I think your first match is better for you and the 2nd better for your opponent. However both are within reason and the winner will be decided by who attacks better.
Printable View
I'm still working on my laloon game.
Using pentaloon, how do you deal with TH9 CC baby dragon, where CC is centered in the base, and all ADs are behind a layer of defenses (think box with all 4 sides having an AD) and the AQ is protected? I'm thinking I don't have room to sacrifice loons since one is running penta, and I'm guessing it would take 3-4 loons out of the 16 to successfully pull the CC.
Or do you run a variation and zap quake an AD to make an easier pull on the CC, replacing a hound with archers/minions + loons?
If symmetrical base and no anti-3 star, just spam laloon. Bring 2 poison to double poison cc or poison on CC+AQ. 5 hounds, 3 on the side with AQ, 2 without. On dip attacks, bring 2 haste, 3 rage, 1 heal; on same TH level, drop a rage. Usually bases with lvl6 AD are left for TH9s, lvl7 AD will depend on base layout.
If anti-3 star base, it really depends on the base. Sometimes CB golaloon works, sometimes queen pop, sometimes just spamming is fine. We've had bases mauled using mass loons + 4 heals before on a maxed TH9 anti-3 star
Quibble: don't most competent war clans get lvl7 AD first thing upon reaching th9? You're not leaving many targets for th9 that way.
I agree that kill squad is the way to go, unless it is absolutely impossible to draw CC and AQ. But if a th9 tries to make an undrawable base, typically (almost invariably) it will be one of three things:
1) a dispersed and large-compartment layout more easily killed by govalk;
2) a centered base with AD close enough together that your rages and pups will traverse easily;
3) a centered base with AD widely separated and towards the exterior, in which case you are going to be able to snipe at least one and possibly two.
I might be oversimplifying, but that's my experience. A th9 simply doesn't have enough structures or walls to avoid falling into one of those three categories.
So more often we see bases susceptible to kill squads. And kill squad composition is fun! I've seen and/or used golems, bowlers, AQ, queen walk plus BK, even valks--depends on layout. Kill squad defense has become important, too; since loons got overpowered in the last update, we've definitely redesigned some bases to avoid kill squads that take out CC, AQ, and one or even two AD.
A perfect war match, in my opinion, would have both sides have the exact same defense and offense, making the results of the war depend entirely on skill of attacking and base designing. A perfect war match has nothing to do with the percentages of chance of winning. A perfect war match isn't something SC can achieve, but that shouldn't stop SC from changing the matchmaker with that goal of perfection in mind.
^^ This = Perfect Match on paper. Attack skills/base design would be they key. How it should be. Bring on the standard issue war bases!
Or maybe there could be three categories for war, up from the two currently. And they each get different maximum loot amounts:
Full loot = current system. You wants your money, you takes your chances.
50% loot = get a randomized but EQUAL mix of defenses, heroes, troops. You can keep your base design of course, but your existing troops, heroes, wall levels are irrelevant.
This is perfect matching, according to some folks... Success depends only on attack skill and base design, and not on any strategically chosen build paths.
Also, given the prevalence of multiple accounts within a clan, it would be best to have a completely randomized mix of assignments for members to bases, so you can't have your best attacker do all the critical work.
It also removes the problem of worthless bottom-filling accounts instantly, since they won't even do their attacks, or will probably suck if they do. Some combo of this idea plus a roster-engineering fix would be great, though.
0 loot = friendlies.
We can quibble about the relative weights, but consider:
Anyone OK with the current system is completely unaffected. And it doesn't preclude changes to the overall MM algorithm.
Anyone who wants a skill-based match can have it, instantly. Just pay a loot penalty, because MM can't be perfect and you are expecting perfection.
Am I serious? Maybe not entirely. But I'm not entirely joking either.