Please register for Total War Access to use the forums. If you're an existing user, your forum details will be merged with Total War Access if you register with the same email or username. For more information please read our FAQ’s here.
I don't understand why people want a reset. A year from now the leaderboard will have the exact same problems as now. A rolling average with a separate leaderboard for All Time and Last Quarter (or whatever periods: I'd personally add Daily and Weekly too) only offers benefits: - Currently high ranked individuals are happy to preserve their hard-earned rank. I don't like kkwon being op top with drawkiting either, but Crusader and Lotus deserved their spots, for example. - A rolling average mechanism requires no work on CA's part to decide when resets take place or to implement them. - People interested in who's currently best still learn little from the leaderboard compared to something like a Best of the Month. - A reset now will heavily bias the results towards who can figure out how to abusive the latest DLC units the fastest.
Well that's precisely why a more regular set of resets would be better most high ranks support it
But what is the advantage of manual resets over a rolling average score? I can think of none, only downsides and more work for CA.
A rolling average keeps account if previous scores offering little incentive for new players. It doesn't address the smurfing and hoarding and inactive accounts nor the massive changes since the first release where 19 chameleon or horse cav builds were a thing.
A manual reset should hardly be arduous unless someone from Ca can explain the manhours required.
If the issue is just a loss of total wins it's barely an issue as many with large amounts of wins Inc myself have already said.
A rolling average keeps account if previous scores offering little incentive for new players. It doesn't address the smurfing and hoarding and inactive accounts nor the massive changes since the first release where 19 chameleon or horse cav builds were a thing.
A manual reset should hardly be arduous unless someone from Ca can explain the manhours required.
If the issue is just a loss of total wins it's barely an issue as many with large amounts of wins Inc myself have already said.
That's not how a rolling average works. A rolling average is simply the average of the last X units of time. So if we want a leaderboard with a Best of Last Quarter, it would only take the data from the last 3 months to compute each player's rank. So kkwon would remain #1 on the All-Time board, but he wouldn't even be on the Best of Last Quarter board unless he started playing again.
A rolling average keeps account if previous scores offering little incentive for new players. It doesn't address the smurfing and hoarding and inactive accounts nor the massive changes since the first release where 19 chameleon or horse cav builds were a thing.
A manual reset should hardly be arduous unless someone from Ca can explain the manhours required.
If the issue is just a loss of total wins it's barely an issue as many with large amounts of wins Inc myself have already said.
That's not how a rolling average works. A rolling average is simply the average of the last X units of time. So if we want a leaderboard with a Best of Last Quarter, it would only take the data from the last 3 months to compute each player's rank. So kkwon would remain #1 on the All-Time board, but he wouldn't even be on the Best of Last Quarter board unless he started playing again.
This is just a bad idea, their rank changed that day from what didn't happen that day.
They should add decay without resetting.
Imagine the leaderboard if decay was there this whole time, it would actually be accurate and updated.
The game has all the movements of units logged, because it needs these for the replay. With this info it should be fairly straightforward to design an anti-draw kiting algorithm. For instance, making average movements away from the enemy in the latter stages of the battle is a telltale sign and very simple mathematically.
Bit late to the party here but I just want to say this is a really cool idea and seems like it would be quite easy to implement. Something like:
If and only if a match ends in the draw, begin the checking algorithm.
Calculate mean position of each player's army every 10s (just taking the mean over the centre pos of all units).
If in the last stage of the battle before the timer ran out (5 mins, 1 min, whatever works best) the centre pos of army B was consistently moving away from the centre pos of army A and the centre pos of army A was consistently moving towards the centre pos of army B, hit player B with a drawkiting penalty.
It should be very hard to get a false positive, because you assess it after the draw happens. It might be possible to game it by making your army run away in a strange dance rather than a straight line, but you don't need to tell players the parameters of the algorithm (and also not running away from their opponent on a direct vector should make it a lot more difficult for draw kiters to actually draw kite). And if there were any route to a false positive you could do it the same way as the drop hack penalty where you need to get multiple strikes before you get hit with it.
A reset would also mean wiping the number of wins, so do take that into account as well when asking for a reset.
People don't recognize the Leaderboard currently because there are no resets. Hearthstone does this monthly to maintain legitimacy and promote competition.
A rolling average keeps account if previous scores offering little incentive for new players. It doesn't address the smurfing and hoarding and inactive accounts nor the massive changes since the first release where 19 chameleon or horse cav builds were a thing.
A manual reset should hardly be arduous unless someone from Ca can explain the manhours required.
If the issue is just a loss of total wins it's barely an issue as many with large amounts of wins Inc myself have already said.
That's not how a rolling average works. A rolling average is simply the average of the last X units of time. So if we want a leaderboard with a Best of Last Quarter, it would only take the data from the last 3 months to compute each player's rank. So kkwon would remain #1 on the All-Time board, but he wouldn't even be on the Best of Last Quarter board unless he started playing again.
This is just a bad idea, their rank changed that day from what didn't happen that day.
They should add decay without resetting.
Imagine the leaderboard if decay was there this whole time, it would actually be accurate and updated.
It seems people here are not very good at math/programming...
"their rank changed that day from what didn't happen that day."
Ironically, this is true for what you propose: decay. You're proposing that rank decays when someone doesn't do anything. With a rolling average their rank is updated each day based on their performance that day and it stops factoring in their performance from 3 months ago. This makes it reflect someone's current skill.
Decay does not result in accurate ranks at all. For illustration, Ondjage would have a terrible rank by now, but he's still pretty good as evidenced by several short comebacks. Same for Vikingcat. A mediocre player that plays daily could have a better rank due to not having decay. A rolling average fixes this problem by giving people that don't play anymore no rank at all. That's accurate, because we have no idea how good they are at this moment.
Not to mention the implementation of a rolling average is very straightforward, but decay is extremely difficult. Decay can only be implemented by corroborating rankings with a different metric that's used as the benchmark to validate the decay rate. Otherwise it's just an arbitrary decision.
Wonders whats CA verdict on this. Its THE golden opportunity to reset it. A full unit cap rework, a massive dlc rebalance.
Hopefully they reset it, should have widespread community support on this, even on reddit and other sources. Surely there r other mp players wanting this
Comments
Once your army is shattered, all your unbreakable units should rampage.
this can't really be abused, makes sense from lorewise perspective for many units and would help out with some draw kitting.
I once got draw kitted by Dwarf Slayers (!!!!!) because during the course of the battle all my strong fast units were killed off.
- Report
2 · Disagree Agree- Report
0 · Disagree AgreeIt doesn't address the smurfing and hoarding and inactive accounts nor the massive changes since the first release where 19 chameleon or horse cav builds were a thing.
A manual reset should hardly be arduous unless someone from Ca can explain the manhours required.
If the issue is just a loss of total wins it's barely an issue as many with large amounts of wins Inc myself have already said.
- Report
0 · Disagree Agree- Report
0 · Disagree AgreeThey should add decay without resetting.
Imagine the leaderboard if decay was there this whole time, it would actually be accurate and updated.
- Report
1 · Disagree Agree- If and only if a match ends in the draw, begin the checking algorithm.
- Calculate mean position of each player's army every 10s (just taking the mean over the centre pos of all units).
- If in the last stage of the battle before the timer ran out (5 mins, 1 min, whatever works best) the centre pos of army B was consistently moving away from the centre pos of army A and the centre pos of army A was consistently moving towards the centre pos of army B, hit player B with a drawkiting penalty.
It should be very hard to get a false positive, because you assess it after the draw happens. It might be possible to game it by making your army run away in a strange dance rather than a straight line, but you don't need to tell players the parameters of the algorithm (and also not running away from their opponent on a direct vector should make it a lot more difficult for draw kiters to actually draw kite). And if there were any route to a false positive you could do it the same way as the drop hack penalty where you need to get multiple strikes before you get hit with it.Get on, Kroq-Gar, we're going shopping
- Report
1 · Disagree Agree- Report
4 · Disagree Agree"their rank changed that day from what didn't happen that day."
Ironically, this is true for what you propose: decay. You're proposing that rank decays when someone doesn't do anything. With a rolling average their rank is updated each day based on their performance that day and it stops factoring in their performance from 3 months ago. This makes it reflect someone's current skill.
Decay does not result in accurate ranks at all. For illustration, Ondjage would have a terrible rank by now, but he's still pretty good as evidenced by several short comebacks. Same for Vikingcat. A mediocre player that plays daily could have a better rank due to not having decay. A rolling average fixes this problem by giving people that don't play anymore no rank at all. That's accurate, because we have no idea how good they are at this moment.
Not to mention the implementation of a rolling average is very straightforward, but decay is extremely difficult. Decay can only be implemented by corroborating rankings with a different metric that's used as the benchmark to validate the decay rate. Otherwise it's just an arbitrary decision.
- Report
0 · Disagree AgreeHopefully they reset it, should have widespread community support on this, even on reddit and other sources. Surely there r other mp players wanting this
Top #3 Leaderboard on Warhammer Totalwar.
- Report
2 · Disagree Agree