WoL vs. Math

[JewellRavenlock]

Random posted this idea here:

WoL is a poor skill ranking and progression system. It rewards streaks over skillful performances, which is problematic for a game as unpredictable as dueling is. A 5-2 win over a fellow Glass is canceled out by a 4-5 loss to an Emerald using mods, leading to a wash progression-wise from 2+ hours of effort. It creates perverse incentives such as targeting lower skilled players and applying the mechanical advantage of mods against lower-ranked players to ensure more predictable wins. The majority of rules changes in the last few years, as much as I approve and support them, are ultimately band-aid fixes to this systemic problem. I think the WoL progression combined with having to wait for a critical mass is the reason regulation dueling is less popular than the leagues and events.

Here's a simple ranking system that would better reflect dueling skill and inject more volatility into rank progression. You gain or lose ranking points based on the difference in scores at the end of a duel. If we define 5-3 as a "normal" win, then we simply multiply the rank requirements and everyone's existing WoL by 2 to convert to the new system. For example, it takes 30 points to reach the top rank instead of 15 wins. Negative scores are zeroed out weekly as they are now. With this system big wins are richly rewarded and narrow losses don't erase as much progress, making progression more exciting and in line with the player's performance. It also makes rank promotion and delegation more common, which supports the goals of the WoL system.

An additional option is to reduce the loser's point loss by 1 for each mod the winner used in excess of the loser's mod limit, with a maximum of -1. This allows higher-ranked players to use their full range of mods without unduly punishing the lower-ranked player, which helps the lower-ranked player learn mod strategy. The IC explanation is that your character learns more from exposure to high-skill moves in a duel.

Does this make sense as something to try in one of the sports? A new sport? A tournament?

People wanted to discuss it so here's a place to do so!

[Lirssa Sarengrave]

I'm not extremely fluent in game mechanics, but reading this it looks like a logical and positive way for progression, including IC storywise. Just for example, when I lose (except a recent shut out) I tend to lose by a slim margin, but my record is rather abysmal. It would also look better when someone like me manages to have a good night and win a baronial title, but as it stands now it looks like I can't hit the broad side of a barn just WoL.

I am looking forward to seeing further discussion on it!

[Delahada]

Yes, yes, yes! I read this and LOVE the idea. I actually even kind of understand it, which is amazing.

[JewellRavenlock]

Yes, yes, yes! I read this and LOVE the idea. I actually even kind of understand it, which is amazing.

That was my thought after reading it! "Wait.. there is math but I am understanding? Whaaaaat?"

Is there a way we could do a trial of this? Maybe in a tournament? Maybe just in a temporary, trial, separate set of standings?

[DemiBob]

I don't understand the advantages of complicating rank progression. Can someone explain to me the benefits?

[Kheldar]

I don't understand the advantages of complicating rank progression. Can someone explain to me the benefits?

This is not to shoot down the idea. I haven't spent much time thinking about it yet, but I'm in the same boat.


The initial post talks about how WoL causes people to prey on lower ranked duelers, but this wouldn't be any different, and it potentially has the effect of making it harder for new duelers. The motivation to fight newer duelers and get big wins early is still there.

Yes there is the scenario where you net something after a 5-2 win and a 4-5 loss, but there is also the situation where you get crushed in one match and wipe out multiple wins at once.

[Sylus Kurgen]

I don't understand the advantages of complicating rank progression. Can someone explain to me the benefits?

This is not to shoot down the idea. I haven't spent much time thinking about it yet, but I'm in the same boat.


The initial post talks about how WoL causes people to prey on lower ranked duelers, but this wouldn't be any different, and it potentially has the effect of making it harder for new duelers. The motivation to fight newer duelers and get big wins early is still there.

Yes there is the scenario where you net something after a 5-2 win and a 4-5 loss, but there is also the situation where you get crushed in one match and wipe out multiple wins at once.

This is one of those topics that looks nice on the surface but I'm not going to weigh in until I see legit numbers from a few trial tests. Not going to support re-writing the whole system because like 5 people like the idea. This puts me on the same fence as Bob and Kheld.

[Claire Gallows]

This is not to shoot down the idea. I haven't spent much time thinking about it yet, but I'm in the same boat.


The initial post talks about how WoL causes people to prey on lower ranked duelers, but this wouldn't be any different, and it potentially has the effect of making it harder for new duelers. The motivation to fight newer duelers and get big wins early is still there.

Yes there is the scenario where you net something after a 5-2 win and a 4-5 loss, but there is also the situation where you get crushed in one match and wipe out multiple wins at once.

That was my concern. As enticing as it may be to see a 4-5 loss count less when weighed against a 5-3 or better win, conversely does it sound as great if you have four 5-4 win (4 points) and then one 5-1 loss wipes them out? You just put in 5 duels, won four of them, and still break even. That would be demoralizing as heck for me.

Count me as an on the fencer.

[Spell]

This is not to shoot down the idea. I haven't spent much time thinking about it yet, but I'm in the same boat.


The initial post talks about how WoL causes people to prey on lower ranked duelers, but this wouldn't be any different, and it potentially has the effect of making it harder for new duelers. The motivation to fight newer duelers and get big wins early is still there.

Yes there is the scenario where you net something after a 5-2 win and a 4-5 loss, but there is also the situation where you get crushed in one match and wipe out multiple wins at once.

That was my concern. As enticing as it may be to see a 4-5 loss count less when weighed against a 5-3 or better win, conversely does it sound as great if you have four 5-4 win (4 points) and then one 5-1 loss wipes them out? You just put in 5 duels, won four of them, and still break even. That would be demoralizing as heck for me.

Count me as an on the fencer.

[Kalamere]

For trial purposes I think it would be interesting to see how something new works out. We have 20+ years of ranking under a WoL system though, so I have a hard time even considering the idea of one day replacing it.. so I'm not really onboard there, but as far as running an alternate un-official standings or a shadow tournament or something, I think it could be fun.

I've never been a fan of WoL. It's a system that in my opinion does more to measure how often you duel than how often you win.

That said, I'm also somewhat of the mindset that a win is a win. 5-1 or 5-4 doesn't matter, I came out on top. (team leagues being a different animal where I think point diff. works, so won't go into it here). I'd rather see something based on percentage of duels won with some consideration given to strength of schedule (eg: +/- X based on rank difference.. though probably just bonus points, since negatives could influence higher ranks to avoid lower ranks). Although, while I think it's ultimately a better measure of skill, I haven't a clue how it could be accurately used in week to week rank advancement.

If we settled on a system I could probably produce shadow standings on DZ or something. I couldn't include the mods used criteria though. I like it because I think it helps differentiate between the warlord wearing kid gloves against a commoner vs the warlord that truly went for the win; but it isn't captured in the duel reports and I don't think anyone wants to be responsible for sifting through the logs to figure it out.

[Aric Delevingne]

I think gaining rank is hard. There are only so many duels I can get in a week, and it feels like I can only progress if I win twice as many as I lose. That's not always going to happen.

Because there are only so many active people around, there are only so many duels to get in a week. And duels can take an hour or longer. I think it's a good idea to gain something for the time and effort.

When I have a chance, or if someone else wants to, it would be pretty easy to compare where I would be with that proposed system vs. where I am now. That would give some data to help base decisions on.

[Random McChanse]

I agree the idea needs a bunch of trial tests to figure out its worth. We should also analyze the archives to see how common each score is, and do an analysis on a cycle or year's results to see where everyone would stand under the new system.

Some thoughts anyways:

Upon reflection, I think the issue of similar-rank alts preying on new duelers will exist as long as we hide the account controlling the character. Even if we used an Elo system like chess, you'd still rank up faster by defeating lower skilled players by larger margins. This system makes the weekly reset a better cushion for early losses, and allows for the possibility of wiping out your losses with one big win. Compare to going 0-2, where the time investment of dueling yourself back to 1-0 is greater than simply not dueling for the rest of the week.

As for higher-rank duelists preying on lower ranks, that's what the optional rule is for. It lessens the blow for the new dueler while allowing the higher-rank to feel comfortable using their mods. Other ideas for the optional rule are to max out the loser's point change at 0 or higher instead of -1, and/or subtract from the winner's points for each excess mod they used. The latter felt a little too much like enforcing same mods, but it's an option if we want to try it.

I also agree that this system has larger emotional highs and lows. People psychologically respond better to variable wins and losses versus fixed wins and losses. For those who play poker, it's why no-limit or pot-limit betting is generally more popular than limit betting.

[DemiBob]

I agree the idea needs a bunch of trial tests to figure out its worth. We should also analyze the archives to see how common each score is, and do an analysis on a cycle or year's results to see where everyone would stand under the new system.

Some thoughts anyways:

Upon reflection, I think the issue of similar-rank alts preying on new duelers will exist as long as we hide the account controlling the character. Even if we used an Elo system like chess, you'd still rank up faster by defeating lower skilled players by larger margins. This system makes the weekly reset a better cushion for early losses, and allows for the possibility of wiping out your losses with one big win. Compare to going 0-2, where the time investment of dueling yourself back to 1-0 is greater than simply not dueling for the rest of the week.

As for higher-rank duelists preying on lower ranks, that's what the optional rule is for. It lessens the blow for the new dueler while allowing the higher-rank to feel comfortable using their mods. Other ideas for the optional rule are to max out the loser's point change at 0 or higher instead of -1, and/or subtract from the winner's points for each excess mod they used. The latter felt a little too much like enforcing same mods, but it's an option if we want to try it.

I also agree that this system has larger emotional highs and lows. People psychologically respond better to variable wins and losses versus fixed wins and losses. For those who play poker, it's why no-limit or pot-limit betting is generally more popular than limit betting.

So the advantage is higher highs and lower lows? That's just what we need - more people being super happy over wins, or super low over losses.

The fact is that better players will generally beat worse players, and any system that uses ranks or titles will encourage people to prefer duels where they have better chances of winning. I don't see how changing from a simple and easy to understand WoL system to a more complicated Points For/Points Against system will positively impact anything. I can, however, see how it would negatively impact things.

Just my two cents.

[Jake]

This system makes the weekly reset a better cushion for early losses, and allows for the possibility of wiping out your losses with one big win. Compare to going 0-2, where the time investment of dueling yourself back to 1-0 is greater than simply not dueling for the rest of the week.

You've made this suggestion a couple of times in your arguments...but what about the flip side?

What if that 1st duel of the week is a shut-out loss? 5-0

Now I have to fight to recover from a negative score, and how many duels will that take? Maybe my next set of a duels are a 5-3.5 (+1.5), 5-2.5 (+2.5), and 5-4 (+1) wins. Three duels just to break even with that opening loss. What if that 3rd duel was a loss?

If I was a commoner starting at 0, and I started the week at a -5 or -4, I'd stop right there until the reset.

Using the current WoL system, yeah, I might start the week with a loss, but it's only a single point. I only need 1 win to break even.

--------------------

For the long-time Warlords (and above) this is a non-issue. They will have such a reserve of points that they aren't likely to lose rank.

For the beginning duelist...I sort of think this system has the potential to be even more demoralizing.

[Spell]

For the long-time Warlords (and above) this is a non-issue. They will have such a reserve of points that they aren't likely to lose rank.

For the beginning duelist...I sort of think this system has the potential to be even more demoralizing.

tbh I'd only agree if there was an opt out option, because my top Warlord only has 53 WoLs and I'd like them to stay relatively high without risking mad dips due to shutouts.