmf... I want to see the breakdown of responses too!

I collected all the words used on the first 67 pages in response to the thread, and searched them for the just over 3600 planetary entries in the Star Wars universe listed on Wookiepedia. This gave approximately 1700 recognizable suggestions for planets from 670 different posters.*

Voting early and often, huh?

Here are the votes, as either a count of planetary mentions or weighted by the number of planets listed by a poster (ie, mentioning two planets gives each half a vote, three planets gives each a third... etc). I've also dropped planets that got less than three mentions (another 71 planets).

Be sure to click to enlarge!

I love that Kashyyyk is at the top of the list - the vertical element could really add a ton of fun to SWTOR questing. I'd love to play with a charge or updated MGGS to move through the trees. A jetpack to ease the concerns of heights and pvp would be pretty fun, too.

Old KOTOR favorites (?) seem to make it to the top of the list - like Dantooine or the aquatic planet Manaan. I have to confess, I'm not a huge fan of aquatic combat in MMOs: that also disqualifies the highly-ranked Naboo from selection. Ditto with Mon Calamari... which might win the contest for the most comic name on the list.

I'm a bit shocked that Dagobah didn't make it higher than #11. Training sequence daily quests? Lift a drowned fighter from the muck. Now jog through the bog while carrying a muppet. The cave alone would be a phenomenal flashpoint / daily quest.

Though, I'm a bit surprised that the devs polled the audience on this one. The SWTOR universe has so much awesome in it, I'd figure they'd just run with whatever idea geeked them the most.

*Just a simple search with some corrections for basic spelling mistakes (Kashyyyk has 3 ys) / abbreviations. While its possible to map "wroshyr trees" to Kashyyyk / account for feedback that looks like "meesa not a fan of Gungans"... I'm mostly interested in the infamy of a planet.

Voice of the Player: Community Q&A

One of the more awesome ways that Bioware has chosen to interact with the SWTOR community is through the developer Q&As, where they solicit questions on a forum thread and then post responses. So far, there have been six of these: on Feb 10th, 17th, and the 24th, March 2nd, 16th, and the 23rd.

On March 30th and April 6th, instead of taking questions from the community, the devs answered questions they'd seen posed on the forums.

I think this change is really unfortunate - so I'm glad they'll return to the old format next week. The Q&A threads are a great way to take the pulse of the community. Just take a quick look at the words most frequently posted in each thread (as wordmaps)

I'm immediately struck by what I don't see - references to LFG don't make the wordlists (it peaks out at the 102nd most used word). Same with references to bugs, or topics related to bugs. Instead, the pattern that pops out at me is the rise of 'server' - typically as references to merges and transfers for the low-population realms.

A number of the discussion threads have mentioned the relative silence of the devs on same-gender romance options (sgra) - often claiming that it was the 'most asked question' in the Q&A threads.

Words related to SGRA are certainly frequently used in posts... but take a remarkable decline after the guild summit (March 5th) gave a clear indication that they were coming "this year". I think this is a really cool indication of how being clear and direct with an answer can ease community concerns (also, Daniel's handling of the question was brilliant.)

Meanwhile, questions about the legacy system have consistently increased... which isn't too surprising given that it is relatively untested on the PTS and that most of the information about the benefits from the system has come from third party sites.

I'm definitely going to keep working with the forum posts data... maybe explore the evolution of topics in a little more detail.