The clan leader’s CWL survival guide

[Maethriel]

Thanks Onyx for your time and effort putting this all together and then updating/maintaining, very informative and a great help.

[Crusher0029]

Great post, well done

[IndiaRocks]

This post should be sticky. it is easy for us to understand the medals. thx Onyx

[Skrags]

Friendly forum bump for players still struggling with CWL mechanics !

[kiyamu]

Hi everyone,

I wanted to share our clan's and mainly my experience (as the clan leader) on the past CWL and the kind of preparation we did for the current one. The title refers to the following post:
https://forum.supercell.com/showthre...ighlight=guide
It is a great post by OnyxDS in case you haven't read it and it helped me/my clan a lot to survive and strive during the first CWL. The goal of this post is not to repeat or conflict the information provided on the other post. OnyxDS provided an excellent overview and overall guidance, however he mentioned that his clan won't be participating to CWL. Therefore, I thought that real experience from a mid-tiered clan can be another perspective and hopefully valuable to the others. This post will be mostly on what our clan did in the first CWL and doing differently in the second CWL. I don't necessarily argue whether our approach is good (or the best) but I sincerely believe that it helped our success tremendously and I hope it can help other clans as well.

A quick word about our clan

As this post is not about advertising our clan, I won't put a name or tag, but I want to give an overall idea. When we participated to CWL, our clan had about 40 people in it and we would make mostly 10x10 or 15x15 wars. In other words, there were less than 15 active war players, which was already a big threat to our participation. More details below, but we managed to get the following rooster at the end.
- Me, a recent TH12 with a few max TH12 offense, but mostly max TH11 defense.
- A max TH11 that stopped playing for a while. We convinced him just for CWL.
- An early TH11, mostly max TH10 defenses and offense with a few extra on TH11.
- 2 (almost)max TH10s, one being my mini.
- 1 late TH10 who does not war regularly. (Offense-wise he was worse than some of our TH9s at the beginning)
- 4 max TH9s, missing several levels from heroes only.
- 4 mid TH9s, some does not even have LavaLoon enabled.
- 1 early TH9, only has DragLoon.
and that's it. We did not have extra players in the rooster.

With this rooster, we got matched in Gold III league. And this is where the journey started.

Communication is the key

In my humble opinion, to be successful in CWL, you need to be able to communication with your clan mates (at least CWL participants) efficiently. Unfortunately, as we all know, the in-game clan chat is not good enough for this. We were lucky that Discord was already a requirement for our clan and we had all participants in Discord. It does not have to be Discord, but you need to get ALL participants to an external chat application to coordinate hits, etc. (more on this later) This is especially more difficult for an international clan like us due to time difference. However, we survived; so I am sure you can as well.

Time it right!

This is again likely a concern only for international clans, however I realized that I need to start wars at a less convenient time for me so that most of my clan mates will be available when the war begins and war ends. I live in west coast USA and we have members all over the USA, UK, India, and even Australia. I found that if I start the wars around 9 AM Pacific time, it works well such that most players are awake to give preferences (more on this later) and even do war attacks before they go to sleep.

Get preferences + create assignments + complete assignments successfully = Success !
It took a while for me to get this right and I learned a lot from OnyxDS' writing to get this right. You have to approach CWL with a different mindset than regular wars because each member has a single war attack. As OnyxDS mentioned, consistent 2-star war attacks win the war. Most of the players don't have this mindset. For the clan to maximize their winning chance and stars, you need "war generals" (strategists) that can create assignments (i.e., who should hit which village for how many stars). This is easier to explain and done. Here is how we ended up doing, which seems to be working very well.

• Once war stars, I expect all members to post their preferences: villages they can 2-star and 3-star comfortably. In general, I am giving several hours to players for this phase and until I get enough (ideally all) preferences no attack is permitted.
• I look at the preferences and do an assignment that maximizes the total number of stars. I started with the algorithm that OnyxDS mentioned in the initial post and ended up improving it quite a lot. Initially, I was doing this semi-manually (using Excel, etc.), however ended up writing a small program that does this during the first CWL. (more on this later as well)
• Once the assignments are finalized, they are posted and everyone is free to attack at will. Ideally, we still want to start clearing from bottom up, just in case something unexpected happens.
• We don't particularly care what the enemy does; we always strive for the best possible outcome and hope that is enough to win the war.

Note that this is really tough to get right, especially at the beginning. We had the following issues:

• Members would forget to provide preferences and I would have to track them personally, or even worse, guess what their preferences are.
• Members would over and under approximate their preferences. It is e.g., quite difficult for a TH9 to know whether they can 2-star a TH10 comfortably, especially if they are not farming at high leagues and/or doing this regularly in wars.

Even with perfect preferences, failed assignments happen due to e.g., bad luck during war attack. Therefore, I had to re-adjust assignments during the day and post updates when/if needed. This is the reason why it is preferable if attacks happen from bottom to top. I was doing this part manually at the beginning, which made it very difficult, but now my program takes care of this as well.

Results?

By following these simple practices, we finished the league at the first place and won all our wars. Although our village was strong in our league (I would rank us as either the second or the third village), without these practices I don't think we would finish the first or win all our wars.

Also, I recently modified my program so that it can be applied to regular wars. For regular wars, I don't micro manage, but instead only create assignments for TH11 and above. To give you an idea, this is generally 6-7 members in my team and focuses on the top 10 (our of 25) villages. It is also less strict compared to CWL since each member has two war attacks. Regardless, I see that these practices are already benefiting us in regular wars as well; we had finished our last to wars with 1 and 2 stars missing (e.g., 74/75), which would never be possible before.

Changes in the current CWL (this season)

The main change for us was new members. After CWL we had done another recruiting spree and managed to get very good new members: 2 (almost) max TH12s, 1 (almost) max TH11, and 2 engineered TH11s (and also some TH10s and TH9s). Although the engineered TH11 accounts decided to sit out of the current CWL, we still had the problem that many other clans had experienced before. How do we decide the 15 members that war and how do we decide the initial line-up? As mentioned before, I like to think myself as trying to balance fun and winning, so this is what we came with.

Initial line-up:

• People must sign-up for CWL; we don't include anyone by default.
• If the member had a good previous CWL performance, then s/he is included.
• If the member had a good war performance with us, then s/he is included.
• Otherwise, the member is not included.

The war (15) line-up:

• TH10 and above from the initial line-up are included unless they have limited (or bad) war experience.
• TH9s are included only if they had been warring with us for the last month (loyalty and gives a chance to better understand their war skills). If there are too many TH9s, they rotate within themselves.

For example, if we had 10 TH10 and above and 10 TH9 that fits the description, then TH10 and above would be in all wars but TH9s would be in every other war (rotate within themselves).

Of course, some members don't like this or find it "fair". However, I believe that it is as fair as it gets without jeopardizing clan's success a lot. It also hopefully gives lower TH members something to strive for, e.g., faster progress.

Other than this, I also improved my program so that it can now make assignments based on "stretch" preferences. For example, you would say that something is 3-stars when 2-stars is extremely safe and it could turn into 3-star with some luck. We will soon see whether this will help or not (it did in regular wars).

Feel free to leave your own experience and insights or ask questions. Regardless, good luck for this season.

Clash on!

[teamrockie]

Impressive. We ran all 12s. After we find a match in CWL, we will pull player datas from clash of stats for each opponent and construct a spreadsheet. Then we rank the difficulty level of each war on a scale of 1-10. Best attackers will fight the toughest wars and the rest can sub for easy ones. We do have a tracking sheet for individual performances on all regular wars prior to the CWL. That’s how we determine who are the best players. On war day, every player first need to call the enemy base on war map, then he must sketch his attacking plan and post it in the chat room, at least 3-4 guys (including those sitting on the sideline) will look at the plan and discuss with the attacker, finally he can make his attack. So far everyone is on the same page and the clan is more active than ever.

[munaeric]

To lazy to read all that🙁

[littledoctor]

Impressive. We ran all 12s. After we find a match in CWL, we will pull player datas from clash of stats for each opponent and construct a spreadsheet. Then we rank the difficulty level of each war on a scale of 1-10. Best attackers will fight the toughest wars and the rest can sub for easy ones. We do have a tracking sheet for individual performances on all regular wars prior to the CWL. That’s how we determine who are the best players. On war day, every player first need to call the enemy base on war map, then he must sketch his attacking plan and post it in the chat room, at least 3-4 guys (including those sitting on the sideline) will look at the plan and discuss with the attacker, finally he can make his attack. So far everyone is on the same page and the clan is more active than ever.

That is a good, deep plan, and it sounds like you will do very well.

In our low clan, I just prebuild armies on 5 out of my 6 accounts, and start hitting bases as soon as the war starts. Strong two hit top two (if we know we are missing any attacks we'll skip the top base or two), little bases hit little bases, and my th9 builds a custom army for whoever he needs to attack (usually third strongest).

In our high clan, with mixed TH12-Th10 composition, the leadership has a pretty good idea of who can handle what kind of base, so we take a look at the war bases and assign targets based on their layouts. E.g. if a high base has an anti 3 layout we'll put a low person on it, we put the people who can't handle eagle bases on hitting non-eagle bases, and so on. Sometimes we'll say something like "Th11s have to 2 star the top 3, figure out amongst yourselves who gets which base".

[clanayyyo]

Thanks Onyx for such a great guide Wish I had read it before round 1 of CWL

Having won our first CWL league to promote to masters 2, in this second round CWL we face a whole bunch of clans with 2-3x the number of TH12's that we have (we have 3). No advantage anywhere else in the lineup against most of them but as this is a league and not "matched", so be it I guess!

It does look like there will be an awful lot of movement up and down the leagues in these early days as clans "find their level" - and those who went in low (without all their TH12's first CWL) get promoted to a more natural league level.

That level does look to be almost entirely determined by the number of quality TH12's they have? Our oppos with 7 or 8 quality TH12's are being promoted to masters 1. We, with our 3 TH12's will get demoted.

Anyone have a view on how this will play out in terms of TH12's per league?

Overall we are enjoying CWL, it's fun ! But it would be nice to face our level of oppos sometime soonish

Have fun,
Peter

[OnyxDS]

That level does look to be almost entirely determined by the number of quality TH12's they have? Our oppos with 7 or 8 quality TH12's are being promoted to masters 1. We, with our 3 TH12's will get demoted.

Anyone have a view on how this will play out in terms of TH12's per league?

From what I'm seeing it's typically 2-5 in M3, 6-8 in M2 and 9-15 in M1, although there are outliers, such as the clan we met in M2 with 14 TH12s - all as strong as our top ones.

I'm trying to see if I can get any data from the API for proper averages; but there is one major drawback in that the API gives full details of every war and every attack. But omits to describe which league it's happening in. So I can only deduce anything from this if I can reliably statistically separate the league groups. I've downloaded around 8GB of data - I just need to a bit of coding to analyse it when I have time.