Trial by Ember

Our Donezo Manifesto

The Ember squad on a hike in Malibu, California

TL;DR

We were given the opportunity to pursue our passions. We tried to do something big and different. We were equipped with everything we needed to win. But we still fell short, so the original team at Ember disbanded last month. Don’t feel sorry for us, we got paid to live our dreams. When you fail in sports and esports, the public wants to know what happened. There has been rampant speculation since we haven’t said much at all. It’s time to set the record straight.

Challenger Series

The bottom line is that ascending to the League Championship Series (LCS) through the Challenger Series has turned from a low probability event to a very low probability event. We felt good about the odds last November when there was still auto-relegation/auto-promotion and one best-of-five match. Things changed this January when Riot Games announced the revamped Promotion/Relegation system. The new system has no auto-relegation and Challenger teams would need to win at least two best-of-five matches to get into the LCS. While this new system is a much-needed step for ensuring stability in the LCS, it makes running a Challenger team much less appealing because:

1. LCS teams have more experience with pressure. LCS teams play eighteen games live onstage during the regular season and each game is immensely more intense than the ten offstage games in the CS regular season. It’s difficult to replicate the conditions of a do-or-die, best-of-five series.
2. Players know this fact too. Therefore, the best talent would much rather go to an LCS team if given the opportunity. Cinderella teams like Origen, a group of LCS-caliber players who decided to start their own team, are the exception rather than the rule.

And we see this logic being manifested in recent results. In Europe, Splyce, ROCCAT, and Giants re-qualified for their spots over the two Challenger teams in the EU LCS. In North America, Renegades and Team Impulse re-qualified for their spots. Apex was the only challenger team out of four that ascended into the LCS.

Ultimately, it wasn’t my decision to not go for it again but I agree with the logic of the business decision. As for the Ember brand and the Challenger spot, that still belongs to the parent company, Team Elemental, Inc. I don’t know what they will do with those assets as I’ve resigned as CEO of Team Elemental.

The Ember Academy is no more. Our docu-series is partially complete (done shooting, just need post-production work) and our production partners at Finite Films are looking for funding to finish it (five episodes, 22 minutes per episode). You can reach out to me on Twitter or email them directly at info [at] finite-films.com if you are interested. Here’s the teaser and they also have one full episode complete:

The rest of this autopsy is broken down into two sections — first, my lessons learned throughout this journey, and second, an in-depth analysis of our game against TDK which ended our run. But before we get started…

Default to Gratitude

It’s easy to get angry and upset in a situation like this. You can always find someone who has something negative to say about someone or something at a failed startup or organization. None of that matters because at the end of the day, we still have our health and the love and support of our friends and families. In times like these, I like to default to gratitude.

• I’m grateful for the amazing group of great players who took a chance and played for Ember: Nick Haddad, Greyson Gilmer, Ben deMunck, Colin Earnest, Juan Garcia, William Chen, and briefly Lucas Larsen.
• I’m grateful for our fans who believed, and I hope, still believe in our philosophies. In the grand scheme of things, we made a dent — a small dent but a dent nonetheless — in the world of esports with our open salaries and our emphasis on player health.
• I’m grateful for our investor who gave us a shot at building our dreams.
• I’m grateful for our staff who helped build an amazing brand in such a short period of time.
• I’m very grateful that our Head Coach, Weldon Green, stayed on to guide our players mentally and emotionally through this ordeal. This man has character (more on this later).
• Last but not least, I’m grateful for Bao Lam, my co-founder. As the General Manager, Bao bore the brunt of our loss publicly and privately. He made some good calls and admittedly, some not so good calls. But I empowered him to make those decisions — and I signed off on every major decision, good or bad — so I take full responsibility for everything that happened.

Lessons Learned

A lot of this may seem like common sense, and it very may well be. However, I still find it worthwhile to write down for the sake of benefiting others or reminding myself in the future.

1. Investor alignment: It all started off with a handshake. Getting into business with someone is very much like getting married. You probably shouldn’t do it after the first date. But I did it anyway. Our investor was very successful by every definition of the word. He is a successful entrepreneur, operator, and investor. He taught me how to be laser-focused on the right win conditions. In many ways, he was the right investor for me but I was the wrong entrepreneur for him. My ambitions were not aligned with his objectives. He wanted A, but I wanted B and C as well.
2. Board composition and meetings: The timeless “two board seat” dilemma that faces first-time entrepreneurs. I agreed to giving two board seats to my investor, which meant I only got one. Although it never came to a direct board vote, I effectively had no control of the company if we came to a disagreement. And when I had the chance to add an independent director, I didn’t push for it fast enough and the deal fell through. As far as board meetings go, I’m definitely an advocate of formal meetings and minutes. It sucks when you need to revisit an old decision and it was partially decided via email and/or a phone call and two sides don’t agree on details.
3. Hiring: Every founder talks about it, but we did a terrible job at it. The main reason was not having on-site interviews because we wanted to save costs. We thought having a paid trial period would work, and it did — until it didn’t. It’s just hard to get a sense of leadership remotely. Our players would respect some coaching staff and analysts online, but once they came here in person, respect would be lost due to interpersonal skills that you can’t assess through Google Hangouts. Secondly, everyone seems to hate everyone in esports. When we asked for recommendations or opinions on hires, we received mixed feedback on every single candidate.
4. “Game” versus “Business”: Before the company formed, Bao and I decided to split ownership for game versus business decisions. I thought the biggest issues that would happen with the team would be game-related, but it rarely was. Instead, it was often interpersonal issues — a bunch of teenagers and young adults trying to work together towards a common goal. Ironically, that’s exactly what I did as an Army officer — get a bunch of teenagers and young adults to work together towards a common goal. Delegating this was a catastrophic mistake. I should have been way more involved in resolving the interpersonal issues of the team.
5. Character over reputation: Be more concerned with your character than your reputation, because your character is what you really are, while your reputation is merely what others think you are — Abraham Lincoln. The first thought that many people have when there’s a train wreck is how do I look good coming out of this. Thankfully, not everyone thinks this way. And certainly not Weldon Green. When he boarded his flight from Finland to Los Angeles, things were rosy. We just lost to TDK, but we are rebuilding the team and staff. We were putting the final touches on the Ember Academy. After he landed, Bao was already forced out. And within a week, most of our players and all of our staff, including him, would be released. With the birth of his child imminent, and no job in hand, what does he do? He makes sure that all the players are in a good state mentally and have a plan of action going forward before addressing his own needs. I’m so inspired by what we did after things exploded. We stayed positive. We helped one another. We did the right thing. And no one can ever take that away from us.

The TDK Game Analysis

The rest of this autopsy focuses on the TDK game and it is roughly a twenty minute read. You should probably skip it unless you really like to analyze League of Legends games. This analysis was done by our Strategic Coach, Brendan McGee so the “I”s you read below belong to him. However, I’ve made some edits to both form and substance.

Summary: Ember’s failure in the semi-finals was the result of a series of bad decisions that occurred in the weeks before the match (lack of scrim diversity, playing the new jungler with only two weeks of team practice) plus the game day itself (draft, mentality). Before we get into the game-by-game recap, let’s answer two frequently asked questions.

Q: Why did we choose purple side for games one, three and five?

1. Scrim win rate: Since patch 6.4 hit and Santorin joined the team, we had an 84% win rate on purple side, and a 62% win rate on blue side.
2. General strategy: Purple side has many advantages, such as allowing you to trade two priority champions for one, and counter-picking mid-lane. Goldenglue is one of our best players and consistently dominated Alex Ich in scrims, so I wanted to have the counter-pick mid lane to let him shine. The vast majority of teams we scrimmed on patch 6.4 would first pick Corki on blue side. Corki is one of Alex Ich’s most played champions, he has played it twice in LCS, and it’s his third most played in solo queue. Corki is widely considered the best blind pick for mid-lane since it is so strong in lane and has very few weak matchups. However, Goldenglue’s Leblanc and Zed are world class and he wanted that matchup.

Q: Why didn’t you draft Goldenglue’s carry champions?

1. We planned to, but TDK conceded in their draft that Alex Ich is simply a much weaker laner and thus blind picked extremely safe, difficult to abuse champions (Lissandra and Nautilus). This caught me off-guard since Alex has played Lissandra once in LCS and it’s not even in his top eight most played in solo queue.
2. Goldenglue has performed excellently on many supportive champions in both NACS and scrims and in many cases prefers to play them himself. For example, our Zilean win rate in scrims is 88%.

Game 1 Analysis

1) Draft

We didn’t expect TDK’s bans. Teams rarely ban Corki on blue side, and Kog’maw is extremely low priority for most teams now, including us. However, the bans didn’t matter much since we did not plan on playing any of those champions anyway. We decided to stick to the plan and play our comfort champions.

• TDK First Pick [Poppy]: Poppy first pick is a strong tank, and certainly in Flaresz champion pool, but not a huge problem at this point since Nautilus is widely considered to be better (all players agreed on this) and this basically guarantees us our most practiced and strongest top pick.
• MBR First/Second Pick [Braum, Kindred]: Nautilus is not a priority here since they already picked their top. Braum has been a highly contested support pick in many regions as well as at IEM the previous weekend. At IEM, Braum was the third most picked/banned champion and has a 20% higher win rate than Alistar. It functions extremely well in lane swaps (which we intended to do as you’ll see later), is good vs tanks like Poppy (Braum just autos all melees during a team fight to stun them all). It also has great stats for us and internationally. Additionally, Braum is our most practiced support champion and has our highest support win rate.

IEM Highest Pick/Ban Champions

• Kindred is probably the most controversial pick we had. Kindred is Santorin’s most played champion in solo queue and a personal favorite. Kindred is a strong early/mid game champion that can turn early skirmishes with her high damage and powerful ultimate. It is also a strong counter-pick to Elise, and Santorin prefers the Kindred vs. Elise match up. We banned Nidalee (Kez’s most played champion) and picked Kindred (his next most played), so we anticipated an Elise pick. We had been defeated by NRG the week prior and decided to focus on playing Kindred since they had shown such success versus us with it. Additionally, Kindred has strong early ganking pressure which we felt we needed since our plan was to pick Ezreal who needed time to scale. Thus we wanted to make up for Ezreal’s poor early game with a strong jungle match up that could pressure solo lanes.
• TDK Second/Third Pick [Elise, Janna]: The Elise pick was as expected. Janna was relatively unexpected since both Bischu and Trance (TDK’s support players) play a lot of Bard in both solo queue and scrims. However, Janna posed little threat to our planned composition since we intended to lane swap anyway and her strengths are primarily in lane.
• MBR Third/Fourth Pick [Ezreal, Nautilus]: We had planned to pick a lot of Ezreal in this series for at least two weeks. Ezreal had an incredible showing at IEM that weekend (most notable from SKT Bang) sporting a 60% win rate and felt incredibly strong. His arcane shift does not get blocked by Poppy’s W ability, he excels in lane swaps due to his attack speed steroid, and his two and three item power spikes during the mid-game are huge. Our team’s strength has always been mid-game team fighting and Ezreal fit right into this mold.
• Nautilus was our most played top pick and a champion we had planned heavily around either playing or countering. It was surprising to us that TDK did not first pick Nautilus, but regardless Solo felt incredibly confident in this match up.
• TDK Fourth/Fifth Pick [Lissandra, Caitlyn]: Again, Lissandra was a surprising blind pick from Alex Ich and something he hadn’t shown before. It made sense that TDK wanted to blind pick a mid-laner that was difficult to defeat with assassins such as Zed and Leblanc. Finally, Caitlyn meshed well with the Janna pick. TDK wanted to abuse Ezreal’s weak laning phase with Caitlyn/Janna, not only putting kill pressure on that lane but also persistently damaging the turret. This was a non-issue since we planned to lane swap anyway.
• MBR Fifth Pick [Karma]: Since Alex Ich picked Lissandra, we agreed it would be better to put Goldenglue on a supportive champion that could abuse the Lissandra pick with poke instead of all-in kill potential. We had seen Karma successful in scrims, not only on our team but other teams had played it successfully as well. We also had a very high damage jungler, and thus could afford slightly less damage from the mid-lane in exchange for some extremely powerful utility. The AoE shield/speed buff from Karma allows us to engage in our otherwise low-engage team composition and force fights, which you will see us utilizing to great effect in the actual game.

2) Post-draft thoughts

• Top: Farm lane, possible for ganks from both teams to swing match up early.
• Jungle: Kindred favored in duels as long as she dodges cocoon, high potential to stick onto Elise if caught out. Slightly less gank pressure than Elise.
• Mid: Karma favored pre-six (Karma poke much higher range than Lissandra Q), however susceptible to all-in from Lissandra at six.
• Bot: Highly favored for TDK. Caitlyn/Janna extremely oppressive lane due to Caitlyn range, faster push than Ezreal, and Janna shield.

3) Gameplay

Early Game: We execute our early game plan to perfection. We successfully lane swap, invade the enemy jungle level one to push them off their top side jungle and take the turret.

Late game: As the game goes on you can see we expertly bait Baron/Dragon over and over with perfect vision control. TDK are eventually forced to fall for the bait and we win team fights this way. Again, Goldenglue uses the karma pick to great effect, speeding up the whole team and allowing Nautilus to get hooks in these engages. LOD also uses his slow from Iceborn Gauntlet to catch members of TDK and set them up for hooks.

Misplays worth noting:

• [12:13] Goldenglue plays too far up vs Lissandra when the rest of our team is off the map. He gets three man ganked and dies.
• [14:30] Top turret is low, the enemy has double TP. This is a situation we had drilled many times in scrims. If the turret is low and there is TPand dive pressure, just back off or make sure the jungler is nearby to help and counter TP. However, Santorin is jungling on the bot-side of the map. This is a MASSIVE strategic mistake. I’m not sure if LOD/Stunt were calling for help before this dive, but if they weren’t then they should have been. This was an incredibly telegraphed play by TDK and should have been anticipated by Santorin.

Great plays worth noting:

• [4:00] Our first strategic outplay. I had seen this cheese done in LCK a few weeks back by Longzhu Gaming. We adopted it and used it successfully on every team we scrimmed in the two weeks prior and felt confident TDK would fall for it like all the rest. Of course they do and we secure an early lead for our bot lane against Ohq, significantly reducing the pressure from that lane.
• [21:20] We win a team fight, but Solo recognizes our ability to dominate even further and hook Ohq under turret for an extra kill securing us mid turret. Additionally you can see how the Karma pick enables this with its incredible team-wide mobility.

Game 2 Analysis

1) Draft

The draft goes identical to game one in the first rotations. We all agree that our previous composition worked well and the onus is on TDK to adapt.

• TDK First Pick [Poppy]
• MBR First/Second Pick [Braum, Kindred]
• TDK Second/Third Pick [Gragas, Ezreal]: TDK takes away Ezreal and picks Gragas instead of Elise. Here TDK is indeed adapting, they recognize the weakness of the Elise pick versus our composition, as well as the strength of the Ezreal pick once it gets items. We’re not too worried about the Gragas pick at this point because Gragas’s early game is weak and abusable by Kindred. However, his ultimate does counter Kindred’s ultimate but we’re confident in our ability to outplay.
• MBR Third/Fourth Pick [Lucian, Nautilus]: Lucian is considered a strong laner and for a long time has been the answer to Ezreal in the laning phase. At this point we are super confident in our early game pressure if we can get standard lanes. The reason behind the Nautilus pick is the same as game one.
• TDK Fourth/Fifth Pick [Trundle, Lissandra]: Trundle support is a smart pick from TDK here since we are running a one-tank team composition and Trundle’s ultimate will make Nautilus much less tanky in team fights. We think Lissandra was picked for the same reasons as game one. It also provides good follow up to their strong Poppy/Gragas engage.
• MBR Fifth Pick [Zilean]: Again we decide on a supportive mid-laner. Zilean was by far our strongest pick for the last month. It was so strong that many teams (LCS teams included) banned it versus us in scrims, and TDK banned it versus us when we played them in NACS. Zilean + Kindred ultimates provide us with longevity it team fights, making us very difficult to kill. TDK’s composition was extremely dependent on getting good engages and one-shotting a carry and Kindred/Zilean ultimates help neutralize that threat. Additionally, Zilean’s speed boost increases Nautilus’s ability to create team fights by speeding him up when he’s looking for hooks. Thus, in the event that we get a lead, Nautilus would have help forcing fights around objectives or in the jungle.

2) Post-draft thoughts

Overall the team felt confident in this draft and had momentum carrying over from their victory game one. However, as I reviewed the draft once the players left and went into game, I began to see where the holes were. We had very little way to pressure our early leads. No lane was really gankable for Kindred barring complete misplay from the enemy. Top was a farm fest, so was bot. Lucian should be able to pressure Ezreal in lane and gain a slight cs lead, but the kill pressure was extremely low due to pillar and Ezreal’s mobility. Lastly, of course mid lane was difficult to gank. Our composition heavily relied on the enemy forcing fights onto our team and then our team utilizing their abilities to survive the all-in and turn the fight in our favor using Kindred/Zilean ults and a double bomb stun/Nautilus counter engage. Even Braum’s ultimate was useful in slowing the enemy’s engage and once their engage failed slowing their retreat and making double bombs easier to land. However, this of course assumes the enemy plays into our hand that way.

3) Gameplay

• [1:30] We should not have allowed TDK to lane swap. It is important when we pick an early game laning bully such as Lucian that we get the standard lanes we need. I’m not sure why the team makes no effort to spot the lane swap and respond accordingly.
• [1:50] Both teams invade respective red buffs, a predictable play that we can abuse later but choose not to.
• [6:40] You can see even with Zilean, Goldenglue is putting a lot of poke pressure on Lissandra and maintaining mid priority.
• [7:00] Santorin does a good job recognizing Poppy is overextended and lands a solid gank.
• [11:00] We correctly set up vision for dragon and bait TDK into engaging on us. You can see how the Zilean and Kindred ultimates turn this fight in our favor despite TDK’s super hard engage composition. Santorin’s ultimate is good and we correctly fight front-to-back, killing the enemy tanks since Lucian is much stronger than Ezreal at this moment. Solo’s flash hook secures a final kill and we win the fight convincingly. We correctly move straight to dragon. It is important we stack dragons as an additional win condition. It also adds 5 stacks to Santorin’s devourer enchant, greatly accelerating the effectiveness of his build.
• [12:25] Here begins a catastrophic series of misplays that completely nullify our lead and get TDK all the tempo in the game.
• [12:30] LOD gets caught farming a greedy wave despite Trundle walking over a ward. We talked a bit the previous week about coordinating our recalls. We had just won a big team fight and LOD should have recalled with everyone else to spend the gold and get back on the map with full mana ready to catapult our lead into a mid turret and rift herald.
• [14:15] Solo gets ganked pushing out the top wave. I don’t think he was particularly overextended and this was simply TDK using all of their available summoners to try to make a play. We respond with a teleport from Zilean but Lissandra also teleports and they get a kill 3v2. I don’t think if Zilean had ult’d Nautilus instead of using bombs that he would have lived, although it certainly would have taken longer to kill him. The rest of our team uses this opportunity to get damage on mid, but ultimately we end up taking a deficit here due to a good play from TDK.
• [14:40] Here Santorin is the third person to get caught out in as many minutes. I do not have access to the voice comms during these games but it’s difficult for me to understand what he is doing in this position. We just saw 2 members from TDK topside after killing Solo just seconds before so it seemed like we should have avoided the top side jungle at all costs.
• These three catches bring our lead down to 1k and seriously dampen our tempo. It is crucial with this team composition that we snowball a lead off our early jungle/mid/bot pressure but instead we throw it away in careless misplays. Now the tempo is in TDK’s hands.
• [17:45] Here we again correctly push out mid and rotate onto dragon, securing vision and taking the scuttle crab. However, we start dragon immediately and completely forfeit our favorable strategic position. We should have stood on our pink wards and continuously killed vision until TDK ran out of wards and was forced to come into the river. We can then land poke and possibly even engage if we land a good double bomb/Nautilus hook. However, instead we start dragon and then tank five hits in a straight up standoff.
• We now have no tactical advantage and a series of misplays cause us to lose the team fight hard. First, we tank the dragon too long. Second, we tunnel too hard on getting the actual dragon and all three of our carries step onto a fully charged Gragas barrel to secure dragon. Gragas immediately ults, separating our team completely and leaving LOD and Goldenglue chunked to 40% hit points at the very start of the fight. Stunt ults in the wrong direction, and shields nothing so Braum basically has no function in this fight. TDK are able to do all of their damage to our front line while our back line is left running away from an untouched Poppy who was able to TP flank. The biggest mistake here is all three carries getting hit by the Gragas barrel/ultimate.
• We’re now down 2k gold (a 3k swing) and have lost all tempo to a scaling composition. The game is very hard to win at this point and the ball is in TDK’s court.
• [23:20] We set up a death bush near mid-lane correctly predicting TDK will want to come drops wards around dragon to get another TP flank. However, both Goldenglue and Stunt get restless way too early, step out of the bush, and alert TDK to their presence. Then they stay in the bush anyway. You can see Trance walking up to ward the river and then getting furiously pinged back by his teammates because we accidentally revealed ourselves. This was a pick that got countered by our own lack of patience. With Trundle dead the team fight may have gone in our favor.
• [24:00] Solo recognizes his ability to kill Lissandra and goes for it. A clutch play no doubt and buys us some time.
• [26:00] LOD gets picked off again for no reason. The big wave coming top was pushing to us quickly and he could have just waited at the turret for it. He knew our red side jungle was dangerous. Now we’re down 4k gold and cannot defend mid.
• [27:00] Good Zilean play from Goldenglue wins us a fight and we regain a small amount of tempo.
• TDK engage under turret without Lissandra and a few good double bombs turn the fight in our favor. Flaresz goes way too deep under turret and Trance tries to help. We kill both and go for Baron.
• This Baron call is another massive mistake. It’s certainly a desperation play, but I don’t think it is one we needed to do. The jungler is still alive and healthy, Lissandra is alive and untouched and still has her ultimate. Lastly, Ezreal is alive and healthy and able to stand out of range damaging us while doing Baron. We lose Braum for nothing. We should have used the won team fight to establish Baron vision and catch/prepare side waves.
• [31:30] Again, we correctly push out mid and take control of dragon vision. However, we let TDK into the river without defending the choke. We start Dragon but are unprepared to fight for it and get slaughtered by TDK due to another good Gragas ult.
• We secure dragon at the expense of a significant amount of health on our front line. The call was to retreat but Solo and LOD seem to not be on the same page. The team gets split up on the retreat and Solo/LOD are easily picked off by the re-engage from TDK.
• TDK gets Baron and the game is essentially lost.

4) Conclusion: We lose the game due to careless catches in the mid-game losing us the hard-earned tempo from winning the first team fight. The tempo loss leads to TDK being strong enough to contest us at Dragon and our simple team fighting misplays lose us the fight and shoot TDK into the lead. Overall, the team looked much less confident this game and regressed back to the type of misplays we had worked hard to reduce.

Tally of game effecting misplays: LOD: 3 (2 extremely bad catches), Stunt: 1 (Bad Braum ultimate in teamfight), Santorin: 2 (1 extremely bad catch), Solo: Non-optimal team fighting (potentially his call to just start the dragons, but no way to tell at this point), Goldenglue: Almost none of his Zilean ults are actually effective this game. Additionally, a desperate baron call throws an opportunity TDK handed us to collect free farm, prep waves, and recover a lot of vision.

Break between Game 2 and Game 3

During game two, it became clear to me that the Poppy pick was a big part of what was contributing to TDK’s pressure and success. I discussed with Ken and Michael (the other analysts) the possibility of changing our bans to include Poppy and leave Kalista/Nidalee open. TDK would likely choose one of these OP’s (almost certainly the Nidalee since it is Kez’s favorite jungler and Santorin has had a lot of success in the past with Nidalee) and we would get Kalista.

Kalista is a huge strength for us and gives us play making ability in the botlane and much needed hard engage to get fights onto Ohq. However, as soon as I bring this up with the players they reject the idea of changing bans. Instead, they want to take Lissandra away from Alex Ich. If we could get Alex off Lissandra then he would likely default back to Orianna, allowing us to get Goldenglue his Leblanc/Zed game.

I reluctantly agree to this plan for the draft. And to this moment regret not asserting myself more. We had never come up against a roadblock like this with such pressure. For the 1.5 weeks I had been put in charge of drafting we had always made it a democratic process where I suggest picks and players choose from the options what they feel more comfortable on. Occasionally in scrims I take full control to force them to play things new and outside their comfort zone, but in this pressure situation it was difficult for me to have the foresight to contest the players on what they all agreed was the problem. This was a catastrophic lack of planning.

We should have had contingency plans for changing bans during the series and gotten practice versus characters like Nidalee and Kalista. However, every time I tried to make them play versus Nidalee they universally rejected the idea and said that champion should never be left open (even though many international teams have done so frequently, including at IEM). The players should not have had the ability to reject anything I wanted to do. I should have had the authority and rapport to enforce a training regimen. 50% of this is my fault, I should have taken control. 50% is the result of the power vacuum left by the firing of the last coach. The last coach never really had any authority and his plans were often vetoed by the players. From the birth of the team they had been conditioned to think that they had final say in draft. For me to wrestle that authority back from them would have taken the foresight to understand that this could lead to problems in the future. Nobody saw that, we’d never had a problem with it. It was a ‘why fix what ain’t broke’ situation.

Game 3 Analysis

1) Draft

• TDK First Pick [Poppy]: Of course.
• MBR First/Second Pick [Ezreal, Braum]: We decide to take Ezreal and Braum here. We take Ezreal away from Ohq because it is one of his favorite champions and we don’t want him to have it again. Ezreal’s scaling is way too strong and we want our team to have the scaling marksman so that in the event that the game goes long again we have a fail-safe.
• Why not Nautilus? TDK had shown no priority for Nautilus up until this game. Also, they had already picked their top laner so I felt no urgency to get Nautilus. I was certain we would get it in the second rotation like the previous two games. We also didn’t want to pick mid obviously to save the counter-pick. So the only other option other than picking support was jungle. We wanted to abandon the Kindred jungle pick first rotation to see if Kez would go for it. We were confident in playing Gragas into that match up and thought TDK might see an opportunity to take Kindred away from us. Here is the only place I think that I could have potentially saved this draft and that was by picking Gragas instead of Braum, but again there seemed to be no reason to prioritize it.
• TDK Second/Third Pick [Gragas, Nautilus]: This is a surprise to everyone and we naturally assume it is Nautilus support at this point. I had never seen Nautilus mid before in any region. Nautilus support is widely considered to be bad on this patch and easily dealt with by Braum.
• MBR Third/Fourth Pick [Lissandra, Rek’sai]: Here we take Lissandra away from Alex Ich (or so we think). Lissandra has a good win rate for us in both the top and mid lane. It also punishes Poppy early and gives us a way to make plays. At this point we are still assuming Alex will pick Orianna, Karma, or some kind of support mage. Santorin had no idea what he wanted to play here. With Contractz we would have immediately picked Elise and put a lot of pressure on the map, something we’ve done many times before. However, Santorin does not like playing Elise at all, didn’t want to play Kindred into Gragas again, and thus defaulted onto Rek’sai. It’s already clear to me the draft is in a bad place.
• TDK Fourth/Fifth Pick [Vayne, Alistar]: Now it’s clear that it must be Nautilus mid. TDK have drafted four tanks and a Vayne. Vayne has been seen zero times in all regions on both patch 6.2 and 6.3, and no region had played on patch 6.4 yet. Ohq had two Vayne games in solo queue. However, TDK had legitimately outsmarted us and we had a composition very poorly prepared to deal with them.
• MBR Fifth Pick [Morgana]: We had about 30 seconds to discuss what we wanted to do here. Solo wanted to put Lissandra mid and pick Trundle for top. Trundle does well vs tanks and can out duel Poppy in the late game. However, in the past we’ve cheesed Nautilus top picks with Morgana. Morgana can continually push Nautilus into turret and chunk him, allowing us to get slow chip damage on the mid turret and potentially get it for free after a favorable gank. Also, against four melees Morgana is actually very strong if we have a lead in the game since it is very easy for her to get many targets stunned in her ultimate. However, it was painfully obvious to everyone that Vayne would probably never die if we didn’t get a strong lead in the early game.

2) Pre-Game Mentality

As far as I could tell, the team’s mentality going into game three was pretty low. This was not the easy domination we were all expecting. Everyone knew we played poorly in game two and now we were heading into game three with a really strange draft that did not feel good at all. At this point I suggest that we lane swap. And instead of invading the enemy’s blue buff, let’s predict their invasion of our red.

However, the players objected to the plan, saying they wanted to play standard. Here I should have been even more adamant, and I should have had a more detailed image to present to them. The enemy did exactly what I said they would do and we could have opened game three with a cheesy one or two kill lead into our preferred lane swap scenario. This might have been enough to carry us through to an early game lead strong enough to beat the four tank composition.

3) Gameplay

• [1:50] Sure enough TDK lane swap and invade our red buff. Had we gone ahead with the plan we could have punished this.
• [3:30] We play the lane swap fine, but we don’t have the damage to kill Nautilus mid with Santorin’s early gank.
• [6:00] As stated before this game hinges on us getting an early lead. Any game vs. tanks is like an arms race, and if we were able to secure leads for our solo laners then we’d be able to outpace their tanks’ magic resist with our own damage and penetration. You can’t blame our players for not trying, they recognize that Nautilus has no teleport and once Goldenglue hits level six he recalls and calls for a dive top lane to kill the Vayne.
• However, before the TP’s come down we see Gragas in the area on a ward. I think we could have backed off here and salvaged this situation, but we go in anyway. All the TPs come down and we get slammed by Alistar and Gragas. In the end we trade two for one and start a slow slide to defeat. Nobody really played the fight poorly except perhaps Stunt, who misses the initial Braum Q and thus doesn’t get turret aggro. Instead, Goldenglue gets turret aggro and tanks 3 shots making him low enough to die in the chase.
• [9:30] Here Santorin recklessly ults into the enemy red buff despite both bot and mid being pushed in. Gragas is waiting for him there and the resulting chase gets Goldenglue killed because both solo lanes rotate faster for TDK. Small misplays like this all contribute to the game sliding out of our control.
• [11:00] Santorin ganks Poppy but she is too tanky and turns the gank for a kill with Gragas’ help. Solo doesn’t flash and probably could have lived if he did.
• [13:40] We make a good TP play on the topside of the map. Crucial we make some kind of play to punish the Vayne or else the game is over. The play is setup by LOD/Stunt playing the 2v2 well and chunking out Vayne. However, Stunt ults the wrong direction and then runs up instead of down to his team. If he ran down then we could have fought the rotating Nautilus/Gragas after Vayne was dead, but instead we end up trading one for two. Still a positive trade and something we needed desperately.
• [15:00] and [16:00] Solo gets caught out twice in a row dying both times, giving TDK even more momentum and helping them finish their crucial tanky items before he can finish any real damage items. I’m not sure how much of a mistake the first catch was, but when Alex catches Solo the second time he is definitely overextended. These mistakes cannot be overstated and are a hallmark of our team’s weakness. We die in sidelanes incredibly frequently, and almost never get those types of picks ourselves. This is largely due to a lack of knowledge on how to manage side-waves properly.
• [19:00] A good team fight setup by Poppy getting caught overextended. Solo and Goldenglue truly shine in this teamfight. Solo’s use of flash plus ult while targeted by Nautilus ult basically nullifies the Nautilus damage and locks up the TDK frontline to be killed by LOD.
• I cannot express how much I want to highlight this play by Solo. This is LCK-level presence of mind in team fights. This is the kind of team fighting you can’t really teach. It is the main reason I don’t think Solo should be removed from the team.
• Then, Goldenglue flashes forward for the three man ult and lands a q onto Ohq, which secures a kill on the Vayne. Our success in this team fight reinvigorates the team and we then utilize a huge wave top at [21:00] to outpush Vayne by herself in the bot lane, trading one turret for two. However, despite this we are still down 4k gold due to the many misplays earlier in the game.
• At this point our team excels at picks. We should look to control vision around certain objectives and bush camp. Tanks are easy to kill when locked up in CC and by themselves. Instead we continue prioritizing side-lane farm and not being grouped at the moments we should be.
• [23:30] Santorin gets caught in a weird position leaving the team and this starts a teamfight. We get a dream wombo-combo but it’s too late in the game for it to matter. The enemy has a huge amount of magic resist and Goldenglue gets Gragas ult’d out of the fight. If the Morgana ult had clicked on all four of the frontline, we might have won the fight, but we had basically lost the arms race at this point and couldn’t kill the tanks or the Vayne. Still, Solo and Goldenglue played admirably in this fight. The Lissandra engage from Solo would have been game winning in other circumstances.
• At this point the game is basically over and the players know it. At [25:40] LOD gets caught again and at [28:00] we lose a pretty one-sided teamfight and baron and that’s essentially the game.

4) Conclusion:

• This game felt extremely difficult from the outset due to the draft. I got legitimately outsmarted by TDK and didn’t have the correct answer to Nautilus mid.
• I failed to force the team to do my level one plan and we miss a crucial opportunity to adapt and punish TDK’s habitual level one. This is standard best of five material but I failed to get the players to do it.
• Getting caught out in side lanes was still a huge problem this game, the tally is: LOD: 1, Santorin: 2, Goldenglue: 0, Solo: 2, Stunt: 0

Break between Game 3 and Game 4

Having watched this game and clearly seen the issues with the draft, this time I prepare a mock draft scenario for the players to look at (below). I want to take Poppy and Ezreal off the table, then secure Kalista and Alistar in the first rotation. As stated before, this gives us huge playmaking potential on bot lane, and Ohq would basically be forced back onto Vayne or Caitlyn, who are both extremely vulnerable to Kalista/Alistar.

Proposed Game 4 Draft

TDK could do a number of things here and what’s shown is only one possible scenario. Whether they take Nautilus or not, send it mid or top, or pick Graves, it doesn’t matter. What matters is we have an extremely strong hard-engage composition, one that we have practiced many times which allows us to throw all of our considerable weight onto Ohq in team fights. Once the Vayne dies, we have a similarly tanky lineup that can continue to fight, the key difference being that we have an extremely tanky ADC as well. Versus a tanky lineup like TDK’s, Kalista does very well because she builds BotRK 1st item, then lots of tanky and lifesteal heavy items after that. She can easily survive the relatively weak burst of a Nautilus and then quickly lifesteal the damage back up while dealing %hp damage to their front line. However, again the players reject the idea, and to this day I really don’t know why. The only explanation I have is that it was lack of scrims and training where champions like Kalista and Nidalee were left open. They were just too scared to play versus Nidalee.

Game 4 Analysis

1) Draft

• TDK First Pick [Poppy]
• MBR First/Second Pick [Ezreal, Gragas]: In the end, nothing changes and the players are convinced Gragas was the problem last game, not Nautilus. Also, Ohq cannot be allowed to have Ezreal. At this point the players are essentially drafting for themselves since they rejected my ban proposal. Nautilus/Gragas would have probably been better here to prevent the Nautilus cheese again, and Ohq would have probably gone Vayne anyway.
• TDK Second/Third Pick [Nautilus, Elise]: It’s safe to say the players are anticipating the Nautilus mid again and we’re already thinking of how to counter it better than Morgana. Elise for Kez didn’t have a great showing game one anyway so we’re not too worried about that. Although Elise does have more pressure in the early game than Gragas, Gragas however out scales.
• MBR Third/Fourth Pick [Trundle, Braum]: Trundle is an excellent anti-tank pick since he can strip tank stats with his ultimate and become even tankier himself. Also, if he has a lead he can actually duel Poppy in the late game. We decide to stick with the Braum pick since it is effective versus tanks and still a strong laning/team fighting support champion.
• TDK Fourth/Fifth Pick [Vayne, Alistar]: TDK predictably take Vayne and Alistar. We really have no answer to this at this point, and the draft has gone completely off the rails. Everyone on the team is in a poor mentality and rushing through choices without really consulting me. At this point we need a strong scaling mage in the mid-lane that can one shot Vayne and still deal strong magic damage to the tanks.
• MBR Fifth Pick [Orianna]: The idea is that melee champions have a very broadcasted group up and run at you death-ball style. Orianna prefers this situation since anytime the enemy is coming towards you it is very easy to land an Orianna ult. Plus, Orianna deals very high magic damage when she gets four to five items, enough to deal with tanks as long as we are not too far behind.
• Overall, I think we all felt that the draft was sub-par and we had lost the series. The players went into the game, again rejected the idea of cheesing our own red buff, and played exactly how you would expect with that mentality. I take responsibility for the game three and four drafts. I had the conceptual ideas of how to adapt, but I lacked the leadership to override the players and force my agenda. Having been made strategic coach only a week before, I had already built rapport with the players due to my previous work. However, I didn’t have time to truly build my confidence in the actual drafting procedure to deal with extremely tense, high pressure situations such as this. I had done it once before and it worked. During a scrim set with TSM I started reading off bans. When the players asked why we were banning that, I responded “just trust me, I have a plan.” They trusted me, ended up liking the draft and we beat TSM in that set. If I had only pulled rank that way during this series, things might have turned out differently.
• Another failure on my part is when Goldenglue asked me what summoner spell he should go on Orianna. He said he wanted to go ignite so they could try to kill the Nautilus early and prevent him from getting too tanky. I told him he should go teleport, and one other player (I believe it was Solo) agreed with me. However, Goldenglue said he wanted to take ignite anyway since he really felt we needed more damage in the early game.

2) Gameplay

• The early game (first ~10 minutes) was extremely passive and consisted of the standard double lane swap. We traded both side lane turrets and entered the early-mid game with a tempo advantage. This tempo advantage was because TDK spent so much time trying to kill Orianna that they fell behind in the turret pushing game. They also decided to spend time stealing blue buff instead of having everyone push our bot tier two turret.
• [8:15] Because of our tempo advantage we are able to push an extra turret on the top side of the map and trade two turrets for three, garnering a small gold lead for ourselves. This was a good call from the team. The only downside is that Orianna dies in the mid lane from a Poppy roam.
• However, at this point it becomes obvious that our team’s mentality has hit rock bottom. We are making zero plays on the map, conceding all map objectives without even trying to fight, and slowly losing the game by attrition.
• [15:45] Here we make the first real play of the game. Santorin moves bot to gank an overextended Poppy who is low on hp and mana. Solo recognizes the chance to kill her with his ultimate and calls Santorin down. However, Nautilus immediately starts to TP in and Elise can be seen rotating to help. Orianna has no teleport and is unable to respond. If Orianna had TP then we might have been able to match teleports, kill the Poppy quickly and clean up the rest of the fight. Instead we lose Gragas and our top’s summoners.
• [18:00] and [19:30] and [21:00] Solo gets caught out three times pushing out the bot lane. Because of our extremely passive play style we have no control of either quadrants of our jungle, and TDK is able to rotate on both of our side lanes to catch anyone who ventures out while keeping Orianna pushed into turret. However, these catches are avoidable.
• [23:45] We catastrophically lose a baron team fight and TDK take Baron and the series is essentially over.

3) Conclusion: Not enforcing my new ban strategy left us with a difficult choice to make in the first rotation. We probably should have gone Nautilus and Gragas, but that would not have made a huge difference. Ohq would get Ezreal, which carries just as hard in the late game vs our team as a Vayne.

Not enforcing teleport choice on Orianna lead to us being unable to set up a single play on the side lanes after the lane swap phase was over. We were left trying to scoop up farm when it came to us, giving up all rift heralds and dragons and the majority of our jungle camps.

Player mentality was at an all time low during this set. You could tell because of the number of times Solo died in a side lane. It appeared we had just conceded the series at this point and were playing it out out of courtesy.

Conclusion

Thanks again to our fans who supported us. I hope you will continue to support our players as they move onto new teams. #MBRWIN