gone in 60 seconds

April 29, 2009

I’ve withdrawn my last post on dual specs because I don’t believe the numbers are right. Those low numbers for dual spec adoption just aren’t credible. I don’t know whether there are still issues with the armoury itself, or something that I’m doing wrong, but I’m working on it.

I guess the number of patches and hotfixes in the game tells us something. OTOH it may just be my stuff-up.


You may not have noticed yet, given all the 3.1 (and now 3.1.1!)  fun, but a whole slab of character stats have disappeared from the armoury. The most interesting ones that have gone are the detailed BG performance stats. Strangely enough, the raiding performance numbers are still there – the missing ones all seem to be PvP related.

Now if I were a, like, y’know, paranoid kinda person, I’d be putting forward the following conspiracy theory. PvP stats are about class performance and raid stats are about group performance. Raid stats tell you something about a guild because performance depends on the ability of the guild to coordinate, to lead, to control the Leeroy Jenkins element etc etc. And indeed there are websites out there that do exactly that:  rate guilds by how many, and which, raid dungeon bosses they’ve downed.

But PvP stats, being all about the mano a mano thing, tell you something about the relative performance of classes. And that subject really does seem to be Blizz’s bête noire these days. Just exactly why they’re so focused on it escapes me – but then I never go anywhere near the official fora so maybe that’s why I’m in the dark.

My conspiracy theory would be that they don’t want anybody to be able to just run class balance through the ol’ spreadsheet to see what comes out the other end.

I made a modest contribution to using the BG data to look at class balance here on this blog. But I believe I was pretty careful to say, like dude, we don’t expect the classes to be balanced across any narrow set of performance measures. Classes that can tank and CC and heal are people too.

But anyway I’m not the paranoid type, so that’s enough of that. It’s just a silly game. Let’s move on.

the 19 twink warlock

April 8, 2009

I’ve got a test page up for my BG twink reports. Check it out here. I’m still mucking around with various parameters – especially the selection criteria for toons in the sample, and the filtering to hide the long tail of odd individual choices that seems to appear in every aspect of WoW.

On that last point, the danger is that some important twink items are likely to be difficult to obtain, and therefore less popular, so I’m reluctant to do too much filtering of the results. The most popular choices may not be the best choices here.

The results do match the various guides on the net for twinking warlocks so I’m reasonably happy with progress to date. The thing that concerns me is that the sample size is very small – even with a million characters in my database, by the time I try and select a sample of the most powerful BG toons at each x9 level, I’m down to a hundred or so. That’s not a lot.

For that reason I nearly decided not to show any percentage popularity measures – the exact numbers just aren’t all that meaningful. But some choices are clearly a lot more popular than others and it seemed important to report that.

Anyway, any feedback would be good, especially if you have a twinked 19 warlock. If you think the report misses important details, I’m happy to hear that too. Make your argument calmly and I’ll take it into consideration. This type of report is an experiment and it just may not work so well because of the small sample size.

Just to complete the picture, here is a set of battleground class performance charts for each of the x9 BG levels plus level 80. The y-axis now shows average deaths per game and not the inverse, so the sweet spot of high-kills-low-deaths is in the bottom right hand corner of the chart.

The sample consists of all players at each level who have played 100 or more BGs. The data is from patch 3.0.9.

There’s a lot of interesting things to note in those charts, especially when you compare the same class at different levels. Some are effective at all levels, others appear to change roles as they level up.

If you want the executive summary, these are the points that strike me:

  • DKs are OP
  • Rogues aren’t, even though people think they are
  • Warriors are fragile, despite all that armour
  • Warlocks are still a force in PvP as long as you don’t mind dying a lot
  • Baby Paladins may be easy meat but the adult of the species sure isn’t
  • Hunters seem to be the consistent high performer, but that is probably because they just play the same role (of ranged attacker) at every level

Do exercise some caution when interpreting these results. In particular remember:

  1. Some classes have fewer attacking players and therefore a lower average kill rate just because they have a healing tree. Other classes may spend a lot of time CC-ing instead of attacking.
  2. Some BGs have objectives that conflict with straight PvP. For example in Warsong Gulch, the classes that spend most time running the flag will have a lower kill rate because of that.
  3. This is data aggregated across every BG accessible at the level. There may be specific features of individual BGs that make certain classes more effective there, despite these charts.



BG Class Effectiveness, Level 19



BG Class Effectiveness, Level 29


BG Class Effectiveness, Level 39

BG Class Effectiveness, Level 39


BG Class Effectiveness, Level 49

BG Class Effectiveness, Level 49


BG Class Effectiveness, Level 59

BG Class Effectiveness, Level 59


BG Class Effectiveness, Level 69

BG Class Effectiveness, Level 69


BG Class Effectiveness, Level 79

BG Class Effectiveness, Level 79


BG Class Effectiveness, Level 80

BG Class Effectiveness, Level 80

If we take some of the battleground stats from the armoury and use them as “performance indicators”, we can get a measure of class performance, as well as individual performance. There are some problems with this, as we’ll see in a minute, but the results are interesting nevertheless.

The performance indicators I like best are killing blows per game and deaths per game, since these give some indication of the performance of the character or class, as opposed to the performance of the team. (Of course they are not completely independent variables. For example an effective team may try to protect the squishies and the healers who should then have a lower death rate.)

Here we take a sample of level 19 WSG players with more than 100 games played.  We can chart these two indicators to give a view of the effectiveness of each class. In this graph I use the inverse of deaths-per-game, so that the sense of each axis is the same. Basically – the further away from the origin the better, along both the x- and y-axes.

This is what we get:


Class performance, 19 WSG.

There are a few interesting things here. The first is that there a clear grouping of classes in the top lefthand corner (around priests) whose role is not primarily individual combat. Priests heal, Druids CC, heal and do the bear flag run thing. What surprises me is to see Paladins so close to this group. Does anybody know why that might be? What do pallies do in WSG at 19?

Out along the bottom right, you see the attack classes – they die a lot more than the druid/priest/pally set, but they do a lot more damage too. The most unexpected thing here is the power of the much-maligned warlock. Although they are the squishiest class (closest to the origin on the y-axis is bad, remember) they can certainly dish it out – more than rogues. The stand-out class is the hunter – significantly more a damage dealer than the rogue, for basically the same death rate. I suspect that a lot of players don’t know that, since the rogue seems to be the most popular choice amongst the serious WSG player at level 19.

The division between attack classes and “support classes” (perhaps an unfair term since running the flag is a bit more than “support”) helps explain the observation I made in the last post – that there is not a high correlation between killing blows and deaths. Several classes are doing something other than killing.

Before we leave this subject, I want to show that these averages are interesting and informative, but they must be taken with the proverbial grain of salt. Averages tend to be influenced by outliers, and to abstract from the fact that the better players (or the richer twinkers perhaps…) can get a good performance from most classes. If we take three classes that are close to each other on the above chart – mages, warlocks and warriors – and plot the individual values used to form the averages, then we get a different picture:


Character performance.

You can see that a lot of players have similar performance and the averages tend to be skewed in one direction or another by smaller groups of outliers. On the other hand, the averages do provide some real insight since you can see that mages have no strong killers at all and that must reflect something about the class as well as something about the player.

Warsong Gulch chart porn

March 20, 2009

I’ve been having a bit of fun with the battleground stats data from the Armoury. There are plenty of good strategy guides on the net for each battleground, but it’s interesting to see how well the strategy is reflected in the data. A good chart really is worth a thousand words…

Warsong Gulch is our example here. You probably know that there are a number of traps in what seems like a simple game. Trap number one is to be on a team where everyone wants to run to the middle of the field and have a big punch-up while the flag carriers run by unchallenged.

The following charts are for level 19 characters who have played at least 100 games, so we’re not talking about noobs here. But you can still see all the nuances of the game in the data.

For a start, teamwork and focus on the flags rather than on scoring kills is vital to winning. So, if we chart games won to killing blows struck by individuals, we don’t find a strong correlation. You can help to win by doing other things than killing – CC-ing, running interference, guarding the flag rooms.

Killing blows vs games won.

Killing blows vs games won.

Being able to drop the opponent does have one important role:  getting the flag back when the other side is running with it. We can chart flags-returned to killing blows, and we get a stronger correlation:

Flags returned vs killing blows

Flags returned vs killing blows

But again, capturing the flag from the other side is about skills other than fighting and the data proves that. In fact, this is my favourite chart since the whole ding an sich of the game is in there. The best characters at capturing the flag are generally those with lower killing blow scores. They’re too busy running and hiding to be killing. So you get a gentle negative correlation as the chart shows:

Flags captured vs killing blows.

Flags captured vs killing blows.

The purpose of this exercise is to see if these indicators can be used to pick out twinks from the data. I’m confident that they can. So, the next chart shows deaths per game vs games won, and suggests what we all know – that the character with the best gear and enchants stays upright for longer. They spend less time running back from the spawn point and more time on-target.


Deaths per game vs games won.

Dead toons don’t win games – it’s not rocket science.

One interesting point is that there is a poor correlation between deaths per game and killing blows per game, which surprises me. I would have thought that twinked characters both killed more and died less. Perhaps there is a kamikaze style of player that kills a lot and dies a lot, and a more um… tactical… toon that can dish it out without getting too much in return.

Deaths per game vs kills per game.

Deaths per game vs kills per game.

One final point – if you look carefully at the last chart, you will see the bane of the armoury data miner – the annoying little outlier that makes all our averages skew away from the median. That character circled on the bottom right of the chart is one lean mean killin’ machine, a real leader of the pack. Here’s his armoury profile – check him out. He’s a twink, no surprise there, but he doesn’t seem out of the general range of twink stats. To me, he is a reminder that skill does play a part in the game too. Mind you, he played nearly 900 games to get that good.

toon limbo

February 5, 2009

A couple of blog readers have asked questions about how many characters in the armoury are not currently being played.

We all have parked alts that, maybe, get a day out at the AH or the bank every now and then if they’re lucky. And the suggestion was made that there are a lot of abandoned toons at the lower levels because players try new classes then decide they don’t like the playstyles associated with that class.

They’re excellent questions. I had this idea that getting answers was going to be hard, but last night I had a eureka moment (perhaps somebody is spiking the local water supply…) After a brief facepalm, it was off to the SQL editor for a bit of date manipulation. And ta-da, here we have an answer:

Months since last played

The armoury supplies the last-login date for each character. It’s just a matter of working back from the date when I did the current armoury scan. I’ve based the chart on months-since-last-played to keep it small and simple; splitting up the data by weeks or days doesn’t show anything very different.

You can see that nearly all level 80s have been played in the same month as the scan. Since the scan itself takes a couple of weeks, to me that means that the vast majority of level 80s are actively played. At level 70, just over 30% haven’t been played for a month and 11% haven’t been played for two months prior to the scan. So the proportion of inactive or low-active alts in the level 70 population might be around 30 to 40%.

The number of inactive toons in the lower levels (I’ve excluded levels 18 and 19 to eliminate the pool of BG twinks parked there, who no doubt are played often) is clearly higher still (20% haven’t been played for 2 months) which suggests that there is indeed a significant pool of baby toons that are never going to grow up to see Naxx…