Fabian* strategy for war

I don’t disagree with this. Or with much of your post.

I think there’s a real danger in being misled by the statement: “we beat the Fabian strategy, so it doesn’t work.”

It’s possible to have very poor matchups in wars, by any way of objectively measuring alliance power. I don’t at all understand how my alliance has 23000 trophies facing a 37000 trophy team (+60%) has a Titan score with 1000. That seems… questionable. But hey. It happened.

When people are posting here that wars go good or bad for them, having some measure of the relative strength and activity of the two teams is important. At least to me.

I think there’s a decent chance that this data would support that it doesn’t matter what strategy is used, if your team has a significant edge in power and similar or more activity, you will win.

If the findings disprove that, i would love to explore that as well. But without those numbers there is nothing to explore.

I agree.

I have two Wu Kongs, but I never put him on defense because his special is not that special without combos. But he boosts my titan damage. Especially two Wu Kongs versus 5* purple titans.

I also swap out my 8/8 4* 4.70 Melendor for 6/8 4* 3.1 Kashhrek on defense because of his Protection from red buff works well in the center. But Kashhrek’s protection from red don’t help versus red titans, because I swap in Kiril 8/8 4* 3.59 or other blue 8/8 4* 3.60 heroes.

Good A.I. controlled heroes are often totally different from good player controlled heroes.

Really hoping the Devs plan for Alliance wars is to get the attack and defense code alpha tested ( us players ) while they work on the code for matching.

1 Like

What is the power of your defense team with and without WuKong?

I can see this being possible, yes, but I have yet to see any evidence of it actually happening. It would be very difficult to gather such evidence, because players’ hero rosters are opaque, apart from their defence teams, and I have not noticed a lot of alliance war matched pairs publishing their full rosters for comparison.

Some wars are slaughters, of course, but whether this is down to mismatched ‘alliance power’ is totally unclear.

Maybe you’re just good at Titans and they’re just bad at Titans. Or maybe they’re really into raiding, and you’ve got a lot of disinterested raiders. Or cup droppers. Or maybe they have a whole bunch of inactive players who have effectively retired but left strong raid defences in place. Or maybe their power distribution has a heavy left skew while yours is normally distributed or right skewed.

We’re never likely to acquire data with sufficient granularity to identify precisely what is happening. It would take quite a lot, in most circumstances, even to say with confidence which alliance is better suited to wars.

I agree with all of that. But collecting alliances’ aggregate raid defence teams’ power will provide no useful measure of the relative strength and activity of the two teams. What’s the best proxy we have? Titan Score (shock, horror!) because it accounts for top team strength, bench strength, player activity and understanding of the game (at least where hitting Titans is concerned).

It’s not perfect, of course. Some heroes are vital to collecting big Titan scores but useless in wars. Some alliances are profligate users of items when killing Titans, while others aren’t. Bench depth is considerably more important for wars than for Titans. Some players will love to be involved with Titans but hate to participate in wars (or vice versa). And some alliances will use stupid war strategies, while others will develop clever war strategies.

But collecting aggregate raid defence teams’ power would provide no more clarity on any of those issues. It would only measure the total power of an alliance’s raid defence teams - a number which would be of very limited use even in assessing whether an alliance was collectively good at defending raids.

1 Like

The left team is mine. The next two columns are the last two teams we faced in war. We lost both significantly.

I’m not sure why their power is much higher than ours. I mean. There’s no incentive for other teams to be that poor at titans relative to their overall power and activity level.

But here it is, twice in a row. >67% attacks used for these teams. And they’re doubling us up in wars.

You can say this is an aberration or doesn’t happen all you want, but that’s exactly why I want to collect this data.

If it’s happening more than you expect, which I think is possible or probable, then having this data to help to isolate Fabian as effective or not is important.

I understand that Titan score SHOULD be a great indicator of team effectiveness. But here’s evidence that it isn’t always so.

And that’s why it should be explored more. Go ahead and tell me we shouldn’t. That’s basically our thing.

I’ll just paste what I wrote above:

"Maybe you’re just good at Titans and they’re just bad at Titans. Or maybe they’re really into raiding, and you’ve got a lot of disinterested raiders. Or cup droppers. Or maybe they have a whole bunch of inactive players who have effectively retired but left strong raid defences in place. Or maybe their power distribution has a heavy left skew while yours is normally distributed or right skewed.

We’re never likely to acquire data with sufficient granularity to identify precisely what is happening. It would take quite a lot, in most circumstances, even to say with confidence which alliance is better suited to wars."

We already know that Titan score isn’t necessarily going to be a great indicator of an alliance’s effectiveness in war - that’s no surprise to anyone. I’ll just paste what I wrote above:

It’s not perfect, of course. Some heroes are vital to collecting big Titan scores but useless in wars. Some alliances are profligate users of items when killing Titans, while others aren’t. Bench depth is considerably more important for wars than for Titans. Some players will love to be involved with Titans but hate to participate in wars (or vice versa). And some alliances will use stupid war strategies, while others will develop clever war strategies.

So I’m left wondering what point you’re trying to make. Perhaps you are arguing that you lost to three alliances whose raid defences had greater aggregate power than yours, because their raid defences had greater aggregate power than yours? But you haven’t provided any evidence of this. Firstly, not to be cliché, but correlation is not causation. Secondly, we know nothing about your bench depth vs their bench depth - if you have terrible benches then that is far more likely to be a cause of your repeated defeats than out powered raid defence teams. And finally, of course, we don’t know how you (or they) are working as alliances. Perhaps you are exceptionally skilled at Titans and exceptionally unskilled at wars. We just can’t tell.

What we do know is that it would be entirely possible for you to be matched against alliances with stronger raid defence teams than yours twice a week every week, to lose every time, and for this to have nothing at all to do with those raid defence teams being stronger than yours, in aggregate.

Which is why, of course, we would be silly to waste time on that data - it can never prove anything, because so much is unknown and unknowable (at least until multiple war pairings start posting full alliance rosters).

Yeah.

, if a team is AVERAGING 400 more power per player, that’s a strong indicator that they MAY be a stronger overall team.

You’re right that we “just can’t tell.” The reason we can’t tell is because we have no data.

If we have data, we can attempt to establish strong correlations and see if there are any trends.

There’s literally nothing to lose by tracking this.

Your argument seems to boil down to “it might not prove anything and I already believe it won’t work so let’s not do it.”

edited for attempted niceness

Incorrect. Already explained:

We already know that the two alliances have near identical Titan scores. If one has a higher score for their aggregate raid defence teams’ power, then the most likely (though not sole possible) reason for the match is that their benches are markedly weaker than their opponents’.

Nope. Already explained:

But collecting alliances’ aggregate raid defence teams’ power will provide no useful measure of the relative strength and activity of the two teams. What’s the best proxy we have? Titan Score (shock, horror!) because it accounts for top team strength, bench strength, player activity and understanding of the game (at least where hitting Titans is concerned).

And:

Perhaps you are arguing that you lost to three alliances whose raid defences had greater aggregate power than yours, because their raid defences had greater aggregate power than yours? But you haven’t provided any evidence of this. Firstly, not to be cliché, but correlation is not causation. Secondly, we know nothing about your bench depth vs their bench depth - if you have terrible benches then that is far more likely to be a cause of your repeated defeats than out powered raid defence teams. And finally, of course, we don’t know how you (or they) are working as alliances. Perhaps you are exceptionally skilled at Titans and exceptionally unskilled at wars. We just can’t tell.

Nuh-uh. Already explained:

So while more data points are usually good, this particular one has so little relevance to the considerations at hand, and such a high probability of being extremely misleading, that I wouldn’t waste my time with it. Any time it was used it would need to be caveated to death, so there’d not be much point to it.

And:

Which is why, of course, we would be silly to waste time on that data - it can never prove anything, because so much is unknown and unknowable (at least until multiple war pairings start posting full alliance rosters).

While it simplifies the process of me replying when a post raises points I have already addressed in some detail, it does leave me wondering about the poster’s desire to read and comprehend. Perhaps I should include a TL/DR in each comment?

TL/DR: In absolutely ideal circumstances (which we do not have) collecting aggregate alliance raid defence power data would tell us absolutely nothing about the relative alliance strength of matched war pairs. The best we could hope for is a complete lack of any trend. Far more likely (randomness being what it is) is that we would uncover an entirely spurious correlation that would lead us to erroneous conclusions.

Edited for niceness.

Just fought an alliance using the “Half of their defense teams are camouflage trap teams- ONE good healer and FOUR 1.1 heroes ( 3*- 5* ).”

5 of Top 5 and 9 of Top 10 scores were our alliance with six players scoring 400+ points and final score 4432 (ours) to 1801 (enemy).

Nice point!

For reference, I never use 2*+ battle items against titans since I don’t need to versus 5* titans.

Our alliance actually has players who don’t hit titans ( don’t look at me like that, not the leader here, plus recruiting is tough ) but we have successfully nagged into hitting during war.

2 Likes

I am reading what you’re saying and comprehending it.

Your suggesting I lack reaching and comprehension is a personal attack - due to our difference in opinion. Let’s just note that specifically right from the start here.

Your suggestion is that MAYBE my alliance has a deeper bench because our Titan score is equal with our enemy’s Titan score, AND that is why we got matched with them?

When you say bench, I am thinking you mean each player’s hero roster? Or are you speaking of clan activity levels?

If it’s the former, why would having 6x5 deep in heroes positively impact our relative Titan scores?

If you mean the latter - that’s why we should look at total number of attacks utilized.

If there is a PROBABLE outlier as far as matching goes, and we are trying to determine if this strategy works, using these numbers to exclude those outliers is valuable for everyone.

If Fabian works for a group and they’re consistently matched against weaker teams, we should be able to look at that and realize it isn’t an indicator of Fabian effectiveness.

Your assertion that power doesn’t matter in any way is absurd.

Go and raid right now. Find someone with 400 power less than you. Fight the raid. Notice if it is easy or hard. Reroll until you find the absolute most challenging setup of heroes imaginable even. But still 400 less.

Then go find a 400 stronger team. Same thing. Find the easiest team. Compare the difficulty.

Now imagine that multiplied across an alliance war. Because that was our last matchup. (Noting that as power increases a 400 power variance is less meaningful.)

So this is valuable and important to consider. We disagree on this point. And we also disagree about the risk and reward of collecting this information. Fantastic.

Raid and AW have completely different strategy, both for setting the def team and also in how do you attack as off.

I think another thing worth thinking is, AW is NOT match of ideally equal alliances. It is a match based on SOME criteria. And then you got someone with better strategy than your team. Because choosing your def team is also important strategy.

Don’t forget that boards are crucial factor as well, if you get bad one, you just need to accept that. For each one, no matter how many of bad ones are in a row.

In AW we’re matched by some arbitrary criteria so that it isn’t completely random and that possibility of interesting war is there. No sense in matching hyper strong with pathetic, no fun for either side.

But we’re not matched by how good strategist we are as a team, how organised we are in the attack and so on.

Also, don’t judge someone’s strength of whole hero roster by shown def team.

My off depends who I’m attacking. I adjust for every opponent the best I can. My deff TS is around 2400, with double or triple colors and luck, I can put down team od 2700 in raids. However, not in AW. Because board is harder, not easy to make it combo (titan board is easiest), there are arrows, and I need three flags for team of 2700… if I manage… more than 2 healers or some other tricky bastard and it’s really struggle.

Also in raids, if I go for weaker team with less trophies, board will be harder than when I go against stronger team with more trophies than me. Even with same heroes between those two players, just of different strength.

All this ‘board is harder’ is subjective, by how much I need to think to do some double move, or dig for my doubled color… Titan boards can be hard or good, raid is harder if I’m stronger, AW is mostly hard to start it… not all board are the same.

And AW were designed to be hard.

But you know, everything mentioned is that strategic part of this game.

If you insist on rainbow team, if each of you is hitting on their own, if you don’t dare to double color, no risk… then it’s definitely different result than in opposite situations.

But above all, other team is also team of human beings who are trying their strategy. You didn’t lose because of their def, but because of their off. Off collects point, def just distributes them.

3 Likes

No of course not…

But if someone has a defense team value of 3500 power, then you KNOW they’re AT LEAST able to field ONE ~3500 team.

And if they have a 2000, maybe they’re hiding a 3500 power roster… and there are some really oddball reasons they might do this. But we know they’ve got at least 2000.

The number isn’t meaningless at any rate. It’s an indicator of the floor of the given player, even if it’s possible their ceiling is a lot higher.

Each time we collect data from a war we are going to have 40-60 ‘data points’ and my expectation is that oddball setups would be normalized over time as we collect more and more information on what people are seeing.

And ultimately, the best we can hope for is some more information on what sort of strategy is or isn’t viable from a more objective perspective.

Is it worth it? I’m not sure. My gut says Fabian isnt effective anyway.

But I do love comparing numbers … :slight_smile:

It absolutely is not. This is the way you give the impression of being thin-skinned - you take an attempt to assume the best of you (that you perhaps choose not to read or comprehend) and get all sensitive about it. How else ought I to interpret your apparent failure to understand anything I’ve written? There are other ways, I’m sure, but they’re none too charitable.

I’m sorry: this is so obvious I can’t face explaining it. I’ll leave you to read the forum and learn a bit.

Thus demonstrating that you continue not to understand anything I have written. Aggregate raid defence power provides no useful indication of relative team power for matched war pairs. I guess I could just do more pasting of previous explanations for you, but you seem to be getting agitated again. I’ll leave you to reread it all.

If you genuinely want to have a civil discussion about this I’d be delighted to do so. But if you want to return with no intention of understanding the points I’ve repeatedly made (pretty clearly, I think) then I’m not sure I can be bothered. Let me know.

So, the BEST CASE SCENARIO is that I am stupid?

That’s your assertion, and how you’re proving that what you said isn’t a personal attack?

THAT is what you’re gonna go with?

How about this scenario:

We disagree and have different opinions.

Because: I think you’re wrong. I don’t think it’s a comprehension issue or an intelligence issue. I think it’s your perspective or, or maybe my perspective that results in our difference of opinion.

But I am not calling you stupid as a result.

Now again - read what I said: go and fight someone with 400 less power and then 400 more power. Because that WAS the matchup we had in our last war… it happened. I don’t care how or why. I’m not complaining. But if we are looking at success or failure for a given strategy, that would make sense to exclude that war.

Your assertion that 6x5 deep being of more benefit as obvious is fun. Go ahead and explain it. I am expecting As many holes as a piece of Swiss, that was shot with a gun a real lot. And other things that make holes. Because instead of providing holey reasoning for your holey assertion you instead suggest something is obvious because your explanation of it would be too poor.

That’s my take.

You have said NOTHING to prove this objectively. Only that you subjectively believe it.

I will require NUMBERS to provide this, you know. Facts? Are you familiar with this?

Guess what one way of getting numbers is? For us to compile them as a group over time.

War result as a function of aggregate team power would be a very interesting statistic to me.

I think I see where you may have flaws in logic.

First, seeing someone’s def, means that they can construct one such team.

Most often that’s also their strongest team, so it’s a ceiling not a floor.

Second, having decent 3* heroes, 6 in each color, give you opportunity to create a lot of titan damage by stacking colors strong for the titan, compared to having single rainbow strong team.

Third, be good at playing match 3 is somehow often forgotten as important factor. Because it’s not just match 3 or more, it’s also which side, which hero to attack first with your heroes’ specials…

And all your team has to be good in everything to score. And opponent’s team had to be less good.

But main part - off gains points, def just distributes them.

Your team chooses who to attack. If you all opt for smaller opponent with good def, no wonder that you need more power and you have nothing left for stronger opponents.

I have no clue how long you’re here, playing and reading, but do read posts from this year in game strategy, and other famous topics.

I honestly think that you had seriously bad strategy if you couldn’t win team where each player had def 400 points lower then your off teams…
And they had better strategy than you.

Again, off wins the war, not def. And you’re in control of your off, not your opponents.

2 Likes

A ceiling in that it is likely the strongest team they can field. A floor in the sense that they could have a much deeper bench.

I maintain that a person fielding a 3000 defense team is FREQUENTLY or OFTEN (>75% of the time) going to have a significantly stronger bench than someone fielding a 2600 defense team.

As you approach the higher and lower levels perhaps this becomes less true.

… and yes. The team with an average of 400 more power did win. It was the other team. I think that is to be expected. You seem to agree. I think it’s an important stat to track, even if it isn’t the be all end all.

Edit: wanted to say thank you. Your post seemed very logical and well presented. I appreciate that. Even if I still mostly disagree with your premise :slight_smile: (that’s kinda my thing?)

I’m just going to ignore all the thin-skimmed, ‘woe is me’ victimhood - because I’m a better person. :grinning: (But you should stop reading insults into everything you see - I’m sure it’s bad for you.)

So where were we? This is the bit you were struggling with:

Aggregate raid defence power provides no useful indication of relative team power for matched war pairs.

And this is why it’s clear there’s something you are not understanding, for some reason. Because here is what I wrote above:

"We already know that the two alliances have near identical Titan scores. If one has a higher score for their aggregate raid defence teams’ power, then the most likely (though not sole possible) reason for the match is that their benches are markedly weaker than their opponents’.

We can’t measure this in any way, but it’s an offsetting factor that ought always to be taken into account. And when one thing is measurable and another, arguably more important thing (for wars, anyway) cannot be measured, there’s a tendency to overweight the former and underweight the latter.

Moreover, of course, simply summing raid defence strength fails to account for intra-alliance strength distribution, a far more important factor than alliance aggregate strength, for wars. And we already know that these alliance fare equally well, more or less, where Titans are concerned."

How else can we understand that aggregate raid defence power can an awful indicator or actual team power? Let’s see…

[…]maybe they have a whole bunch of inactive players who have effectively retired but left strong raid defences in place. Or maybe their power distribution has a heavy left skew while yours is normally distributed or right skewed.

And you want numbers? Cool, but:

We’re never likely to acquire data with sufficient granularity to identify precisely what is happening. It would take quite a lot, in most circumstances, even to say with confidence which alliance is better suited to wars.

Now if you don’t understand this stuff, that’s one thing and I can be sympathetic. But if you’re just pretending it doesn’t exist and not addressing it, despite my repeated posting of it in direct answer to you, then I’m less sympathetic.

That’s not my assertion, by the way: that’s your misinterpretation. (Accidental? Let’s say so.) Here’s what I wrote:

We already know that the two alliances have near identical Titan scores. If one has a higher score for their aggregate raid defence teams’ power, then the most likely (though not sole possible) reason for the match is that their benches are markedly weaker than their opponents’.

This has been discussed a million times in the forum. I should really just link to a thread but that seems too hard right now. Cliff Notes version:

  • High Titan scores are achieved mainly by doubling, tripling or even quadrupling up on the weak colour.

  • To do this effectively, you gotta have redundancies - ideally double or triple redundancies - in each colour.

  • This implies that 20 heroes is the optimal number for Titan damage. (I think that’s unnecessary, personally - I think 15 does perfectly well.)

So if we see an alliance full of strong raid defences generating poor Titan scores, there’s a good chance they lack bench depth (a very common characteristic of new players, who tend to throw all their resources at one team - which might well have high aggregate power but be awful in practice.) Despite their apparently high powered defences, they’re a weak alliance (in war terms).

That’s not the only possibility, of course (as I’ve set out above and you’ve ignored). The alliance might have a bunch of members who are not particularly active. (This also means they are weak in war terms, and is another reason to disregard aggregate raid defence power as a metric.) Items, blah, blah, blah, I’ve already said it all.

Disagreeing is fine - it’s good, even. But when I go to the trouble to engage seriously with your contention, set out many of the problems with it in considerable detail, repeatedly, then you ignore all that for post after post, failing to address it while insisting you understand it, I’m left confused.

This is particularly so when you decide to get the hump and complain about how hurt you feel: I keep inviting you to consider and respond to these ideas, as I have taken the time to respond to yours. You continue not to do so.

So I guess that you don’t really want to have a serious conversation. That’s cool, too. But it’s a terrible way to make a case for something if you really believe it.

1 Like

I try to respond to your points in each paragraph / each subject:

YOU: “Most likely (though not sole possible) reason for the match is that their benches are markedly weaker than their opponents. “
REPLY: This is pure speculation. This may be true at higher levels of play. Perhaps you’re unable to account for the experience and variance of an alliance in with a ~30,000 titan score. I’m seeing a wide range of teams with this score. One alliance we were matched against had 9 members, for example. I only started tracking Power recently.
YOU: We can’t measure [benches] in any way, and since we can not measure benches, we shouldn’t measure overall power, because [brobb doesn’t think it is important in any context].
REPLY: Again, this is your opinion. I don’t agree with it. My expectation is that at your level of play the determining factor in wars has zero to do with aggregate power and more to do with bench depth. I’m guessing however, that at your level the aggregate alliance power is probably VERY close, and as a result wouldn’t stand out in our data collection anyway.
YOU: [MAYBE they have inactive players. Maybe they have a left or right or middle skew].

REPLY: Yes. MAYBE. With enough data points outliers would stand out and could be excluded. Regarding inactive players, this is why it makes sense to also track total # of attacks in each war. Ideally, we would track # attacks * raid defense power, but this isn’t realistic. Over time and with a larger sample size, I think this would level out at any rate. Keeping in mind that not just stronger teams have inactive players, they would occur on both sides probably at roughly equivalent rates. Same things with left shifts and right shifts. They aren’t exclusive to any particular alliance.

YOU: Cliff notes on why redundancies in titan scores are optimal: 1 doubling / tripling etc. 2 – redundancies in each color 3 – 15-20 heroes is optimal # for titans.
strong raid defenses generating poor titan scores = lack back depth = new players = poor war team.
Or they may have a bunch of inactive players. Or they don’t use items.
REPLY: YESSSS. EXACTLY. New players have poor bench depth and generate poorer titan scores. Guess what kind of alliance I’m in? A generally low level alliance where maybe people don’t understand or practice these principles. GUESS what kind of alliances I’m routinely matched against? There are many possible reasons for why we succeed and fail, and outside of gathering data there is nothing you can say to prove to me that overall power is not one of them.
So please try to step outside of your castle and consider what it’s like for other players. Because my experience, and that of a LOT of people posting here varies from yours. Which is fine of course. But when you’re gonna use YOUR experience to try and define mine, we have a problem. I am willing to consider the possibility, so you should be willing as well. We run these numbers, chances are pretty good they support that Fabian doesn’t work. They probably also show that at high level gameplay, team power does NOT matter, further supporting what you’re saying.
But maybe in the 40-80,000 range, they show that team power is actually a VERY strong predictor of who will win. (Maybe not, too. That’s how data collection works.)
You’re right that correlation isn’t causation. But if we can establish strong correlation between alliance power within a specific range predicting the end result, then there is value in gathering these numbers.
The bottom line is that I disagree with you that there is a risk of this data somehow misleading everyone involved. I see what you’re saying. I don’t believe it’s applicable to all cases.
And, as far as my thin skin goes. I’m not crying in a corner over your words, just trying to point out to you what you did. When you tell someone that they can’t read or comprehend, you’re calling them stupid. That isn’t a stretch, in my mind anyway. I feel that discussion should proceed under the premise that those involved are reading (and comprehending) others’ statements and responding to what they feel is pertinent and worthy of being responded to. (and if we feel something was missed that really did warrant response, we post it again and bring attention to it. We don’t accuse the other person of being stupid. That’s how I think it should work, anyway.)

2 Likes

Thank you for delivering a considered response. I hope we can talk about this stuff seriously now that you’re engaging.

Allow me to respond properly later on, though I want to flag as a preliminary (but overarching) matter that you have referred to “overall power” and “team power” in your comments, again. We have no way of measuring overall power or team power.

All we can measure is raid defence power, which we know is an awful measure of actual power. Fundamentally, that’s the point you seem to keep missing.