WPPR v6.0? :)

I’ve always considered it an open door policy in terms of public WPPR system dialogue leading to things.

If you have solutions, bring them to the table anytime. Many people have! The annual WPPR’s for B-division proposals used to be really fun on an annual basis - usually happened a week before PAPA every year.

Pinside discussion has been surprisingly good at helping us work through how people perceive the upcoming changes, and we have taken suggestions from that dialogue and incorporated it into the WPPR rules that were ultimately implemented.

I guess you can call us posting WPPR v5.3 proposed changes in May of the current year a step in the right direction of what you’re looking for (compared to September announcements for 2015 and 2016 changes)?

Or are you looking for more of this WPPR v6.0 discussion type of dialogue?

If it’s the latter, this WPPR v6.0 discussion came from the comments of a player . . . not the IFPA. I fully support anyone that is willing to propose solutions to make the system more accurate and is willing to post them publically, and as you’ll notice I won’t hesitate to get involved in the discussions :slight_smile:

Least PAPA you walked away with 500 bones :wink:

1 qualifier as currently configured does not equate to best, no, it’s not long enough. But the shift from 0 to 50+ for adding three games is still absurd. Even ludicrous, to quote President Skroob…

No idea why the font changed on that one; I didn’t touch the settings. Maybe the system decided it was worth shouting?

The 500 came and went . . . the memories nightmares last forever :smiley:

2 Likes

How many diapers is that??

2 Likes

Yes, Josh, how many diapers does it take to cost the same as a new machine?

Holy cow! It was worth it.

1 Like

Hi All,

First off, the suggestions Josh posted/quoted were suggestions I made. I’m grateful that he has posted them here to gain other peoples opinions (which seem to be far more considered than on Pinside where everything becomes a drama so quickly!)

The suggestions are certainly not trolling - just ideas or suggestions which I feel could do with investigating. Just because you don’t agree with another persons post doesn’t make them a troll.

I’d like to expand a little more on the ideas behind my suggestions.

Firstly: Standardised Tournaments
This is not designed to inhibit innovation. The more different formats the better. The reason behind it, is that the ranking for each tournament which currently follows a set template does not always reflect the challenge required to win, reach the latter stages.
By having standardised tournaments, which have a set pre-determined value, this value can more accurately reflect the skill required. it will also stop formats being created which are long-winded and boring for the lesser to average players as they are not involved for the majority of the tournament due to the back-loading of games.

Secondly: Number of Games
I may not have made myself entirely clear here originally.
There are many tournaments, which have either a league style qualifying over a period of time, or a number of games played over a number of machines, where I have seen a player play just a single game in qualifying (usually a players partner or child) just to boost the numbers. That player is then counted as having played in that comp, with the same worth as the player who finished second.
What I am proposing is that the total number of games played by everyone, is then divided by the total number of players, this would give a more accurate representation. This can be submitted at same time as the results to the IFPA, meaning that no further work is required by them.
As for pump and dump tournaments, you only count your best score on any machine anyway, so it would make no difference if you played 1 game on a machine to qualify, or 200 games. it would count as 1 game.

Most contentious point? Last 20 results
The idea behind this is not to punish those players who cannot attend 30+ comps a year.
To keep the maths simple.
Let’s say 2 players of equal ability have a 50% chance of winning a tournament.
Player 1 enters 40 tournaments a year, therefore they will have 20 winning contributions
Player 2 can only enter 20 tournaments a year they will have 10 winning contributions and 10 minor contributions
Player 1 will always be ranked higher, despite them being of equal ability, purely because Player 2 can’t attend as many comps - for whatever reason.
Bring in a 3rd Player, who wins every tournament, but can only enter 10 comps a year.
They will be ranked 3rd in this scenario.
Ranking is awarded by attendance, as well as skill.
These figures are obviously taken to an extreme to show the point, but the numbers hold up in any scenario. (I have investigated this to some point in the current rankings already)

By changing it to the last 20 tournaments it gives a fairer indication of consistency of results over time rather than just picking out the peak performances.

Who is the better player? A player who finishes top 4 every tournament, or a player who wins 1 in every 10 tournaments they enter?

As for players not entering tournaments because they fear lowering their ranking. I doubt this would happen to any large scale degree anyway.
The best players would win the lesser tournaments anyway, and with the suggested tournament value change above there wouldn’t be such a large discrepancy anyway.
I (and the majority of people I know) enter tournaments for the fun, socialising and competition. WPPR points are an additional minor factor, I certainly wouldn’t not attend any comp for fear of lowering my ranking.

There would also still be degradation of points, so a win 3 years ago would be worth considerably less than a recent win. You’d have to still compete regularly to maintain your ranking.

Last one Tournaments gaining extra value based on the players present
The reason behind trying to minimise this is that if the top ranked players don’t compete, it’s minimising the opportunity for other players to earn more points. The fact that the vast majority of the currently top ranked players compete in the USA means players outside the USA are at an unfair advantage.
(In answer to the comment about the opposite of a vicious circle being good - yes it is, IF you are in that circle).

I hope that clarifies a few points. In no way am I suggesting that if these changes were implemented that everything would be perfect, but I feel that they could produce a more accurate system - even if the suggestions were tweaked somewhat.
Having spoken to a lot of people, currently the feeling in the UK (and other parts of Europe) is that the ranking is only really workable for the players in the USA who compete in large numbers of events.

I don’t think design by committee works in this case.

I value that ifpa observes, collects data, discusses with a core group, and then acts/reacts.
Imo the general community doesn’t understand the history and evolution and motivation behind the WPPR rules.
If we left it up to direct input, WPPR would end up with some weird upper right flipper that had nothing to shoot for (looking at you WOZ). :stuck_out_tongue:

2 Likes

First, thanks for the interesting discussion. Happy to have people putting ideas out there.

I understand the point you are trying to get across here, but I disagree with the math(s). Yes, your proposal in your example leads to an unbiased estimate if everyone plays at least 20 tournaments. However, the variance of the Skellam distribution when comparing two players leads to too much noise in the system that can’t be corrected. I have not actually walked through a real example with reasonable poisson parameters, but my gut feeling says 20 is not enough, and expanding the window more than this defeats the purpose of your proposal. I don’t think things being random is fair, the system should try to eliminate the random effect and try to estimate the underlying fixed fixed effect.

I think the rating (which I think is using some kind of glicko measure) is the best means to accomplish what you want, not a modification of ranking. Rating should converge over time correctly. It captures confidence in its updates. Of course treating ranking in event as a head to head match probably has massive flaws. Not to mention the issues of inconsequential events that people play just for fun. Maybe they play a league for fun, but can’t make the final. It is very inaccurate to call this a loss to a weaker player.

Your assumption here with the ‘last 20’ is that the values of the tournaments change to be more consistent between all events.

It sounds like you agree that based on our system right now, which values the quick-one-night tournaments far less than the full-weekend tournaments, that this wouldn’t be a good idea?

For me personally, a majority of my tournament play is small local events (monthlies, launch parties, etc) designed with the purpose of being worth small WPPR amounts. This kind of change would all but kill my attendance at these events under the current point structure because even winning these events will be detrimental to my ranking.

My plan years ago was to try and incorporate the Eff % as part of the ranking calculation, and use that average as a way to combat the people playing all the time having such a large advantage. This would be able to handle the various point discrepancies between events because playing in PAPA (worth 100), would be the same as playing in 12 monthly events worth ~8 each, with respect to the impact on your Eff % of WPPR’s earned based on your various finishing positions.

Golf I think gets things closer to accurate because they are able to offer events with similar point amounts (having a professional tour allows them to set these restrictions and limit the number of events even offered), and they also take an AVERAGE based on the number of events actually played, or a minimum of 40 I believe over their valid period.

This simply becomes impossible with the current way tournament values are so different based on size/strength of field/format, and any system that somehow makes everything worth ‘the same’ is a huge step backwards for the accuracy of ranking what a win at a given tournament should be worth. Unless we end up gutting the ~2500 tournaments per year on the calendar, and not allow any tournaments worth less than X to even be considered in determining someone’s ranking.

1 Like

We love to hear from more people on feedback. I’m not big on the “many people feel this way” hearsay opinions. I also talk to a lot of people (privately and publicly), including the Country Directors who in turn speak to a lot of people, and I think the trend over the years since 2007 has certainly shifted to become more pro-Europe with many of the changes made. If people think the system is only worthwhile if you live in the US, I would love to hear the reasons why and the potential solutions to combat those potential issues. I’ve been fortunate enough that most people don’t seem worried about pissing me off with their bashing of the system, and they shouldn’t be because I have insanely thick skin about it. I f*cking LOVE people that bash the system because that’s something I can work with. I can think about the reasons why those people aren’t happy, and see if there are changes that could be made to help their arguments, or further support the reasons why we do what we’re doing.

The largest group of people that have been equally unhappy with every version of the system we’ve come up with has been the UK and Australia. With both being islands maybe the fact that travel is simply more challenging in those regions makes for an increased likelihood of negative opinions?

The “hearsay” is not only what I have spoken to people about but also posted by those individuals on other forums (pinballinfo). The fact that it is mainly UK and European people on there is maybe why it’s not being heard in the US?

People are specifically saying that they are not entering competitions because of the longwindedness of tournaments.

I agree 100% that there needs to be different values of tournaments based on the format, but feel that there is simply too much variance between them currently.

With regard to travel, there simply aren’t enough competitions in the UK. The biggest bar/arcade in the country has 15 machines with a realistic capacity of 100 max.
Any large tournaments rely on people bringing machines, and hiring a venue. As you know there are no competitions with a pump and dump format so no prize money is ever generated as in the US, nor is money generated to cover cost of venue hire, I barely cover the cost of Trophies and have personally lost money running tournaments, as have others.

I’d be interested to see what the max pts tally that could be earned in the UK is, and where that would place that person in the world rankings? Keith Elwin has an eff %ge of 68.5, where would that have based him if he was based in the UK?

There are players in the top 40 who have a significantly lower %ge than players below 250, purely based on availability of competitions and the points they’re worth. There’s a saying regarding the value of a house Location, location, location.It seems the same can be said about the potential to gain points.

I think that if the current system continues there will become an even greater divide between the US and the rest of the world, effectively meaning you must compete/reside in the US to have any chance of breaking into the top 50.

EVERY single UK player is ranked lower than they were last year(unless they’ve only just started competing and are still filling their 20 games), as V5 comes more into effect. We can’t all be getting worse?

1 Like

Design by committee and hearing input from the community are two very very different things as I see it. I have absolutely no interest in—and doubt others are particularly interested in—design by the community. To me, that would be where everyone votes up or down on which ideas they think should be implemented into WPPR vX and then the IFPA goes off and builds it. The IFPA on the other hand, is a committee, albeit a small one, that doesn’t include every TD in the world for obvious reasons.

In the past when whatever version of the system was announced, it seemed completely clear to me that the IFPA was not soliciting feedback at all. Let’s look at the 5.3 announcement as an example…

Since the release of WPPR v5.2 for the 2016 season, the IFPA has continued investigating ways to make the World Pinball Player Rankings more accurate for how we rank players across the globe.

These latest changes will be incorporated into WPPR v5.3 which will be implemented starting January 1, 2017.

Please note that for all 2016 IFPA related activities (qualifying for 2016-17 SCS, 2016-17 ECS, IFPA14, etc), version 5.2 will be used for that qualification process.

There is nothing in the content that says these are proposed changes, as Josh has mentioned in this thread. It instead clearly states “these changes will be incorporated… starting January 1.” If 5.3 is just a proposal, and there’s potential for updates based on feedback, prior to implementation, then I think it should be made clear that’s an option and there should be a clear way and timeline for the community to give that feedback to the committee. While there may be other instances, this is the first time I’ve seen Josh actively ask for feedback instead of simply saying he doesn’t agree and pointing out reasons why something won’t work. It would be refreshing to see a little more thought put into why something can work, and see some of those can work ideas implemented proactively instead of reactively, which is how it appears to be happening to me at this time.

1 Like

This is crazy. No, not crazy. This is incredibly generous. Do you not charge entry fees to cover the cost of the venue?

1 Like

In the same announcement:

“For full detailed information on the 2017 formula, click HERE. Any questions please don’t hesitate to contact us at ifpapinball@gmail.com. Please check back when we announce that WPPR v5.3 has been implemented for 2017 as the details may always change prior to launch.”

Details have always changed prior to launch because of the dialogue that has been created.

Go back and read the v5.0 and v5.2 announcements on Pinside. Each have about 500 posts and you can see where that dialogue impacted implementation.

We can certainly be more explicit that these are PROPOSED CHANGES, but I’ve always gone with the “everything is proposed until it’s actually implemented” thought process every year.

Certainly the lack of posting to either Tilt Forums or Pinside leads to the discussion staying at a micro level. It wasn’t until I randomly stumbled upon the Aussie forum that I was able to engage in a dialogue with that player base. I don’t think many people here follow the various local forums. I’d love to see that local dialogue branch out into more productive international dialogues through Tilt Forums. It’s a great place for it with far less noise than Pinside (probably the only other global forum option).

Much easier to go the other way and compare the Eff% of the best UK players to Keith’s events.

On Keith’s active resume if you assume he earned his 962.07 points based on a 68.5% win percentage, that’s 1404 WPPR points that were available in those events.

Here’s the UK top 6, plus yourself, because surely you would be top 20 if you played in NYC as you mentioned on the UK board :slight_smile:

Martin Ayub - 465.29 WPPR points (which would ironically put Martin in the exact same spot ranked 38th)
Craig Pullen - 508.53 WPPR points (would go from 72nd to 33rd in the world)
Greg Mott - 325.17 WPPR points (would go from 96th to 90th in the world)
Peter Blakemore - 236.85 WPPR points (would go from 99th to 180th in the world)
David Mainwaring - 195.86 WPPR points (would go from 103rd to 250th in the world)
Matt Vince - 361.95 WPPR points (would go from 123rd to 71st in the world)
Wayne Johns - 267.46 WPPR points (would go from 331st to 141st in the world)

Interesting changes for sure based on any of these guys performing at the relative level they have performed at, but playing in every event Keith played in.

1 Like

For reference and info

Cheers, Craig

Some additional fun with numbers, just because ‘access to tournaments’ is the biggest issue we fight globally with respect to the WPPR system . . .

Keith’s active resume represents 1404 WPPR’s ‘available’, and I started to wonder where that fell against ‘the norm’ with respect to points available for players competing.

In the top 500 ranked players, the average number of WPPR’s available in any given top 20 resume is 1048.

Broken down further:

top 50 --> Average participation in 1578 WPPR’s worth of action (KME, JOE and BEK are the only three players in the top 10 to have a ranking based on a participation rate that is lower than that)

top 100 - 1457 average
top 150 - 1379 average
top 200 - 1288 average
top 250 - 1234 average

So clearly as we move down the list, based on the efficiency of points earned, players are absolutely playing in fewer ‘points worth’ of events.

Some fun participation numbers . . .

  • Francesco La Rocca is #1 in the top 500, having played in 2570 WPPR points worth of action based on his top 20
  • David Mainwaring is #3 having played in 2162
  • Steve Bowden you would think would be up at the top, but he is NOT . . . 1840 points worth of action

Here are the other UK folks from the prior example:

Martin Ayub - 1417 points worth
Craig Pullen - 985 points worth
Greg Mott - 1340 points worth
Peter Blakemore - 1829 points worth
David Mainwaring - 2162 points worth
Matt Vince - 1104 points worth
Wayne Johns - 889 points worth

At some point we can’t all be “FLR’s” (and nobody is lol), but at what point is it actually right to blame the player for lack of their participation in events? (with respect to saying the system isn’t working using those players as the example)

There’s obviously some line there with respect to a participation rate that is expected of a top 100 player IMO. The question there is where is that line . . . and can we use that expected line to help make things more accurate by adjusting someone’s rank based off of their activity that goes beyond that line as not being ‘fully positive’. Fully positive being NOT NEGATIVE (because I do believe that has a huge negative impact on promoting people to play).

For example, if we say the average participation rate of the top 250 (1234 average) would be expected by any player looking to be in the top 100, is it fair to look at any participation beyond that 1234 as some sort of ‘unfair advantage of access’?

Let’s just assume that’s fair . . . anyone who hasn’t played in at least 1234 points worth of events, sorry, get out and play ‘to the average’ to be ranked properly.

Someone like Francesco however, who is ranked 36th in the world based off of earning 484.17 points in his top 20, we go WAIT A MINUTE THERE BUDDY. You played in 2570 points worth of events, more than double the average player, so a more TrueWPPR value for you would be to take that ‘beyond the average activity’ and apply his Efficiency percentage to that amount (that way he’s getting SOME bonus for playing in more events - just not at the full rate he’s getting now).

So 232.48 would be his WPPR total if he played in 1234 points worth of events. His other 251.69 points earned on his resume would get a WPPR tax of 81.16% based on his Eff %, so he would only net out 47.42 of those points in determining his ranking.

FLR’s total would be 279.90 WPPR points, ranking him 129th in the world.

2 Likes