18 replies to this topic

[Daidachi]

I originally posted this in another thread, but thought this might be better served being moved to its own post.

___

I think we can all agree that the scheduling for the current tournament has caused a great many logistical hassles.

With that in mind I would suggest we take a page out of the FA cup in English football (and to a lesser extent the Europa League) for next time.

Bear with me as I explain the analogy.

I understand the idea behind giving everyone a fair run, and I applaud it. But the practicalities make this a somewhat stressful experience for all involved.

Whether it be shoutcasters, PGI staff who have to confirm rescheduled matches and review uploads from participants, let alone the participants themselves, who are furiously trying to take part, whilst still maintaining some semblance of healthy work/life balance.

In the FA cup, teams from all levels of professional competition are permitted entry. The process to reach the finals is much longer for teams outside of the top level professional leagues however.

What I'd suggest is based off placing in this tournament, as well as existing comp play rankings (IE MRBC), brackets are constructed for teams.

Bracket A might consist of teams that have consistently placed highly at the top levels, Bracket B the next level down, and so on and so forth. This might of course need to be adjusted for regions where there are fewer teams (so only 2 or 3 brackets for AP, whilst there might be 10 let's say for NA).

Rather than playing every single team in the region, organise it elimination style - teams in the lowest brackets (little to no comp experience) start the tournament off by playing against teams in their bracket, which means they aren't starting off getting wiped out by top level comp teams (and top level comp teams aren't facing weeks of steamrolling lower ranked teams).

This means they are playing against teams of a similar level to them, but you can still have at max 2 games a week and get through the process. Once they've gone through their bracket and there's a last man standing, that team advances into the next bracket up, who start playing the following week.

High level doesn't mean an instant pass into the regional finals obviously, so if you started the regional finals off once bracket A have played through, that would work.

Regional finals would continue in current format, as you'd have few enough teams by then to make for a fairly streamlined process.

This would mean:

- increased ability to shoutcast. Rather than having a morass of matches to cast, and only so many hours in which they can do it, the schedule would make for manageable schedules for each region.
- More interest in comp play from new teams. If they learn it and do well, they'll be much more likely to do it than if a current player looks at the list and goes 'man, I don't really feel like getting wasted by ISEN/EmP/Lord live on twitch'. And getting new people involved is great, especially when they don't feel they're wasting people's time.
- fewer cancellations/dropouts. Teams who make it through multiple brackets might have a lot of matches, sure - but they are far less likely to care when they can see that they are accomplishing something!
- A more even standard of competition. This makes for a better product to view and advertise the game, and the company with.
- A much smoother review process, because rather than having a glut of matches at the start, and the pace smoothing out as you get closer to the end, you have a steady number of matches per week.

Just a few thoughts I put together while at work anyways - cheers for reading through, would be interested to hear any thoughts others may have.

[Daidachi]

Tamerlin replied on my original post, so I'll put my response here.

"The problem with this is the FA cup is a "lost one game and you are out" tournament. And there is no "fair" way to establish brackets when you don't know who will be in which team. One way would be to use combined Pilot Skill Rating, but that isn't perfect. "

My response:

Two factors to this - one is the review period that was established before the current tournament, which can be used to check team composition and the history of team members.

The other is that the current tournament itself is a very good opportunity to check player performance against what their current PSR is, and establish what adjustments need to be made to make it more closely reflect a players actual skill level (hint: less focus on team win/loss. If a player does 56 damage, even if their team wins their PSR shouldn't stay even).

[RjBass3]

I like the FA Cup Analogy but may I suggest instead the US Open Cup analogy?

In the US Open Cup, all MLS teams are auto placed in the 4th round. All NASL and USL teams are auto placed in round 3 and all other teams (whoever wants to participate and are not in one of the major three leagues) start in round 1.

It is a single elimination tournament where as all teams who enter the tournament can get knocked out of the tournament in one single loss.

So the 1st round begins, and all the teams still left standing move onto round 2 to play the other half of those entry level teams. The entry level teams who win move onto round 3 to play the NASL and USL teams. The winners from that round move on to play the MLS teams in round 4. The winners of round 4 move onto the quarter finals with the winners having to win the best of three games, then onto the semi finals (also the best of three games), and eventually the final (single game but for drama and the fan base in MWO you could make it the best of five games or something).

In the case of the MWOWC, all teams in MRBC division A & B start the competition in the 4th round, C & D in the 3rd round and all other teams in round 1.

But then this all goes back to PGI's saying "we don't have any previous records to go off of" which to most of us makes no sense considering that many of the teams from last season's MRBC league are participating in the MWOWC. They obviously didn't want to pull from MRBC though for some reason.

[Tamerlin]

Yeah, but MRBC does not have max roster limits, so it's hard to use their results directly.

[Daidachi]

Tamerlin, on 08 June 2016 - 09:35 PM, said:

Yeah, but MRBC does not have max roster limits, so it's hard to use their results directly.

Agree as far as directly goes, but given the requirement to upload at the least screenshots for MRBC results (and most teams will upload video for the cbill bonus it gives your standing), a couple of dedicated people could sift through that data to determine individual player performance in matches combined with PSR inside a week (assuming same number of signed up teams as this time).

If in doubt, put them in a lower bracket - cream will rise if it's good.

Also RJBass - that's a really solid idea!

[PhoenixFire55]

Tamerlin, on 08 June 2016 - 09:35 PM, said:

Yeah, but MRBC does not have max roster limits, so it's hard to use their results directly.

Most top tier teams use like 10-12 guys all the time anyway. Larger units might have up to 40 people on their teams rosters but they are there "just because" anyway.

I'm not saying MRBC based seedings should be used, but seal clubbing isn't exactly pleasant for both clubbed and clubbers. Its a rather complicated matter how can this be avioided while keeping the tourney even for all and if it should be avoided in the first place.

Still, the bottom line is that playing 90 matches on 90 different days for qualifiers is silly. Even swiss pairing like RHoD did would have been better. Just how many teams will simply stop playing after 10-15 games when it'll become apparent they can't make it to regional finals?

[Ibrandul Mike]

I believe a week could be enough, after a few months of preparation and with enough persons working on it. You would need to watch every single game. You would need trained people to watch the matche. Using good checklists to give each individual pilot his rating (positioning, level of awareness and so on). Then you need all the statistics for the match. Hit percentage, how much damage to which piece of the mech vs. Twisting and of course efficient targeting and firing... effective damage spreading through twisting... and so on.

Or you could just say thank you to the people looking at the results screen and trying to see the individual players performance. Because with not enough information and only untrained personel... you can just toss a dice to determine the round a team starts in.
Of course there is a big difference in the skill levels of players. And of course there is a difference in the team leadership capabilities. And much more. Good teams will get out on top no doubt. Less games would be better overall.we all agree on that. Another seeding system might have worked better for the tournament (yeah most likely). As far as a seeding system for the next MWOWC is thought off... perhaps teams could be seeded based on their standings in this MWOWC. Might be easier even if not as acurate as possible.

Edited by Ibrandul Mike, 09 June 2016 - 04:45 AM.

[Daidachi]

PhoenixFire55, on 09 June 2016 - 04:35 AM, said:

Most top tier teams use like 10-12 guys all the time anyway. Larger units might have up to 40 people on their teams rosters but they are there "just because" anyway.

I'm not saying MRBC based seedings should be used, but seal clubbing isn't exactly pleasant for both clubbed and clubbers. Its a rather complicated matter how can this be avioided while keeping the tourney even for all and if it should be avoided in the first place.

Still, the bottom line is that playing 90 matches on 90 different days for qualifiers is silly. Even swiss pairing like RHoD did would have been better. Just how many teams will simply stop playing after 10-15 games when it'll become apparent they can't make it to regional finals?

This pretty much nails my thought process. If you want to encourage both top level players and new comp players to take part, there needs to be a form of segregation - but at the same time, you don't start all of them at the same time, with the winners of each group facing off - that will lead to a number of really good teams being disadvantaged by their skill, and some really lop sided/boring semi final matches.

I mean as an option you could (I guess) use a seeding system similar to the champions league/Europa League/world cup, with one top team in each division, and the top 2 teams from each division progressing to the next round - but it'd be even more of a pain to set up the seedings without outright copying the ranking data from MRBC in its entirety, and would still lead to disparity in match performance between teams - though not to the extent we're seeing now.

Also, at the end of the day you want two things - one, new entrants to the scene to go into their matches knowing that they have a fighting chance against their opponents, and grouping players of similar levels together helps them buy in to that. Secondly, you want the experienced comp players to really feel like they have something to play for - that they aren't just wasting time until the finals.

Both these things make for better performances to shoutcast, because one leads to opportunities to point out to new viewers/participants the things experienced players inherently know, and the other because it provides for matches where win or loss can be put down to miniscule mistakes by either side.

[Kageru Ikazuchi]

Teams may not have the same names or members, so new teams could be formed with experienced players, or old successful teams could have a large influx of inexperienced players

There probably needs to be a per-player ranking, based on their team's performance in past official tournaments, to seed future official tournaments. Then, when the next tournament comes up, PGI can sum up (or average) the experience of the players on any particular team to seed them for the qualifying rounds.

[Pineapple Salad]

Kageru Ikazuchi, on 09 June 2016 - 08:07 PM, said:

Teams may not have the same names or members, so new teams could be formed with experienced players, or old successful teams could have a large influx of inexperienced players

There probably needs to be a per-player ranking, based on their team's performance in past official tournaments, to seed future official tournaments. Then, when the next tournament comes up, PGI can sum up (or average) the experience of the players on any particular team to seed them for the qualifying rounds.

A good team is more than sum of it's members. A team with average players but good teamwork and tactics always beats a bunch of skilled individuals.

[Daidachi]

Pineapple Salad, on 09 June 2016 - 11:06 PM, said:

A good team is more than sum of it's members. A team with average players but good teamwork and tactics always beats a bunch of skilled individuals.

No argument there, but calculating average player level gives a good starting base for seeding.

And if it leads to a few upsets, is anyone going to complain? Hell no! Good competition is what any competitive team LIVES for, in any sport.

[Kageru Ikazuchi]

@ Pineapple Salad (the "quote" function doesn't work on my work computer ... weird).

I agree that a bunch of players are not necessarily a "team", but giving each Player an individual score solves the problems of turnover, teams dissolving and players joining someone else, or recruiting ringers (or rocks).

Even if the score is as simple as: non-participant = 0, bottom 40% = 1, 41%-70% = 2, 71%-90% = 3, 91%-100% = 4 (based on a Player's team performance); it would give PGI a basis for the initial seeding of qualifier brackets or pools and make it unnecessary to have a 90-team round robin.

[Pineapple Salad]

Kageru Ikazuchi, on 09 June 2016 - 11:52 PM, said:

Even if the score is as simple as: non-participant = 0, bottom 40% = 1, 41%-70% = 2, 71%-90% = 3, 91%-100% = 4 (based on a Player's team performance); it would give PGI a basis for the initial seeding of qualifier brackets or pools and make it unnecessary to have a 90-team round robin.

How would you measure an individual player's performance though?

A light pilot that excels in harassing the enemy team but only pulls 100 damage can be just as good as an assault pilot that does 1000 damage. In fact the light pilot enables him to do so.

[Kageru Ikazuchi]

Pineapple Salad, on 09 June 2016 - 11:57 PM, said:

How would you measure an individual player's performance though?

A light pilot that excels in harassing the enemy team but only pulls 100 damage can be just as good as an assault pilot that does 1000 damage. In fact the light pilot enables him to do so.

I wouldn't bother ... only base it on his team's standings in previous official tournaments. With a 12-man roster, any one player only affects the team's score by 8.33%. Any more granularity than a scale of 0 to 4 (or maybe, maybe 10) is more effort than it's worth.

Also, since it's based on the team standings, your point -- that a 100-damage light can be worth as much for a win as a 1000-damage assault (or vice versa) -- applies even more so, when you don't take an individual's stats into account.

[P H O X]

In any case, there must be something done about it. ATM, thist is not about work/life balance. Its more about playtime/divorce-papers balance.

Even if we know we could not flield all 8, we have to be there to be not disqualified. So worst case: three evenings out of the window for a screenshot.

The roster should be also bigger, so that the playtime could be shared onto more shoulders.

[Daidachi]

P H O X, on 10 June 2016 - 04:58 PM, said:

In any case, there must be something done about it. ATM, thist is not about work/life balance. Its more about playtime/divorce-papers balance.

Even if we know we could not flield all 8, we have to be there to be not disqualified. So worst case: three evenings out of the window for a screenshot.

The roster should be also bigger, so that the playtime could be shared onto more shoulders.

There is also the small fact of time differences to take into account.

Even within the same area (NA for example) you have a big gap between East and West Coast, and trying to find a happy middle ground between both is a considerable hassle.

The same I am expecting to happen with AP (due to the 1.5 hour gap between NZ and East Coast Aus/3 hour gap between East Coast Aus and WA/Singapore), admittedly to a lesser extent since they have longer to prepare and pull forward matches that are going to be a scheduling hassle.

[Aresye]

They didn't even listen to league admins about how to run a tournament in the first place, despite the fact they ASKED for it. You really think they'd consider using existing league tables? lolololololol

[Daidachi]

Aresye, on 11 June 2016 - 02:44 PM, said:

They didn't even listen to league admins about how to run a tournament in the first place, despite the fact they ASKED for it. You really think they'd consider using existing league tables? lolololololol

There was a certain element of bravado when the schedule was put together initially, which I think has faded now that the realities of organising an event of this size has actually dawned on PGI.

While I don't expect them to say that they should have listened to comp player feedback, I do think that behind closed doors they are likely discussing what they need to change for next time they run it (or at least will have a debrief at the end of the tournament).

Whilst using previous events as a bellweather is probably the smartest thing to do, they could use rankings from this tournament exclusively. It'll be of questionable use since the current tournament is driven by attrition more than anything, but it could be used.

[TheArcher]

The one thing I find odd is that this is called a "Tournament". From that, I infer a short-term competition, say 2-4 weeks. It sounds much more like a League, with four months of matches to determine the regional finals competitors. Then the Tournament starts.

I did look up the definition of Tournament, and it doesn't say anything about time span, so I guess my "short-term" connotation is just from seeing 4 day to 4 week tournaments on TV. (Soccer, golf, etc...)