Mechanics: Attack Speed

Attack Speed is an intuitive concept governed by unintuitive, arcane calculations and mechanisms. Everyone can understand that more attack speed means attacking faster, but the exact nature of the math can induce headaches.

To make this easier, we'll start with an oversimplified formula.

Attack Speed = Base Attack Speed * (1 + Attack Speed Bonuses)

Every champion has their own Base Attack Speed. These range from 0.579 to 0.694 attacks per second depending on the champion. Because of how the formula works this value is very important to a champion's ability to scale with Attack Speed. To reach 2.0 attacks per second a champion with 0.694 Base Attack Speed only needs 188% Attack Speed, whereas a champion with 0.579 Base Attack Speed needs 245%, a 57% gap worth roughly 1500 gold. A higher Base Attack Speed clearly results in substantially greater benefits from Attack Speed bonuses.

Champions also gain Attack Speed as they level, similar to how they gain AD. These range from 0.98% per level to 3.84% per level, including level 1. This Attack Speed functions the same as any other Attack Speed Bonuses.

So a level 18 Shaco with Madred's Bloodrazor and Berserker's Greaves has 65% Attack Speed from items, 54% from levels and a Base Attack Speed of 0.694 attacks per second. Attack Speed Bonuses are added together, so filling out the the formula we get:

Attack Speed = 0.694 * (1 + 119%)
Attack Speed = 0.694 * 2.19
Attack Speed = 1.520

If we replace Shaco with Nautilus, we only have 17.6% Attack Speed from levels, and a Base Attack Speed of 0.613, giving us:

Attack Speed = 0.613 * (1 + 82.6%)
Attack Speed = 0.613 * 1.826
Attack Speed = 1.119

As you can see, both Base Attack Speed and a champion's innate Attack Speed per level can have a huge influence on their scaling.

However, there's more to Attack Speed than that. The oversimplified formula ignores the effect of Attack Speed Debuffs, such as Frozen Heart or Malphite's Ground Slam. Factoring those in the equation changes.

Attack Speed = (Base Attack Speed * (1 + Sum of Attack Speed Bonuses)) * (Product of Attack Speed Debuffs)

Because of how this formula works Attack Speed Debuffs are incredibly powerful. Consider what happens to our Shaco when fighting an opponent with Frozen Heart's 20% Attack Speed Debuff.

Attack Speed = (0.694 * (1 + 119%)) * (1 - 20%)
Attack Speed = (0.694 * 2.19) * (0.8)
Attack Speed = 1.520 * 0.8
Attack Speed = 1.216

Shaco's final Attack Speed is reduced to a level equivalent to having only a 75.2% total Attack Speed Bonus. Because the 20% Debuff affects all attack speed as a multiplier it's as if Shaco lost 43.8% Attack Speed. The more Attack Speed a champion has, the greater this effective loss becomes.

Multiple Attack Speed Debuffs are multiplied together, so if our Shaco was affected by both Frozen Heart and Ground Slam the calculations are as follows:

Attack Speed = (0.694 * (1 + 119%)) * ((1 - 0.2) * (1 - 0.5))
Attack Speed = (0.694 * 2.19) * (0.8 * 0.5)
Attack Speed = 1.520 * 0.4
Attack Speed = 0.608

In this case our unfortunate Shaco is actually attacking slower than his Base Attack Speed. He effectively has -12.4% Attack Speed, making such an auto-attack heavy champion essentially worthless. This crippling combination is a key part of why Malphite has been a top ban.

As a matter of trivia, Attack Speed has both a cap and a minimum. You can't exceed 2.5 attacks per second, or go below 0.2 attacks per second. The circumstances for approaching either number are extremely rare, but not entirely improbable. There is also one confusing exception to the rules for debuffs, the passive effect of Randuin's Omen. For whatever reason, Randuin's Omen's passive is treated as a negative Attack Speed Bonus rather than as an Attack Speed Debuff. You won't often encounter situations where these tidbits are practicable, but they're interesting nonetheless.

What is useful is applying the understanding of Attack Speed mechanics to champion theory. Not too long ago AD Kennen was used in tournaments in place of traditional AD carries, and a large part of his success came from his excellent Base Attack Speed and Attack Speed per level. Combining his 0.690 attacks per second with his 3.4% Attack Speed per level, Kennen has the highest innate Attack Speed in the game. Despite only having a nominal AD steroid, Kennen's superior base stats and scaling allowed him to be a serious contender at the tournament level.

Similarly, Twitch is currently flying under the radar. Of the traditional AD carries Twitch has the highest Base Attack Speed and Attack Speed per level. That he also has an extremely strong Attack Speed steroid only adds to his hidden strength. It's only a matter of time before someone figures out an appropriate team composition and Twitch rises from the depths of obscurity.

Understanding Attack Speed isn't quite as critical as other mechanics, but for the teams that want to be on the bleeding edge of composition and champion theory it's another important piece of the puzzle.

Patch Day: Late August

Today's patch notably comes without any new champion whatsoever, not even a note about requiring further testing on the PBE. Of course, this doesn't mean there's nothing to talk about.

Alistar is getting some much needed nerfs, and in the right places. He's been banned consistently for quite some time, largely because he's been the ideal jungler for the current meta. His ganks are amongst the strongest in the game, while his farm requirements are amongst the lowest. Nerfing his movement speed and his headbutt range cuts into his ganking power significantly, but he'll still be a very strong jungler regardless.

Corki gets a thorough smacking with the nerf bat. The modification to his passive is the most notable of his changes, as it reduces his auto-attack damage significantly. Due to the mathematics of the true passive it's roughly a 10% auto-attack damage nerf late game. That's a very strong nerf, and will hurt Corki significantly.

Diana's nerfs are reasonable, and she'll likely still be incredibly strong afterward. Her base damage and innate tankiness remain very high. The key change is the cooldown nerf to her ultimate, which forces Diana to be much more thoughtful about its use. She is now, as she should have been in the first place, very reliant on her Q for sticking power.

The Eve changes will significantly help both her laning and her jungling, but she remains an odd champion. She's much more at home in a gank composition meta, which no one is currently running. Should that become common she'll see a resurgence of popularity, but until then she's not likely to see the limelight.

Ezreal is the least-nerfed of the three popular ADs. What made Ezreal's W strong wasn't its damage as much as its attack speed slow. Landing a W in a skirmish bottom meant winning the skirmish, as your opponent's damage potential plummeted. This hasn't changed, and buffing Q only makes Ezreal stronger. Expect him to utterly dominate whenever played by someone who can aim.

Gragas' ultimate has long been known for its ridiculously short cooldown. With cooldown reduction he's been able to throw potential game winning barrels every 36 seconds for some time. The nerf is warranted, but saddening for long time Gragas players. The mobility nerf is notable but less important, as ultimately the mana cost won't matter.

Graves' changes can be hard to wrap one's mind around due to the arcane and mysterious nature of Attack Speed. However, the end result of the changes is that with a typical build Graves will be auto-attacking 10% slower than before. This is roughly equivalent to the Corki nerf, though via a different vector.

Nerfing Karthas' wall hurts his laning phase more than anything else. It will still be great for cutting off fleeing enemies, or countering an initiation.

Katarina's damage scaling has been nerfed significantly, likely as a counter-weight to her passive's change. Previously getting a reset on her ultimate required near-maximum CDR and three dead enemies, meaning it was pretty common for her to get her ultimate back with only one foe left to attack. With this change 2-3 deaths will ensure a reset, warranting a slight nerf on her other damage sources.

It surprised me how little Twitch is played. These buffs are very slight, and likely are aimed at putting him on the radar rather more than anything else. I expect that eventually people will learn the power of a Twitch kill lane, and we'll see him become extremely popular.

This was a nerf-heavy patch, but they were all warranted. We'll see if Syndra's patch doesn't bring with it a greater proportion of buffs for needy champions.

Jungling: Super Fast Ganks

I am remiss for failing to discuss my recent video on super fast ganks. There are a number of important points to be made about the ganking path explored in the video and how it shapes your jungling.

What makes this route compelling is fairly obvious. It's a double-buff, level 3 gank which comes so absurdly fast that even solo lanes may still be level 1. In the video Ashe Bot is only level 2 because she mindlessly pushed through the first two creep waves. The advantage of having both buffs and all of your abilities during a gank, which comes right as most lanes look to start skirmishing, is monumental.

This route fills the time gap between the common level 2 and level 4 ganks nicely, and combines many of the best features of both. Unlike a level 2 gank both of your buffs have been secured, so the cost of being counter-jungled is minimal. Unlike level 4 ganks your targets often haven't had time to invest in their escape abilities. While not flawless, it's a powerful tool to place in a jungler's arsenal.

The use of Karthas, a fairly controversial jungler, was an intentional choice designed to highlight the universality of this route. You don't have to play Nunu to use this route, or for it to be so ridiculously fast. Any jungler given a sufficient leash to save Smite for the second buff camp can follow this route. The only requirement is a little extra help from your team.

When I employ this route my goal is typically to snowball a sidelane. While you can gank middle with this route, the short lane ensures an earlier level 2 and easier escapes; if my goal was to snowball mid I'd rather do a faster level 2 gank. Whether the gank succeeds or fails you'll still have nearby camps you can clear before making a second pass, making this setup ideal for camping a lane.

The biggest threat to this route is a Smite steal at your first buff. This was actually employed by SK during the EU Regionals to shut down CLG's jungler. On top of the Smite steal they sent their team to secure CLG's other buff as well. While you're not likely to encounter a counter-jungling strategy of that level in solo queue, it's a danger to consider.

Super fast ganking is an important tool to have in your arsenal as a jungler. Practice it both with help and without, and then use it demolish unsuspecting lanes. While professional teams have been employing this method for months now, people still don't seem to expect it in solo queue.

Quote: On the Proving Grounds

"That's one small map for ARAM, one giant sink for our time."

-Neil Armstrong

Thought: Competitive Ethics

The finals between Curse and Dignitas this weekend have created quite a stir. There are a wide range of opinions flying around, and mine would only add to the chaos. Instead, I'd like to try to throw some facts and objective analysis into the mix.

The first and most important fact is that while this is League of Legends' first major match-fixing incident, it is not the first (nor will it likely be the last) case of match-fixing in eSports. StarCraft, StarCraft 2, and even Counter-Strike have all had match-fixing scandals in the past. Match-fixing is as problem for all eSports, not League of Legends specifically.

Second, match-fixing is damaging to any sport, but particularly eSports. Currently eSports are attempting to gain mainstream traction. Whenever an incident of this nature occurs it undermines the credibility of eSports and the organizations associated with them. We are still in the crucial developmental phase of eSports, and scandals like this roll back the efforts of thousands of people.

Finally, League of Legends is currently in an especially sensitive position as an eSport. It currently dwarfs all other eSports in terms of viewership, and in many ways is doing for eSports in North America what everyone hoped StarCraft 2 would do. It's also taken off like wildfire in Korea, the home of eSports. League of Legends is, in many ways, the eSports leader worldwide, which means anything adverse that happens in its competitions can have far ranging impacts.

So while it's unfair to ignore the history of match-fixing in eSports when considering the events of this weekend, League of Legends is naturally under greater scrutiny than other games due to its privileged position and the tenuous situation of eSports. Despite the small scope of this incident it is nonetheless a big deal because of the context.

In the long run I don't believe this particular case will do more than provide fodder for meaningless "DotA vs LoL" or "LoL vs SC2" discussions, but had this been a higher profile event or involved throwing matches rather than simple prize-splitting and ARAMs it could have been much, much worse.

Prognostication: Recap

It's been three months since I made my predictions regarding junglers, so it's time to see how well I did. The answer, not so well.

Amumu: While he did become much more popular, it was nowhere near to the extent of the now permanently banned Alistar and Malphite. If they weren't banned, he might not have become popular at all.

Hecarim: My prediction was completely off. While the community now recognizes Hecarim as a good jungler, Skarner has reentered the scene with a vengeance. Skarner is in the top 25 most played champions, while Hecarim barely makes the top 80.

Shen and Kayle: While I'm technically correct that they didn't become dominant forces in the jungle, it's a heck of lot easier to predict what won't become popular than what will. I award myself no points for these.

Sejuani: Riot hasn't given her any buffs my prediction here was fairly meaningless. She has seen a slight uptick in use, but nothing else. I maintain that if she's buffed she'll easily become a top pick.

In the end, my predictions were mostly failures. I will probably take another stab at it in the near future, but not until after I've nursed my poor ego's bruises and wounds.

Mechanics: Penetration and Reduction

Penetration and Reduction, whether Armor or Magic, are important stats for any champion looking to do significant damage as a game continues forward. They can also be fairly befuddling.

The first source of confusion is the separation of the stats themselves. This is the unfortunate, but necessary, result of game mechanics. Reduction is a debuff, granting its benefit to anyone attacking the affected target or targets. Penetration is a personal stat that benefits only the holder. Because Penetration does not benefit the team it can be balanced to give a greater personal benefit, whereas Reduction is usually found only on expensive items and support champions. Without this separation either some support champions would lose a critical benefit or carries would see diminished effectiveness from several key items.

As a further complication both stats can appear in two forms, flat values and percentages. In effect, Penetration and Reduction are actually four stats in all. Because of these two forms the order in which the stats are applied becomes highly relevant, and has actually changed over League of Legend's history.

In the earliest days of the game these stats were applied in the following order:

% Reduction
Flat Reduction
% Penetration
Flat Penetration

This, however, lead to a number of problems. Because % Penetration was applied before Flat Penetration the combination of the two was extremely potent. Building Armor or Magic Resist felt largely useless because the vast majority of it would be meaningless. Between Last Whisper and Armor Penetration runes and masteries a tank's 200 Armor would shrink to 85, and a non-tank's Armor would often be completely irrelevant. Because of these issues Riot altered the order.

The new order became:

% Reduction
Flat Reduction
Flat Penetration
% Penetration

In addition, both Last Whisper and Armor Penetration runes received nerfs. The net result was that our hypothetical 200 Armor tank would still have 104 Armor after penetration, and a non-tank would actually have a significant chunk of armor remaining.

At that point % Reduction was only found on Nidalee's Bushwhack, so its place in the order of operations wasn't particularly problematic. However, by the time of the second Kayle rework a large number of champions were now sporting % Reduction effects. Wukong, Jarvan, Renekton, Trundle and now Kayle all had them. This was a potential disaster, as not only was Flat Penetration applied after % Reduction, but so too was Flat Reduction. Riot wisely headed off the problem by changing the order again, and that change is still in place today.

The order as it is now:

Flat Reduction
% Reduction
Flat Penetration
% Penetration

Since the release of Wukong very few other champions have been given % Reduction, and with good reason. The combination of % Reduction and Flat Penetration remains very strong, though not abusively so.

So after all that history, what are the practical applications?

First, and most obvious, if you're one of the 10 champions with % Reduction then Flat Penetration can be very valuable. How valuable depends on the strength of your % Reduction ability and its ease of use. Wukong and Jarvan love Flat Penetration because of their simple yet strong % Reduction Abilities. Jayce and Shyvana, on the other hand, don't really care as their abilities aren't as effective or easy to use.

Less obvious is when to stack Flat Penetration on champions without % Reduction. Few AD carries stack Flat Penetration because it is hard to come by and delays their vital items. Bruisers sometimes grab Brutalizer if they're ahead in an attempt to press their advantage. AP carries, however, will often attempt to exploit a strong mid-game through a combination of Sorcerer's Shoes, Haunting Guise, and Abyssal Scepter, even if they aren't already fed. It's a tough call that depends a lot on a champion's role, skill set, and the game's circumstances. The most important rule to keep in mind is that Flat Penetration, while less effective late-game, is nonetheless effective; unless your item slots are all full it's better to build other items rather than foolishly sell your Flat Penetration items.

You can also build team compositions around stacking Reduction effects. This was once popular, but became rare after dragon and Baron became immune to these effects. It used to be that a team with Nidalee and Trundle could tear through Baron in the blink of an eye. This team composition was nerfed in order to make Baron less than the be all and end all of the late-game. It's still effective against tanks and their teams, but not so effective that anyone has found a compelling reason to revive such team compositions.

Hopefully Penetration and Reduction have become less mystifying, or at least more interesting, as a result of this post. It was far more history-heavy than math-heavy, but hopefully my chronicle of the long forgotten past was not as dull as some history lessons can be.

Mechanics: Cooldown Reduction

Cooldown Reduction (CDR) is an odd mechanic. Most stats are less effective as they are stacked, or at least cause other stats to become better investments past certain points. CDR alone becomes more effective the more of it you have.

This can be unintuitive initially, especially because of how information regarding CDR is presented in game. We see the results of CDR in the numerical values of our cooldowns, which follow this equation:

Cooldown = Base Cooldown * ( 1 - CDR% )

To the mathematically inclined this is obviously a linear equation. Each additional % of CDR will reduce the cooldown by the same amount as the last. A 10 second cooldown is reduced by 0.1 seconds per % of CDR whether you're at 0% or 39% CDR. At face value CDR appears to have linear worth.

But the value of CDR isn't determined by how much we reduce the cooldown, but by how often we can use the ability. That equation looks like this:


# of Ability Uses = Time Frame / ( Base Cooldown * ( 1 - CDR% ) )

This equation isn't linear at all. Even though the cooldown decreases linearly, as a divisor its effect on the equation is anything but. Consider a graph of the equation using our hypothetical 10 second cooldown:



Looking at the graph it becomes self-evident why Riot capped CDR at 40%. Past that point the increase in value accelerates to game-breaking proportions.

Even though CDR's effectiveness at 40% isn't as dramatic as at 100%, it's still incredibly valuable. Many champions hit a threshold at or near 40% CDR where their crowd control or other abilities chain together in a way that isn't possible otherwise.

For example, Olaf's Axe Throw has an 8 second cooldown which is reduced by 4.5 seconds if you pick up the axe. That leaves 3.5 seconds of downtime between axes. However, with 40% CDR the cooldown is only 4.8 seconds, leaving a mere 0.3 seconds between axes. At that point Olaf can perpetually throw axes as fast as his animation allows, provided he instantly picks them back up. Many Olafs abuse CDR for super-fast dragon kills and for decimating targets in their immediate vicinity.

Crowd control also increases in effectiveness dramatically with CDR. At maximum rank Rammus' Taunt lasts for 3 seconds with an 12 second cooldown, leaving 9 seconds of freedom for an opponent. With max CDR the cooldown is only 7.2 seconds, cutting that freedom by more than half to 4.2 seconds. Some crowd control effects, such as Zilean's Time Warp, practically become permanent with CDR.

Not all champions can or should aim for maximum CDR, but as a stat CDR is perhaps one of the most undervalued.

Jungler: Rengar

Rengar is now available. While I suspect he's better suited to top lane, he's a very capable jungler.

Rengar's passive makes him extremely effective at ganking side lanes or ambushing opponents in the jungle. The prevalence of brush in these places essentially gives him free, infinite gap closers on any targets that stray near to brush. Normally brush is a safe-haven for champions escaping a gank or through the jungle, but not against Rengar.

Rengar's overall kit is ideal for gank-heavy jungling. His auto-attack reset/nuke/attack speed buff gives him the damage needed to deal with buff camps and champions. His primary clearing tool is also a defensive steroid and sustain source, giving him built-in tankiness. That tankiness allows him to build damage/utility items early on without being a liability in team fights. He also has a ranged slow/root and, after level 6, the ability to bypass wards. Rengar's exclusive snowball item, filled with utility delicious for jungling, is just icing on the cake.

What Rengar doesn't have is strong initiation. While he can suddenly appear and potentially slow/root someone, his lack of followup CC is crippling unless you're catching an opponent who's way out of position. If Rengar is jungling, you'll be looking for your initiation elsewhere unless he's very fed.

Still, he looks to be an extremely fun jungler, and Shaco-level in his potential to abuse thoughtless or careless opponents.

The rune page suggested in the spotlight is fine, but you'll want to skill QWE R>W>Q/E rather than maxing Q first. The defense and clearing power of W is simply superior to the meager returns from ranking up Q.

Jungling: Dragon Control

When it comes to dragon control the current meta stands in stark contrast to all that preceded it. For the majority of League of Legend's history the jungler could solo dragon fairly easily, demanding perpetual wards to keep dragon safe. It is only very recently that junglers stopped being considered a notable threat to dragon.

In the past junglers used to be able to solo dragon after their first clear, or even sooner in some special cases. As a result Riot increased the defenses of dragon, pushing out how quickly any jungler could manage this. Since then the jungle revamp nerfed Madred's and Wriggle's, making dragon even riskier. The jungle revamp also shifted the meta away from the bruiser junglers who excelled at taking dragon. Finally, warding became much more prevalent, especially with supports coming to lane with 3-4 wards. Combined, all these factors make solo dragon too risky for too little reward.

Today the norm is for dragon to be continually warded, but only taken when there's a significant advantage. Whether it's because the enemy jungler is ganking top, a nearby enemy has recalled, or an enemy was successfully ganked/killed, dragon is typically taken or attempted only when the enemy is limited to giving up or running headlong into a highly unfavorable situation. Attempting dragon at any other time is dangerous, similar to attacking an equal opponent under their tower.

Because of this the value of Wriggle's has declined greatly. I've stated in the past that Wriggle's is overrated, and it's even truer now. With dragon control being a team responsibility rather than a jungler-specific function it's simply unnecessary to build an item specifically for controlling dragon. Even on junglers who have potential synergy Wriggle's is no longer the best option.

Dragon control has fundamentally changed. Even though Wriggle's may see a resurgence in the future in double/triple Wriggle's compositions, dragon is no longer controlled by a jungler but by a team.

Jungler: Katarina

The most interesting aspect of jungle Katarina is how quickly she went from the darling of Saintvicious' imagination to the garbage heap. She is the best, most recent example of what happens when players get caught up in an idea without thoroughly testing it.

It's too soon to definitively call Katarina a terrible jungler, but she currently suffers from several debilitating problems. She is fragile to the point where she can't start boots without an extremely specific rune page, and even then she won't be in a position to gank after her first clear. Even Mundo and Shyvana, well known for often starting Cloth Armor, can start boots. Katarina also has no clear build, as AP items are expensive and largely out of the question for a jungler. Finally, she is reliant on her lanes to provide crowd control for ganks and team fights. Without solutions to these issues jungle Katarina is simply bad.

All of this would have been known by Saintvicious and the community if they had simply tried to jungle Katarina on the PBE. Yet, without having done so, many players were forecasting her coming preeminence. The combination of spammable movement speed boosts and a gap closers is certainly unique, but ganking is only one facet of jungling. Failing to cautiously consider the implications of Katarina's lack of sustain and verify her potential through tests lead to Saintvicious' disappointment and dramatic reversal.

It's a shame jungle Katarina hasn't worked out as her playstyle would have been fairly fun and unique. However, she serves as a useful example and reminder that theory and experience must be tempered by each other.

Update: Platinum

A couple of days ago I hit platinum for the first time in my LoL career. In many ways this isn't particularly special, thousands of people have done this before me, but it marks an important milestone on my path to improvement.

What makes this moment notable isn't Riot's arbitrary determination, but my own investment in my goals. Reaching platinum is akin to arriving at a landmark on a mountain hike. More than anything else it serves as both a reminder of how far I have come and how far I have yet to go.

Even as I continue forward I'm keenly aware of how many hurdles are before me. I've still played less than 150 total ranked games, meaning my practical experience at this level is limited. I am extremely specialized in jungling, and while I'm somewhat comfortable in most other roles it's painfully clear that I lack the intuition and knowledge necessary to excel, particularly at top lane. Even though I am a proud jungler I lack confidence in direct counter-jungling. There are many competitive jungle champions that I have not mastered. Though I have climbed high on the ladder, the chasms of knowledge left unfilled are daunting.

My next goal is to reach 2000 Elo, one which should be considerably more difficult. It's entirely possible that when I next play I will have a losing streak comparable to my present winning streak, dropping me from platinum back down to 1800 Elo. My ultimate goal is to reach the top 25, which requires nearly 2600 Elo right now and potentially more by the time I'm anywhere near that goal. Having reasonable milestones in between grand delusions is important.

Thanks to everyone who has read my blog and watched my stream so far. When I eventually become super popular and famous and have 10000 readers/viewers you can talk about how you knew about Montegomery before he was cool.

Patch Day: Rengar/Mid August

It's patch day once again, and that means analysis. Unfortunately there's not whole lot to analyze this time around.

Rengar will be coming out later after Riot finishes balancing him on the PBE. He's got the chops to be a great jungler or top lane, but we'll see how he does in his final incarnation.

Garen received huge changes in an attempt to make him a competitive top lane. These changes are something of a wash for the comical Doran's Blade Garen build, but considerably increase Garen's strength later in the game if you aren't being silly. Because his passive is broken by jungle creeps he still isn't a viable jungler.

The Katarina remake is highly controversial as it appears to shift her playstyle from that of a burst assassin to more of an AP bruiser style. Due to all the changes Katarina can jungle, but she's very Shyvana-esque in that her first clear has sustain issues and she doesn't bring any crowd control to the table. Except to see her shift from mid to top lane, and only appear in the jungle occasionally.

Several champions have had AP ratios tweaked to make AP more desirable on them. The minor tweak to Kayle's W isn't going to dramatically alter her top lane play, but AP Teemo was definitively buffed. AP Varus will likely remain a joke, but the modifications to his ultimate will help make him a competitive AD carry.

Honestly the most important change this patch is that Katarina no longer has a semi-sentient octopus nesting on her head. If not for that and the changes to everyone's favorite Shakespearean couple, this would have been a boring patch.

Thought: Queue Sniping

Riot's new Accept/Decline system for queuing is almost welcome, though certain aspects of its implementation are wanting. That said, I may have found out why some people decline so abruptly and so frequently.

A little history is warranted. Duo-queuing used to average the Elos of the two players, but this ran into several problems. First, the inherent communication advantage between the two players wasn't accounted for. Second, if there was a large Elo gap between the players it was far easier for the better player to abuse the entire enemy team than for the entire enemy team to abuse the lesser player. As a result, Riot changed matchmaking to inflate the Elo of duo-queues to compensate.

As long as there have been ranked games there's also been the practice of "queue sniping". Upon seeing a friend queue for a game, a player also enters the queue hoping to be matched with or against the player. The lower your Elo the less likely you are to be matched together, but it's fun to try. Ultimately it's never been a problem because dodging cost Elo (though it doesn't anymore) and also time, and has remained simply a fun thing to do between friends.

The new Accept/Decline system, designed to weed out players who have gone afk, can be used to circumvent the duo-queue handicap via queue sniping. All you need are two or more players with similar Elo and a form of voice communication. Simply queue at the same time as your friends, vocally announce when your queue pops, and accept or decline depending on whether your friends' queues pop as well. This doesn't guarantee you'll be on the same team, but you have an even chance.

The problem is that players now have some measure of control over their queue; it becomes possible to game the system. Consider three friends queuing together who only accept a match if all three pop the same queue. Half their games will have all three friends together, and the other half will have one friend against the other two. In the former case the three have a huge advantage over their opponents in coordination and thus have an inflated chance of victory. In the latter case it's incredibly simple for the one friend to feed information to the other two (e.g. jungler/roamer positions) and clandestinely throw the game. In any case it ultimately only requires a little patience to gain an unfair advantage and inflate one's Elo.

Obviously this isn't good sportsmanship, but in the anonymous cesspool of the internet a great many people simply don't care. The allure of near-guaranteed Elo gains is bound to attract anyone who has either figured this out themselves or heard from somebody else.

The last I was on the PBE Riot had removed the Accept/Decline system, but I don't know if that's an intentional change we'll see in the next patch or simply a quirk of that environment. That seems an extreme solution when Riot could simply impose penalties on declining games, but in either case something should be done

Quote: On CS

"I find your lack of CS disturbing."

-Darth Vader, Star Wars IV: A New Hope

Thought: Mana Regen Itemization

Mana regen itemization is the most worthless in the game. Aside from the barest minimum required to lane or jungle effectively, mana regen is the worst stat on which to spend your hard-earned gold. The cost of mana regen is simply too high for what it actually accomplishes.

Because of LoL's mechanics and practical reality it is almost always better to invest in more damage than to invest in mana regen. Killing an opponent outright is more valuable than having the mana to kill them on the second or third try. This is why mana regen is most commonly built on AP carries who aren't burst-oriented, they can't instagib their opponents anyway. Even then, damage is so critical that despite the potential for Chalice to yield a whopping 35.8 mp5 by itself (approximately 2000g worth of stats) it was very rarely purchased even by attrition champions. Only recently has Chalice become popular, and primarily because it is now the component of a damage item. Mana regen is simply too costly for too little benefit.

Furthermore, mana regen itemization is overly-focused on AP champions and supports. AD champions and bruisers are essentially limited to Manamune, Philo/Reverie, Soul Shroud, Tiamat and perhaps Chalice; two support items, a troll item, an item whose upgrade is for AP champions, and one item actually tailored to the needs of an AD. In short, if you're an AD carry or bruiser you have few, if any, meaningful mana regen options.

It's for these reasons that Riot has been moving away from significant mana costs on AD carries and bruisers. Without building any mana regen at all most recent champion releases for these roles only care about mana in the earliest stages of the game. By the mid or late-game they hardly care unless they completely ignore managing their mana. There simply isn't the itemization to support mana-intensive kits for these champions, and this is the easy solution.

It's not, in my opinion, the "right" solution, but I understand why Riot chose this direction. Giving bruisers meaningful mana costs would dramatically unbalance the game. As it stands bruisers have significant issues becoming tanky enough to survive teamfights while still having enough damage to be relevant. Forcing them to invest in mana regen without careful reblancing would essentially eliminate all but the manaless bruisers as competitive options. Thus, fixing mana regen the "right" way is an incredibly complicated, tricky, and perilous operation. Rather than risk collateral damage Riot went with the simple option.

Still, that leaves us with mana regen itemization in its current, dilapidated state. Athene's Unholy Grail gives me hope that Riot will slowly, and carefully, tweak mana regen to be a more meaningful and investment-worthy stat (at least for APs), but until then it will remain an uninteresting, bland aspect of itemization.

Quote: On Brush

"If you can't see Garen out of brush, then you can see how Garen can be in brush."

-Dr. Seuss

Jungler: Diana

Having had the opportunity to jungle Diana I can declare, with the confidence and certainty of experience, that there is absolutely no reason to jungle Akali unless Diana is banned. As junglers, and arguably as champions, Diana is in every way superior to Akali.

Diana's superiority is not rooted in the ease of her jungling, but that is a notable attribute nonetheless. Starting with W at level 1 and maxing Q first, Diana clears quickly while taking little meaningful damage. After her first clear she practically auto-pilots through the jungle. She becomes so fast she could easily be played Udyr/Shyvana style, but that would be a waste of her considerable talents.

What makes Diana so much better than Akali is superior damage. Diana, without items, does more damage with her passive and W than Akali does with her passive and E (which is often unusable in teamfights and chases). On top of that Diana's Q and R are roughly equivalent in damage to Akali's. Diana simply does more damage with fewer items.

Essentially, Diana can afford to ride her high base damage into the mid-game and primarily build defensive items. For a champion based around jumping into the thick of things and going toe to toe with the enemy team's primary damage dealers, this is an incredibly important trait. You might try the same on Akali, but to significantly less effect.

Overall Diana is a very strong jungler. Possibly the strongest that's ever been released. I'm fairly certain she'll be nerfed significantly in the near future, at which point Akali may obtain a competitive edge. For now, Diana is simply the best at what she does without contest.

Jungling: Mid Lane Matters

The most important lane in the game for a jungler is the mid lane. All lanes are to varying degrees important, but the mid lane is critical to maintaining jungle control. If the mid lane loses, especially if it's a decisive loss, a jungler's life can become utterly miserable.

The importance of the mid lane bears similarities to the importance of the four center tiles in Chess. The central location of these tiles gives them power because they provide access to many parts of the board, which is why much of Chess involves fighting over these locations or finding clever ways around them. The mid lane in League of Legends is very much the same, it has the easiest access to every objective on the map.

As a result the mid lane has a dramatic influence on how well a jungler can control the jungle on both sides of the map. A weak mid lane can't contest or support invasions unless their opponent is inattentive, as otherwise they risk being intercepted and blown away by their laning opponent. For similar reasons a strong mid lane can support or be the instigator of invasions. A decisive loss mid lane opens up the entire jungle to the enemy, and can quickly collapse the other lanes as well.

This doesn't happen to the same extent with side lanes simply because of distances. The distance from top or bottom lane to any nearby buff camp is greater than the distance from the mid lane, and can be covered by a single ward. It's only when those lanes snowball to the point where they fearlessly push down a turret and past river that they become significant jungle threats.

Maintaining control of the mid lane is the key to maintaining jungle control. Unless your mid lane is a complete loss, it can be in your self-interest to help them get back on their feet. Any jungle farm you might lose by helping them might be lost anyway if your jungle becomes porous.

Update: More Glitches, Streaming, and Season 3

Unfortunately I was overly optimistic about my Sony Vegas being fixed. I've now been going back and forth with customer support for a couple of weeks because rendering isn't working. Once that's fixed I should finally have some new videos to post.

Streaming should pick up again now that I'm no longer on vacation and occupied. I won't necessarily be providing insightful commentary all the time because...

...Season 3 has been announced. This now means there's limited time (though how limited isn't clear) before Season 2 ends. I'd like to hit platinum before then, even though Riot hasn't announced any rewards for doing so. That means many more ranked games, a lot of learning, focus, and effort. I can't guarantee I'll personally be interesting all the time while playing.

We'll see if I can hit platinum (and stay there).

Quote: On Low Elo

"The path of the skilled player is beset on all sides by the profanities of the ragers and the inanity of low Elo."

-Jules, Pulp Fiction

Thought: What Happens to Jungle Akali?

Over two years ago I started jungling Akali, and after a few months was confident enough to post my first guide. It was laughable, viable only in games with players as bad as I was, but as I grew and improved at the game my guide grew with me. Its current iteration is, in my conceited opinion, the definitive work on jungle Akali. The gravity of the situation should therefore be self-evident when I state that Diana, in all probability, makes jungle Akali obsolete.

I've long held that, given a different meta, jungle Akali would be a competitive option. For the past year the meta for junglers has changed several times, from early-gankers to fast-clearing farmers to tanky-cc. Akali, while an adept jungler, has never matched any of those paradigms and thus has remained in obscurity. Despite this it was still possible that arranged teams might break from the standard meta and use her. That possibility largely vanished with the introduction of Diana.

The problem is that Diana is essentially jungle Akali, only better. The draw of jungle Akali was largely her uniqueness as an AP jungler and assassin, in particular the power of her post-level-six ganks due to her extremely long range, gap-closing ultimate. While their ultimates are practically identical for the purposes of ganking Diana also can pull enemies to her and slow them, a form of crowd control many times more powerful than the slow on Akali's Twilight Shroud. Diana's ganks both before and after level six are significantly stronger.

In addition many of Akali's minor flaws are not shared with Diana. Akali is largely single target damage and her AoE ability is awkward when chasing down opponents. Diana is almost entirely AoE damage, ideal for jungling, and her AoE abilities have an integral role in how she chases. Diana also can reveal foes during ganks, making the brush in side lanes less of a safe haven. For all her viability as a jungler, Akali's skills are designed around laning more than jungling.

Akali does have a few notable advantages over Diana. Akali has built-in sustain, allowing her to enter the jungle and leave with significantly more health than she entered with, while Diana only has a shield. Akali also has a stealth ability, and despite being countered by Oracle's it has its advantages. Finally, Akali is mana-independent, meaning she doesn't have to itemize around mana or suffer when passing blue buff to an ally. These benefits don't make Akali better at jungling than Diana overall, but they are places where Akali is superior to her.

Those small advantages are possibly the saving grace for jungle Akali. Because many players will be trying out Diana in the jungle, some will possibly note the similarities between Diana and Akali and begin to consider Akali's viability. Perhaps jungle Akali can ride on Diana's coattails and break into the community mindscape. Someone cleverer than I might even demonstrate a niche for her that makes her more desirable than Diana. It might be wishful thinking, but it's not impossible.

For now Diana is, barring some critical flaw yet to be discovered, superior to jungle Akali. This won't stop me from keeping my guide updated or my skills sharp, but it's unlikely that my team will field my favorite ninja as a jungler anytime soon.

Thought: Chalice versus Double Doran's

By far the most popular opening for a mana-based AP is double Doran's Rings. This is not without reason, it gives a player the mana regen needed to hold a lane, the health needed to survive a gank, and the AP needed to clear creep waves or burst down champions. However, Chalice of Harmony has been slowly chipping away at that monopoly.

Chalice has been increasing in popularity for a few reasons. The mana regen it provides is greater than that of two Doran's Rings in any circumstance where you actually care about your regen. The magic resist is better survivability against another AP's burst at 800 health, which most champions pass around level 6 (coincidentally around when you really need to worry about burst). It even costs slightly less and eventually builds into the powerful Unholy Grail. The primary, and arguably only, drawback is the lack of AP. Chalice is simply a very powerful item, and super effective in the lane most apt to buy it.

What has been holding back Chalice, besides simple inertia, is the lack of AP. 30 AP can be 100 damage or more on a level 6 burst combo, often the difference between a kill and a near miss. It also gets amplified by Deathcap, whereas Chalice has no stats which are. Chalice can therefore feel weak offensively, especially if you're used to double Doran's.

Still, Chalice is simply powerful on any champion with above-average mana issues. There are many who, even with Blue buff, can quickly exhaust their mana. At 25% mana with Blue, Chalice is easily a whopping 45 mp5 or more. Imagine being able to have two Blue buffs at once, and you have a pretty good picture of what Chalice does.

On all but the most mana-independent AP champions Chalice is simply better than double Doran's. There's so much to gain and so little to lose. For an item once considered one of the worst in the game, Chalice has certainly come a long way.

Patch Day: Early August/Diana

Today's patch comes with Diana to follow later. If the Zyra patch was any indication Diana will be released sometime this weekend or early next week. In the meantime, there are patch changes to analyze.

Amumu's changes dramatically reduce his damage when he builds AP. It was fairly common for Amumu to pick up at least one AP item when jungling, such as Abyssal, so this does have a significant effect. The mana cost reduction on Tantrum honestly isn't a big deal, hardly sufficient compensation for dropping 0.5 in AP ratios. However, it might be enough for Amumu to forgo Philosopher's stone and rush other items instead.

The change to Brand's ult should greatly reduce its randomness. Sometimes Brand would dominate due to lucky bounces, and sometimes he'd feel like he had no ult at all. I'd posit we'll see more Brands, but then the substantial Malzahar buffs didn't exactly bring him back to the forefront.

Cho'gath's changes are, like Brand and Malzahar before him, quality of life changes first and foremost. Cho'gath should be somewhat easier to use now that his delays and cast times are streamlined. Again, I'd guess we'd see more Cho'gaths, but quality of life changes are no guarantee of a lasting increase in popularity.

Continuing the theme of quality of life changes we have Eve. The shorter stealth fade time is fairly important, as it was a ridiculously long time before. Similarly her ult's cooldown was extremely long before, they literally shaved 30 seconds off it at rank 1. With these changes her play should feel smoother, particularly relating to stealth.

Finally we have Hecarim to round up the quality of life changes. E was previously fairly unresponsive, with its pushback going at odd angles and freezing Hecarim thereafter. If this is fixed it will make his ganks substantially easier, along with his greater initial velocity. Similarly faster fleeing targets will make his ultimate feel much stronger when placed well. Overall he should be far easier to use for the many players who have felt he was clunky and difficult.

Most of what remains are minor bug fixes and nerfs, but it's worth noting that Janna, Xin, and Yorick were nerfed due to overly strong laning phases. While small, their nerfs should make them easier to deal with. Also, the reversal of Urgot's nerf should make him competitive again (he had, as I suggested he would, fallen completely off the competitive ladder).

Overall this patch largely has quality of life changes, a few nerfs, and a few minor buffs. After the plethora of buffs and remakes last patch this is a welcome reprieve. I'll talk more about Diana later this week.