What are you referencing?
Printable View
This is detailed statistical overview (click to see bigger image) of our match up in our last war. There is only a difference of 64,000 weight between clans in our favor but it is all erased with their offensive level. The figures exclude walls, hidden traps and resource buildings. They have 6 nines and we have 5 nines under offense level column (Next to TH level). The TH11 has Level 4 miners that 3 starred our top two bases. If that TH11 was not in the line up we would have had a chance of winning. This slight advantage of having level 4 miners tip the war in their favor. Their number one had inferno towers and our number one didn't have them which may have played a factor. We lost this war 30 - 27 we did have 3 90%+ 2 stars. This is what engineering is all about having better offense than your opponent and just enough defense to win. Our next step is to make sure all bases have everything built and exclude any .5 bases.
https://s26.postimg.org/92l1cjt89/Wa...ics_Forums.jpg
Our last war (ended 10 September) was a 15 v. 15. And problematitic on the matchup.
Us (no engineered bases)
7 TH9s
8 TH8s - 2 had th7 level troops and defences still
Them
4 Th10 (3 were engineered) The top base had 3 max xbows but no infernos yet, but level 3 miners.
7 TH9s none appeared engineered - mid to max
4 TH8s all mid to max
Normally I'd chalk this up to the low level heroes on their end, nothing above teens, but only our #1 had heroes over 20 20/25 heros and was maxed, anyway still not a fair fight. Improved I'm not sure, that was a horrid match we probably never had a chance to win.
This would all make sense if we knew, whether we had to engineer to make a fair match. I don't care about how you play, but isn't the point to better match playing styles and hopefully make the fights fair. It seems more confusing and a thread like this gets so long and contains more griping than useful data.
Thanks for trying SC and keep at it.
Not sure if any adjustments were made but last 7 wars we did the same 10 bases
First 5 were terrible matchups
Last 2 were what we consider fair 1+ th11 or 10 up top but fair for both teams.
I hope what I'm seeing is correct.
Last war and this war either team can win comes down to attacking and defending
How I would design the matchmaker - rank attackers and defenders on each team. Use statistics to predict number of stars against defence for each attack against each possible defence. (this statistic would rely on what the strongest troops, or combinations of troops are available, rather than a simple addition) If three stars is probable - choose the weakest attack that gives two stars. If two stars is the best - select the weakest attack that gives three stars. Move down the defence from strongest to weakest until no defenders are left. Determine expected number of stars won for each team based on this. (note two attacks per player means that there will be more attacks available than defenders). Repeat for second team. If the number of stars predicted is within one or so (maybe two) then it's a fair match.
Probably needs a small adjustment for the extra attack strength left over, but basically this could give a pretty good start on preventing the offence/defence mismatch.
Seems so easy....
Our top TH10 draws a TH11 every war now. I'm starting to see a lot of engineered clans again. Have they done a recent update on MM?