…or make this war matching test on a separate offline backup
An idea to do the testing in live game, but do not affect the wars in live games immediatelly:
Do a match-making test maybe 1 day (or x hours) earlier based on new logic for half of the alliances, and the new logic for the other half, and collect the data.
And after that do a real matchmaking based on the old rules, which we be used for the next war.
So, you gonna announce the next test in the live version?
I mean with a in-game message and not just something here on the forum…
Or you keep doing it secretly and annoying your customers, if the matchmaking still isn’t working?
That comparison is a little much.
I think performing a “dry-run” of the new matchmaking is a fair point.
However, the challenge is this: how do you evaluate if the matches were good or not? What is a simple heuristic to predict the outcome of a given war with high certainty? We could compare e.g. hero powers but the whole reason for this project is that the hero powers alone are not a good enough indicator.
In fact, if we had such a heuristic and it was better than what we use for matchmaking, shouldn’t we be using it for the matchmaking instead of the current model?
Please note that there was no bug in our previous matchmaking attempt. It worked as designed - unfortunately, the results were not as good as we hoped.
12 Likes
We’ll discuss this tomorrow. I am sure that I’ll post about it on the forums. I am not sure if we’ll send an in-game message.
9 Likes
This is a bad thing. I mean, how many players visit the forum? 5%? Maybe 10%?
On the other hand, I have another example of how your new / old war system is really bad.
We’re facing an enemy with 13 teams of 4000+ team power while we have only 6 teams of 4000+ team power. How is this fair?! How can we defeat them? Not all the alliances / players are willing to invest $10k to get a good roaster for each player.
1 Like
We often come up against stronger defence than ours. Last war we faced a team with 10 out of 17 with 45xx defence where we have 6. But we try and defeat whatever opponents we are matched against.
If we win these match ups it’s Tequila shots all round, if we lose on to the next one
A short update: We will run the next test on the 29th of September, so this weekend’s war will be using the old matchmaking for every alliance.
We’ll post a separate announcement about the test on the forums on Monday. We do not plan to send an in-game message.
8 Likes
I do find it strange you don’t use the in game message platform to keep all your customers informed of developments and that your running a live test. Perhaps you are worried that you will receive another influx of complaints about war matching, Taking the mushroom approach will only anger players more.
5 Likes
I would also vote for an in-game message.
Too many players are upset and angry and too few of the complete player base know this thread.
2 Likes
Errmmmm, it has been released. We have all been living it. We are testing it for you. Yes, it was a mistake.
I understand you’re trying to improve the playing experience. I appreciate that…But screwing everyone over with the lights off, is still screwing everyone over.
3 Likes
The response to this and the apparent lack of regard for the customer base continues to astound me.
On one hand it’s good that there is at least some engagement from the dev level here to explain, but the tone deafness of the response when it comes with respect to legitimate complaints and concerns of the customer base is alienating to say the least. I’m pretty much at a loss at this point. 
3 Likes
Fortunately for me, i (still) haven’t come up with this bad mismatch since you tried to implement the new matchmaking method, but as already has been suggested, if you are going to test it on live game, playing with people’s patience and efforts to build successful alliance that tries to climb the ladder, you better assign a WIN for both alliances, no matter the result.
I mean what sort of match is 150k alliance (with 60k war score) against 330k alliance (with 138k war score)? Obviously bad mismatch, that probably the obvious winner will not enjoy as well. At least make sure to compensate the weaker alliances when you are testing this method on poor people back.
3 Likes
Pretty bad decision.
But was expected as your company seems to not care about their customers.
Edit:
You then at least grant 5 points to everyone, regardless if it’s a win or loss?
Or you still keep messing up those, who get a war, with a messed up matchmaking?
Probably the most informative posts (SGG side) i read until now in 3 years.
Thank you.
I will not complaint about these tests, as it is your job and not mine.
Lecture you guys about this seems a bit silly.
Mistakes are part of improvement process.
And you are not bound for sure to let us know every single decision you made.
Several people were asking to change war matching alghoritm.
They’re trying to do it. Stop.
They should apologize for many things, but not this.
9 Likes
People are getting shafted badly with completely unfair war match-ups in this test with nothing to show for it.
Which is 100 percent preventable with even a modicum of respect for your customers.
As others have already posited, just give everyone in the test groups 5 war points, win or lose. That literally costs SGG nothing and it’s not all difficult to implement. It’s not even giving players any specific items. It’s just doing what’s right by their customers, as they test product changes to improve their product their trying to sell to those very same customers.
This is the antithesis of good customer service.
2 Likes
I was totally annihilated this last war and had the greatest mismatch against me as well, still i don’t feel entitled to receive anything for that.
Yes, they could let us know before hand, but i think they were genuinely think it was not necessary if all went well.
It did not. Shame.
But there are good intentions behind it and an attempt to give to us players what we want.
I can sacrifice 5 war points for that.
I usually am on the front line to complaint, and this change even screw my shuffle routine, but i really apreciate the effort here and feel kinda ungrateful to shout on them.
2 Likes
Making sure people aren’t getting shafted on unfair war match-ups isn’t being “entitled” to anything though. Like others have mentioned simply eliminating the cost of losing war so that both teams get 5 war points isn’t giving anyone anything. It’s just eliminating the cost to the teams during these tests. Which again, should be very simple to implement and there’s not a single solitary good reason for not doing it.
The pathway to hell(and in this case customers leaving for getting shafted during this test for no good reason) is paved on good intentions.
They’re doing it so they can try to sell a better product to you. I don’t really think asking them to eliminate the cost to the players during this test is even close to unreasonable.
It’s really no different than forcing players in your beta environment to pay to test the new product. I don’t know, to me that is just not good practice at all, regardless if it’s for the “betterment” of the game.
Come on man. You know the complaints cant stop. Even when they’re trying to fix stuff.
People are still having nightmares about Telly tanks, 20% defence bonuses, and not getting a 5* every time they drop a few dollars.
And it seems like in these threads there’s always a clamor for “free” stuff, whether it be emblems, 5 war chest points, gems, etc etc etc. That gets pretty nauseating too.
Kudos for the post though.
5 Likes