Analysis of the Post-Hotfix 5.2 Battleground Scaling Changes

Cynwise, Twin Peaks FC

The hotfix applied on March 12th addressed the most obvious bugs present in the battleground scaling changes of patch 5.2.  More testing will be required, of course, but I think we can say that we’re looking at the final, intended version of the scaling.

Battlegrounds for characters under level 90 in 5.2 have been changed so that a character’s effective level is scaled up to the top of the bracket. A level 10 character becomes 14, a 37 becomes 39, a 69 stays 69. For characters who are not at the maximum level of a bracket, this has the following direct effects:

  • Health pools are increased and normalized.
  • Primary attributes are increased slightly.
  • Heirloom items scale to the new effective level.
  • Normal gear does not scale.

The goal of this change was to normalize Hit between characters at opposite ends of the bracket and help balance low-level PvP. This change takes steps in the right direction for both goals, but there are a lot of issues that undermine that forward progress.

Let’s take a look.

THE QUESTION OF HIT AND SECONDARY STATS

WoW uses a system where you gain combat values as you level, but their relative effectiveness declines. You need more of a given stat to remain at the same power. I covered this in my previous post On the Broken Battleground Scaling of 5.2 as well as in an earlier post, The Challenge of Fixing Low Level PvP, so I’m going to assume you’re familiar with the concepts covered there. The point which is relevant to analyzing the 5.2 scaling change is that if you stay in the same gear as you level, it will do less for you. For primary stats (Strength, Agility, Intellect – or Attack Power and Spell Power), they increase linearly but decline relative to the character’s environment. For secondary stats (Haste, Hit, Crit, Dodge, PvP Resilience, etc.) they decay directly as more rating is required to achieve 1% of a given stat.

So the first problem that is introduced is that characters at the bottom of the bracket have their secondary stats reduced upon entering the battleground. A item with +8 Hit gives 4% Spell Hit at level 20, but only 3.2% Spell Hit at level 24. When a level 20 character zones into Warsong Gulch, they will need to stack more Hit, Haste, Crit, etc. to be as effective as they were before. For secondary stats like Haste, Crit, Mastery Dodge, PvP Power and Resilience, this results in the character performing worse inside the battleground compared to outside of it. Casts are slower, regen is slower, Crit is lower.

Hit isn’t quite so cut and dry. Lower levels will need more Hit to achieve the same rating as before, but the amount of Hit required to cap is also lowered (from 10% to 5% for melee hit and 20% to 4% for spell hit) because everyone is now the same level. You will need more Hit at the lowest level to hit the at-level PvP Hit cap, but once you hit it it’s good against everyone in the bracket.

5.2 Scaling Analysis - Do Low Levels Need More to Hit High Levels?

The chart above is from a Hit Analysis spreadsheet to compare the hit ratings necessary at the tops and bottoms of each individual bracket. The values in red are those places where the lower end of the bracket needs more hit (a positive number) to hit cap against the top of the bracket under the new scaling system; green values are where they need less hit. (Less Hit is the goal, which is why it’s green.)

As you can see, there’s a uniformly positive impact on casters, and nearly all brackets have a positive impact for melee. That’s the good news. It looks weird and nonintuitive because your Hit definitely goes down, but it’s easier to cap your Hit against a level 79 now at 75 than it was before.

The problem is, do you want to cap your Hit against the top of the bracket? The scaling means that you’re slower, less keen, less able to defend yourself against attacks. If most of your opponents are also at the bottom of a bracket, you’re less effective against them than you were before. Take a look at the same chart, only this time comparing low levels hitting low levels:

5.2 Scaling Analysis - Do Low Levels Need more to Hit Low Levels?

Uniformly melee and casters are worse off trying to hit people of the same level if they’re in the bottom of a bracket. Why would the majority of people be in the bottom of a bracket?

I never thought you’d ask.

Cyn - Warblade and Vicious Pyrium Set - SW - procs

BRACKET SKEW AND GEAR AVAILABILITY

In the previous post I described brackets as either skewing high or low. Each bracket has a different combination of stat weighting, gear availability, and possible expansion overlap which gives one end or another unique strengths. High skew brackets are characterized by:

  • Better gear is available at the top end of the bracket than the bottom.
  • Rating decay is relatively flat across the bracket.

High skew brackets tend to avoid expansion breaks and are characteristic of most of the middle levels.

I’d consider the following high skew brackets: 15-19, 20-24, 25-29, 30-34, 35-39, 40-44, 45-49, 50-54, 55-59, 65-69, 75-79.

The 20-24 bracket is a little bit of an odd duck amongst the high-skew brackets because of the preponderance of Starter Edition players. The population skews low, but the bracket itself favors high.

Low skew brackets are different. They usually are:

  • The best gear is available at the bottom of the bracket.
  • Rating decay hits sharply across the bracket. This can happen either at the beginning (after an expansion break) or across the entire bracket (due to expansion level compression).

Low skew brackets are usually situated near expansion breaks due to the stat scaling and gear combinations afforded by endgame development. (A good reference to understanding how this works is Ghostcrawler’s blog post The Great Item Squish (Or Not) Of Pandaria, where he illustrates how each expansion introduces item inflation at the endgame.)

The following brackets skew low: 10-14, 60-64, 70-74, 85-89.

The 80-84 bracket used to skew low, but the introduction of Mists gear is starting to make it skew high despite the dramatic combat rating dropoff at 83. There are still classes which excel at 80, depending on gear availability, but there are sweet spots for gear at 81, 82 and even 84 which twinks pursue.

All clear on bracket skew? Okay, good, because the second issue the new scaling model introduces is that it skews all brackets high, eliminating any benefits of being low level. It’s not just that it turns the low skew brackets high – it’s that it skews all of them higher.

Compare a low skew bracket like 70-74 against a high skew bracket like 75-79. At level 70 you have access to Sunwell gear and Brutal PvP gear. At 71 you can have the same gear, but your secondary stats are 25% lower, and it goes downhill to 74. With one or two exceptions, the gear at 74 just isn’t any better than what you had at 70. With the new system, everyone PvPs at level 74 in level 70 gear, which means you can level and gain some abilities (and maybe those one or two pieces of gear) without any drawbacks. It’s not great – everyone is worse off than before – but it’s not terrible, either.

In a high-skew bracket like 75-79 the situation is much worse. Level 75 characters only have access to Wrath questing gear (about ilvl 154-160) – which is still about as good as the Sunwell/Brutal PvP epics you had in the previous bracket. Level 79 characters have access to Cataclysm greens (about ilvl 277-289). The gear disparity is magnified in this bracket: the low end (normally 75) is now 79 in level 75 gear, while the high end is unchanged!

Because high skew brackets have better gear at higher levels, lower level characters are further behind relatively under the new system. They’re locked out of the better gear (because of their level) and the gear they do have functions worse (because they’ve lost their scaling.) Low skew brackets are affected as well, but not to the extent that a high skew bracket is.

In this case, I think scaling has made brackets a little more unfriendly to the bottom. How unfriendly it has become depends very much on how skewed the bracket was before the change.

Generally, it’s now always optimal to level to the top of the bracket if possible.

Screen Shot 2013-03-08 at 9.00.32 PM

A BIAS TOWARDS HEIRLOOMS

In my post The Challenge of Fixing Low Level PvP, I didn’t think that level disparity was the real source of inequality in low level PvP. Ability distribution, player experience, and wild variance in gear levels still seem to me to be a bigger issue than levels and hit. Heirloom gear is where the scaling change directly meets my own concerns about gear disparity, since I think enchanted heirlooms are practically essential PvP gear for leveling at this point.

Heirlooms have created a new strata of player in the battleground ecosystem, leveling twinks.  Sometimes called heirloom twinks, these players rightly gear as best they can for PvP with stamina-heavy heirlooms that give them solidly good gear in nearly every slot as they go. Leveling twinks are different from XP-off twinks because they are in the same bracket with players in quest whites and greens, yet have dramatically better gear than their opposition. XP-off twinks are shunted off into their own battlegrounds and (generally) play against similarly well-geared opponents. As XP-off twinks aren’t leveling, they will often go to great lengths to secure the best in slot gear – lengths that you might not even pursue on an endgame character – because they know it will be good for the life of that toon. Leveling twinks might lock XP and run some dungeons to fill in non-heirloom slots, but generally they have blue-quality gear and good enchants.

The third issue is that heirlooms have increased value and power under the new scaling system. Because they scale to the top of a bracket with a character, they don’t suffer any stat decay. They’re as good at 36 as they are at 39, at 55 as they are at 59. Heirloom items have the stats they are supposed to have at the top of a bracket even if the character is at the bottom. Normal gear doesn’t do this – if a belt gives you +8 Intellect at 40, it will give you +8 Intellect at 44. A heirloom that gives you +8 Intellect at 40 will give you +11 Intellect at 44.

This seems like a small issue, but I don’t really think it is. New players are already at a disadvantage if they try out PvP on their first character. People like me bring heirlooms and enchants and gems and consumables which are really really difficult to acquire until you have high-level assistance and knowledge of the game. Heirlooms contribute to the imbalance of low level PvP, and now they’re more imbalanced. Not a lot – quick testing showed me each piece was like 5-10% better in terms of raw stats – but better.

This change makes sense in the confines of scaling, but not in terms of achieving PvP balance.

Cynwii - Loch Modan Bush Disguise

AN AMUSING SIDE EFFECT

One thing I noticed while testing the new system amused me more than it should. If you level out of a bracket – ding 40 in the 35-39, for instance – instead of being one level above your opponents, you’re now five. All of the Hit balancing goes out the window because everyone needs 11-25% Hit to get you now. Your heirloom gear scales up, too, so while you’re not quite as good as you can be (unless you carried gear with you to equip later) you’re not that badly off. The slight advantage people used to get for dinging in a battleground has been replaced by a very large one, and honestly it’s kinda funny.

I opened a ticket because I saw a level 69 Druid running around my level 60-64 Eye of the Storm this week. This is how that happened. (Addon nameplates are reporting the incorrect level. The stock UI should display correctly.)

Cynxi - Pandaren Rogue - Melee at Rest -SW

INCREASED HEALTH AND THE GOOD PART OF THESE CHANGES

There are some things about this change which I don’t like. It skews brackets high, it rewards heirloom gear in low levels, it makes Hit a PITA to manage, it destroys secondary stats like PvP Resilience and PvP Power. It makes my characters feel different (read that as worse) inside a battleground than outside them, and that’s awkward. Don’t like it.

But there are some good things, too, things that any analysis needs to acknowledge.

Combat seems slower now. Burst is lessened in many higher brackets because secondary stats like Haste, Mastery, and Crit have been slaughtered at the same time health pools have been increased. This effect is very pronounced at the higher levels where health pools have shot up 100k or more. Here’s how my warlock looks:

5.2 Scaling Analysis - Cynwise 85 Lock

(This is from my post-hotfix 5.2 Battleground Scaling Data spreadsheet if you are following along with the math.)

Without trying to stack any Resilience, her effective health is up 110k – almost 40%! If damage goes down and effective health goes up, fights are slower and more strategic, you don’t get 1-shot by an Eviscerate crit or Chaos Bolt, and – perhaps most importantly – healers are more potent. (Your opinion of this depends on if you’re playing a healer or facing a healer premade.)

These scaling changes make battlegrounds more healer-friendly. That’s not a bad thing, and not something I expected.

The changes in lower levels are not quite as dramatic, but they’re still present. Here’s my heirloom-clad Mistweaver monk:

5.2 Scaling Analysis - Cynwii 39 Monk

I mention the heirlooms because they fight against stat decay while providing level 44 stats in the second example, but in general you see that there’s a solid improvement in health with only a moderate decrease in secondary stats. (Secondary stats from 35-60 decay at much slower rate than those from 85-90.)

It’s a small boost to survivability, but it certainly does change the feel of some brackets.

PASSING JUDGEMENT

Now that we’re past the buggy part of the implementation – and make no mistake, it was buggy – it’s tough to get past it and move on to adjusting to how the new scaling system works. Now that I’ve had a chance to sit back, do some math, and more importantly think about it – many, if not most of my objections below are qualitative and personal.

  • Hit scaling is non-intuitive if you play at the bottom of a low-skew bracket. Hitting 4% Hit at 85 was pretty easy, but hitting 20% wasn’t ever a goal of mine. If you miss some of the 89s in there, you miss them and move on.
  • DPS specs feel less bursty, and I was enjoying playing a bursty class. The survivability tradeoff doesn’t feel all that great. (Counterpoint: my healer partner absolutely adores the loss of burst and feels like a god.)
  • The hit to secondary stats devalues PvP Resilience and PvP Power. Once you hit level 70, Resilience becomes a Thing You Should Have, but now it’s hard to argue for it in the face of increased health pools.
  • I feel clunky. Casts are slower and hit much softer than they did before. I step out of a BG and it’s like I’ve got a supersized-ICC buff on me! I zone in and things slow down. I notice this more at 85 than at 40 or 60.
  • I don’t like how some of the brackets have skewed even higher than before.
  • Having to choose between awesomeness in PvE (by staying at the end of an expansion) and PvP (by leveling to the top of a bracket) isn’t any fun. I like having my cake and eating it too!

It’s not bad. It’s not great, either, and I understand why people are upset and confused by it. But it’s not as bad as my initial reactions, and it’s not going to break PvP and cause the Battlegrounds to be unplayable.

From an IT perspective, it looks like it was a reasonably simple development change with a lot of downstream impacts. While I’m pleasantly surprised that it does achieve the goal of making it easier for the low end of brackets to hit the high end of the brackets, the tradeoffs in secondary stat reduction, elevated heirloom importance, and skewing bracket imbalance doesn’t seem like a positive change overall.

That said, the increased health pools and reduced burst have made my pocket healer very happy. And keeping my healer happy is important too.

Jury is still out. Let’s see how this change works for a bit.

About these ads

8 Comments

Filed under Cynwise's Warcraft Manual

8 responses to “Analysis of the Post-Hotfix 5.2 Battleground Scaling Changes

  1. I am a happy pocket healer right now! :D

    However, I found if I was in the wrong bracket with Heirloom players that actually knew what was what, I was in a world of hurt. (I was hurt, over and over).

    This is a great analysis, thanks for doing it!

  2. skinnemuva

    Has anything happened with the resilience stacking healers in the 70-74 xp off bracket?

  3. Bwappo

    Great work and thorough explanation, Cynwise. Thank you.

    For players who want to arena at 80 and 85, and for expansion-capped players, battleground stat scaling forces a choice between the two ends of their brackets. I’m not sure what to think about that, since 70s and endgame players don’t face that choice — they can have it all.

  4. Pingback: What to Expect in 5.2 (for Twinks) | Twinking.net

  5. Excellent work (but then your always amazing). I reported the heirloom issues as soon as I noticed it in pvp and was told their importance in pvp is ‘working as intended’ to paraphrase.

  6. Pingback: Suggested Reading: Isle of Giants, PVP and PVE, the Isle of Thunder, blind raiding, and more | Kor'kron 501st

  7. Pingback: On Rejoining the Leveling Brackets | Cynwise's Warcraft Manual