Published using Google Docs
ITN - 420
Updated automatically every 5 minutes

#420 - Into the Nexus: “Ladder Land”

• Garrett • Kyle • NotParadox or Guest

Tracking Link:

https://dts.podtrac.com/redirect.mp3/

Sponsors:

Our Patrons support Into the Nexus at Patreon.com/ITN. If you enjoy the show consider supporting it with 25 cents, a dollar, or however you would like! We thank all our Patrons for their support!

Intro:  Welcome to Into the Nexus! The Heroes of the Storm podcast!

Into the Nexus is back and talking ladder! Former Heroes Balance Designer AZJackson was on Twitter this week talking about a recent article posted by Magic the Gathering creator Richard Garfield. The challenges of game design and letting things like MMR handle your players for you. We also briefly check on World of Warcraft’s new expansion leaks before taking emails.

Patreon Mention: Patreon.com/ITN

Thanks to our Patreon producers Shawn B & Mike R!

Patron Thanks Doc for keeping track

Former Heroes Balance Designer Talks Ladder

AZJackson Twitter:
What a fascinating article.  I know we struggled a lot on HOTS with nearly all of these issues, and this really hits home to me just how much room there is for innovation when it comes to skill expression and skill achievement in games.

I found it particularly interesting how I personally hugely favor the format used in in-person tournaments but never really thought enough about ways to translate those benefits over to ladder systems in online games.

I remember having a lot of discussions about industry innovations on matchmaking with things like Auto Chess and Battle Royales.  I think we still have a lot of room for improvement, particularly in game types like card games/MOBAS that usually use ladders for skill expression.

The Article on Informa

Written by Richard Garfield - Creator of MTG

Recently I started playing a digital game. There was a tournament announced for the top players. Although I wasn’t a top player – the game was new enough and there was enough luck in the play that I thought maybe I could qualify, and so I started paying attention to my ranking.

Quickly I noticed that I was having less fun because I was no longer experimenting and trying new things – I was focusing on what I knew worked. I began to lean more heavily on what other people said was correct rather than finding my own way. Losses were setbacks rather than learning experiences. When I rose to a level that corresponded to my actual skill, I stalled. Then the games became more samey, with the players mostly playing similar styles. I noticed that I felt like a mediocre player even though I was in the top 10%.

The experience brought into focus and made personal some of the misgivings I have had about extensive player rankings in games.

Whatever made me want to play in a tournament anyway? Isn’t that something only the best do? In the non-digital world all levels of players participate in tournaments for almost any game or sport. Most players in most of these games and sports are rarely ranked. The only times I was ranked with a skill rating system in an analog game was in chess, and even then only in tournaments. Yet I could often find tournaments for games that interested me. For example, I played bridge for years, participating in a club and many tournaments but I never had a skill based rating applied to me – until I played online.

In sports it was similar – if I got into a sport then I would participate in leagues and tournaments. I was rarely ranked, and when I was it was based only on the results of a few tournaments – it wasn’t the default mode of play.

If you contrast this to my digital game play it is striking – my play is almost always ranked in some way. If I opt out of that, as is sometimes possible, the player quality is generally so bad it isn’t worth playing – if I can find any opponents at all.

What is a tournament?

It is worthwhile thinking about what distinguishes a tournament from ladder play. I would say a tournament has a fixed length of time, a bounded number of games, and a set of participants that is fixed from the start. Sometimes leagues fit this description – though many online leagues allow players to play as often as they like within a certain time frame – which makes it more a form of ranking or ladder and less of a tournament. Similarly, the more players join and leave the league while in progress the less like a tournament it is. A tournament is less about who is actually the best player, and more about who plays best on a particular day, or handful of days. It is more like a game.

One way to think about it is, can the underdog win? If I am in an 8 person chess league and am 50 points below the other 7 players, it is unlikely at any given time that I will be ranked much above the very bottom. However, if we had a direct elimination tournament, I have about an 8% chance of winning, and a 40% chance to end up in the top half. If I did win, although an underdog, I would have played best on that day for that tournament – and it would be recognized rather lost in the noise of long term statistics.

Are even matches always best?

Matching players with evenly matched players is one of the main reasons for ladder play. This is not without its merits, but it also isn’t a strictly good thing. If you match people precisely then there is no sense of progress because players will win about half their games regardless of how good they get. When playing in tournaments or communities it is much easier to feel a sense of accomplishment precisely because these environments are not highly curated to be ‘fair’. This sense of progress is not just about being able to play against weaker opponents – it is probably even more about allowing players to play against stronger opponents – ones that they can learn from and aspire to defeat.

This is particularly problematic when a player doesn’t value their rating, and sees it as just something that punishes you with harder games. It is always easier to play lower level games – just lose a bunch. Unfortunately that means that the only way to play against harder players is to wait for someone to tank their rating just so they can have easy wins. I want to play against better players, but not players who are cheating the system to get a sense of power. Playing below your skill doesn’t just give a bad experience to the people who are beaten in this way – it also breaks the illusion that winning means anything for the players who beat them when they are intentionally losing – and artificially inflates their rating so that they will have more losses in the future.

Ladders are dangerous motivators

If a player cares about their ranking, every time they play, something of value is at risk. This naturally makes a more antagonistic play environment – especially with players of differing skills. We might otherwise hope that the stronger player would mentor the weaker; perhaps even play in a laid back casual manner – taking the opportunity to experiment. Instead, any adjustment of play will reduce the expected win percent and, against a lower ranked player, it could cost a lot. A player who values their rating must always bring their best game.

It is worth thinking closely about the case of mismatched players playing for a rating. I have been in environments where the higher ranked players saw it as an advantage and would hunt lower ranked players. I have been in environments where the higher ranked players saw it as a disadvantage and would be hostile if matched with a low rank player. It doesn’t matter whether the rating algorithm is accurate or not – if the perception errs one way or the other there will be an unwelcoming environment for lower ranked players. I have advocated in games in the past to simply have a cutoff, and if one player is favored to win, say, 2/3 of the time or more that the game is never rated.

Partly because of this, many digital games have ratings but share them only in a limited way; they will give you a very general category like Gold, and even if you know your own ranking they may obscure the opponents. Often this is a result of tests the developers ran, where they found players quitting games when they are facing a higher ranked player, or even leaving the game entirely when losing a few times and seeing their score go down. It seems like this should be a sign that something is wrong with ubiquitous rating. It is a bit crazy really, having a system to motivate play, which if treated too transparently causes people to quit. Your ideal is a system where players can advertise their accomplishments and see their opponent’s accomplishments which does not automatically put one player lower on the pecking order.

When I fenced on a team, we had a general understanding of who on the team was better. But, even if I wasn’t the best – I had my clutch victories, and my occasional excellent tournament results. Each member of the team was a unique asset – not a number. I have to imagine my experience would have been much worse if everyone’s ranking were plastered on them at all times, and were being adjusted every time we played. And that is a lot better, actually, than we could hope for in most digital environments, where your rank is usually against everyone in the entire world.

Numbers that always go up!

Developers face a conundrum, which is that the game community generally wants a measure of skill which is represented in numbers that always go up. This is one reason for the secrecy that sometimes shrouds how the ratings are represented, a purely skill based rating system must cost the player points from time to time. Many systems cheat that for a while – perhaps only having the players lose points above a certain rank – and until then they can ratchet their way up without worrying about losses.

Usually this is viewed as players wanting something unreasonable, you can’t have a measure of skill which is just accumulative. But you can celebrate accomplishment in an accumulative way – and the accomplishments can have a strong correlation to skill. In games and sports this is often the collection of trophies a player has won through their career. When an accumulative system exists in a digital game it is often the nearly meaningless “How many games have you won?” For most games and most players this is transparently proportionate to “How many times have you played?” particularly if the players are matched to have ‘fair’ games. The key is to get accomplishments the player truly values, and tournaments are a great way to offer something that is skill dependent, without threatening to take something away from them.

To do this in a way that all players can participate, it is necessary that there are different trophies for different difficulty achievements. Putting a numeric value on the trophies and converted into one big “trophy value” undermines their value. It doesn’t matter how many high school trophies I accumulated – there is no amount of them that equals a world championship. But that doesn’t mean high school trophies are meaningless!

Bridge did this quite well – in a way I failed to appreciate when I actively played. Players get points for many events, often even club play. The points are color coded so if someone has a huge number of black points it really mostly meant they had played a long time, but silver, red, gold, and platinum represent accomplishment in events that are progressively broader and more prestigious. Some players will find silver points challenging to get, and treasure the ones they have – while for others the silver are noise and the only ones that represent accomplishment in their eyes might be the platinum points, and over a player career – what they value will shift.

Are ladders the best way to choose our champions?

It is actually not even clear that the top players on a ladder are the best. I am sure that sometimes they are, but often the top echelon of a competitive ladder is filled with a group think and adherence to whatever is perceived as the top level metagame. A long time ago I divided players into innovators and honers. Innovators wanted to find new strategies in games, while honers would take those strategies and squeeze all the value out of them. The deal was that innovators would come up with something new and surprising and do well with it. In the long run, the honers were always the champions, but their reign was broken up by innovation. Of course in practice most players are a mix, and there are occasional champions who arise who do both very well – but that was the pattern in a lot of games as I saw it.

Ladders are a honer’s paradise – grinding out long term value from their incrementally better play. You can win tournaments on innovation. If you bring innovation to a ladder players will learn to play against it, or adopt it themselves before you can get very far. You can’t really get to the top of a big ladder that way, the grind is so long that any value you bring will be absorbed by the honers on the way. They will adopt it, or learn to play around it. This decreases the incentive to innovate which increases the chance of a stale metagame.

An example from Magic tournament history illustrates this well. There was a season where the metagame collapsed to one single strategy, built around the card Necropotence. The community complained it needed to be changed because it was the only way to win. Organized play left it however, and … what do you know? There was a radical new deck type that was entered in the world championship. If all play had been on a ladder – the innovators would have had to give other players their glory. It is even possible there would be no glory to be had, players on ladders can adapt to the changes even after they lost some ground – maybe Necropotence would even have remained supreme after some strategic adjustments. The pressure of thinking on your feet in the face of something new isn’t as big a part of ladder play, but it is one of the things that makes tournaments so exciting..

What else can we do?

One reason digital games skew toward ladders rather than tournaments is that they are so easy to manage online. Running a tournament involves scheduling things, or at least being pretty clever. One should keep in mind, though, that while tournaments present some challenges to the digital world – they are at least as challenging to arrange in the analog world, and yet game communities have always managed to do it.

One such clever approach is something like Hearthstone Arena, which is a bit like an asynchronous tournament. It doesn’t meet my definition exactly, but it does a lot of the same things. When you play it, you play a pool of players that isn’t (to my knowledge) selected to be fair – they are matched to your record within the arena run as closely as possible. Winning is hard, it represents beating more than 200 players. I believe it could support meaningful trophies. I played a lot and only won a few times, I would have been motivated if the system gave me a gold star for each – I cared a lot more about the victory itself than whatever the ‘real’ prizes were, which were pretty much meaningless to me. I played for years, and when I left the game there was really not much I felt attached to – the only thing the system seemed to care about is where I was on their ladder, it didn’t bother to track the biggest accomplishments I had.

I am sure there are many approaches to enabling all levels of players to access tournament play. Just giving players tools to do it themselves is useful. I’ll give one of my favorite tournament concepts here. Create tiers of tournaments, so that entering a higher level tournament requires winning (or perhaps doing well) on the previous level. The base level tournaments will be played often, because everyone qualifies for them. If a player has earned an entry to a level 2 tournament they can spend it and … perhaps get a ticket to a level 3 tournament. For level 2 and higher being in the top 50% will get you another ticket to the same tier. Prizes, whether real prizes or badges and virtual honors, increase with each tier. With this approach, better players will gravitate toward the higher tier tournaments because it is a better payoff. The same principle works with poker – the best players don’t hang out at the lowest stakes tables, it isn’t worth their time. On the other hand, if a player doesn’t feel like playing seriously they can play lower level tournaments and save their high level tickets for when they want to really focus, and it isn’t ‘cheating the system’.

Note that with such a structure, and with 8 person tournaments, winning a level 5 tournament represents defeating over 32000 people. No player is more than a few good tournaments from the top level play.

It is true that as a player climbs the ranks, the wait between tournaments will grow. This is not necessarily a bad thing. Digital environments often introduce artificial time constraints to generate excitement or to create habits in their players. This can be effective, but can also feel manipulative. Here there is an opportunity to have a time constraint for a legitimate reason, a reason that should make the people who are waiting for a game to feel good: they are among the few who have qualified for the rank of tournament for which they are waiting.

It is exhilarating playing in an event for a game you like, pitted against a wide range of player skills, and seeing how far you can make it. Each is a fresh roll of the dice, and you craft the story that your high and low points tell – they aren’t boiled down with all your performances to a number that over time tries to define your value as a player. There are many games cropping up in the last several years that provide that without really being tournaments; I am thinking of games like Fortnite and, to a lesser extent, Autochess. Players don’t expect to win every time with 8+ opponents, and it makes the victories and near victories all that much more exciting. I think a lot of what players get out of these games, they could get out of other games if other games were not so committed to putting everyone on a list.

Links

WoW Leak: Dragonflight Expansion

Links

Emails

Chad

As someone that loves playing Alexstrasza the recent buffs have been encouraging. However, I don't feel it's quite enough to bring her out of obscurity. I've written a few emails about off kilter, outside the box buff ideas and have another one for her. Reduce the cooldown of Dragonqueen from 150 seconds to 100. The opposing team would always be thinking "Ok guys, every meaningful team fight this bitch is gonna turn into a dragon." This also adds thematically that her dragon form is always looming over you. She's an awesome, busty dragon. Can she feel like one when we're playing her?

Daxx:

What if abundance (W) healed over time while it was being "channeled" and at the end you gain the burst heal?

If we could make it advantageous to stand in the circle..  it should at least mitigate some damage while waiting for the big heal

De La Soak

With the repeated targeted nerfs to Rehgar's totem build and his kit in general, how do you see his build and playstyle shifting in home games and in coordinated?

After facing down some rehgars and playing him myself a bit recently, he still feels good but totem doesn't feel as brain dead easy and the early game healing output really hurts.

I always appreciate your thoughts 💪 😎 Thanks!

Wrrqqarrr

I enjoy Alarak quite a bit, and was cool to see this mini-Lightning build revival. But was thinking about the comment last week about tweaking him to be somewhat of a Tank. Leaning into his Fantasy as a sadistic bastard who sacrifices his allies for power, here's my idea for Sadism additional function: Every Ally death increases Alarak's max health (resets on death). This is in addition to default Sadism functionality (which basically stacks his damage based on damage/death of enemies). In theory, this can turn him into something tanky/bruisery, allows for decisions to let allies die to stack your own health. Numbers can be adjusted of course, but thoughts?

Scaboboboobs

Simple question: What are your top 5 most fun talents in the game?  I'll give one... Headshot!  the level 20 for Dva.  When you combine it with Pew Pew Pew it results in some devastating burst damage

Kyle:

Cassia 16 - Static Electricity

Diablo 16 - Domination

Nazeebo 7 - Dead Rush

Tychus 4 - In the Rhythm

Butcher 20 - Slaughterhouse

Outro

Patreon Shout Out

ITN is supported by our badass Patrons. You can become an ITN Patron by going to Patreon.com/itn.

Thanks to our Patreon producers Shawn B & Mike R!

ITN shirt at shirts.amove.tv.

Catch the live show Thursday at 3pm Eastern / 12pm Pacific on Twitch.tv/Amovetv.

Follow the show on Twitter @ITNcast and ITNpodcast.reddit.com.

Thanks to Gorath for our intro music and briangriffithmusic.com (@briangriffith) for our outro..

Personal Stuff

Garrett

Kyle