Let's talk about initial seeding in tournaments


#1

I’m probably in the minority here, but I really don’t enjoy the default of seeding tournaments based on IFPA ranking.

Perhaps I’m wrong, but the reason IFPA ranking is used to see is so that you don’t have top players randomly facing each other early in the event. I get that it would be “unfair” for say Raymond and Keith to face each other in the first round. Sucks for them to fly out to [some town far away] and get a bad draw like that. And presumably we want the finals to be made up of the best players possible.

This is “more fair” for those top players…one person doesn’t have a substantially more or less difficult path to the finals. But I don’t think that makes pinball better.

Exhibit A: Heads Up Pinball Championship. (The Challonge bracket seems to be gone so I can’t verify this). The list of players still alive when I tuned in was almost entirely what one would expect. Of course, this is to be expected…if you’re a top player, your first match or two is a cakewalk. Why would a lessor ranked player bother to show up for this if they know they’re being fed to the sharks right from the beginning? The path to victory for a bottom seed is literally almost as hard as it could possibly be. (In a seeded 16 player pinball tournament, the 16 seed would face: #1, #8/9, #4/5, #2/3. Yes, it could be slightly harder, but not much.

Exhibit B: local monthly tournament at 1up Colfax in Denver. This tournament starts out in six groups of 5 or 6, but it seeded using IFPA ranking (so the top seed has the bottom seed in their group, etc). The groups are evenly distributed so that there aren’t hard groups and easy groups. Each should have about the same level of difficultly. Great news if you’re one of the six best players in the tournament…you’re guaranteed not to face any of the other six best players during your first 8 to 10 matches (you play each person in your group twice). If you’re one of the worst 6 players, you’re guaranteed not to face any of the 6 worst players in the tournament.

Easy path for the good players, hard path for the bad players. Even though the top two players advance out of this group, the number of times a lessor ranked player has advanced is very, very small.

I’m all for making a qualifying/seeding portion of the tournament, and using that to determine groupings or matches later in the tournament.

But (even as someone who benefits from this system at the aforementioned monthly) I think that using IFPA ranking for initial seedings/pairings is really unfair for the players that need the most encouragement.

Edit: To be clear, I only think this is an issue with open events, meaning anyone can join. I don’t think you should be seeded at the start in any open event. I have no problem with initial seedings in a closed event…one where there was a qualification process that has already been completed (SCS, IFPA Worlds, Circuit Finals, etc).


circuit finals tentative first round matchups
#2

I think it’s really the only way to do things, especially at majors and circuit events. For smaller monthly and one off events, random pairings may be more widely accepted for sure.

I hear what you’re saying, but when it comes to competitions that are paying out large sums of money or prizes of large value, I’m not sure that catering to those that need encouragement is the correct way to go about it. I don’t mean for that to come off as crass or harsh, but this almost never, if ever, happens is any other form of competition.


#3

I must be misunderstanding this. Are you doing initial pairing based off ifpa rankings and doing slaughter?

I use IFPA ranking and do adjacent. So initially seedings and pairings are with people of similar ranks/skills then subsequent pairings I will do Swiss so that your being paired with others who are performing on a similar level.

I started with this recently and I think it’s the best most fair way of doing things for all skill levels.


#4

It’s tricky, because past a certain point, IFPA ranking means nothing. I’ve heard this described anywhere from 500+ to 50+. Unless you’re at a high major or happen to have a glut of high ranking players in your area, IFPA ranking is likely to be almost as random in terms of true skill as drawing names from a hat past the top 5 or so.


#5

Okay fair enough. Then the first round pairings might not be perfect, but doing ifpa ranking is only the first round. It’s swiss after that. No big deal. I still think it’s the best way to do pairings for a group matchplay event.


#6

I definitely agree with you here. I feel like in NV, the polarity between strong players and novice players, especially at the beginning of the tournament, was much more so than other major tournaments like that; the significance of using IFPA tank to seed was more pronounced, for sure. It felt like that anyway. @pinwizj would probably tell me I’m wrong though.


#7

The kind of format matters big time. The whole complexion of the format kind of depends on how people are to be seeded if there’s no qualifying component.


#8

Here was the breakdown of ranks for HU:

1 Raymond Davidson 3
2 Trent Augenstein 7
3 Andrei Massenkoff 20
4 Karl DeAngelo 28
5 Bob Matthews 33
6 Johnny Modica 39
7 Tim Sexton 48
8 Fred Richardson 72
9 David Peck 213
10 Robert Hooton 711
11 Craig Sengstock 892
12 Sal Ayoob 954
13 Raymond Ashby 1006
14 Sean Stewart 1082
15 Deborah Tahlman 1372
16 Dan Newman 1402
17 Jeff Rank 1683
18 Tom Neighbors 2111
19 Steve Strom 2277
20 John Barker 3263
21 Mike Lund 4302
22 Marcus Amith 5409
23 Ashley Resurreccion 5685
24 Bradin Wilhelmsen 6035
25 Garrett McCarty 6680
26 Patrick Mueller 8649
27 Andrew Mueller 21147
28 Andrew Moskowitz 50000
29 Chris Herrick 50001
30 Dana Martin 50002
31 James Hastings 50003
32 Alyssa Rex 50004
33 Chris Wren 50005
34 Keith Ipock 50006
35 Peter Vibar 50007
36 Myles Grosovsky 50008
37 Daniel Golin 50009
38 Gary Barker 50010


#9

I may have to correct myself here.

At IFPA nationals, rankings ranged from #3 to #13730. So the gap can still be fairly large. As was mentioned above, I think this can be a testament to how meaningful (or not) IFPA rankings become at a certain point.

I think the situation you’re talking about here would be better applied to tournaments that have an open format with no qualifying component.


#10

Ahhh, just beat me to the point.


#11

Wait seriously? The finishing positions were exactly corresponding to the initial seed?


#13

Those are just the initial seeds. Karl finished 2nd.


#14

I don’t mean to argue that we should be catering to those that need encouragement…I’m more of the position that we shouldn’t cater to the highest skilled players - which is what happens if an open event seeds based on the WPPR ranking.

I think poker is a pretty fair comparison to pinball in that they have a similar mix of luck and skill in the short term (although the luck factor is higher in poker, and pinball gives more opportunities to make decisions per match/game/event/hand/ball).

I don’t quite understand your point here…is it that other competitions always seed based on skill? The heads up championship of poker uses random pairings: https://en.wikipedia.org/wiki/National_Heads-Up_Poker_Championship

Nearly all leagues here form groups by pairing people of similar ranks/skills. The largest monthly creates groups in a similar manner to initial Pinburgh groups would be formed (ie: with 16 players…group one is 1,8,9,16, group two is 2,7,10,15, etc). It’s round robin where you play everyone in your group twice (and groups are 5 or 6 people generally). This round robin is the qualification portion of the tournament. Top two advance out, and everyone else goes home.

Are you suggesting that difference in the “true skill” of a player ranked 500 and one ranked 5000 is basically random? I couldn’t possibly disagree more.

I agree that past a certain point, it probably doesn’t mean much, but if we must declare a point, it’s way into the thousands. Drawing the line at the Top 5 is preposterous.

Absolutely, that’s exactly what I’m talking about. My two examples were of events where players were immediately seeded into “fair” groupings based on past performance.


#15

I will add that I don’t have a problem with how Pinburgh does it. I think it’s because you still have so many opportunities after that first round…and the fact that it does continually move towards pairing like players together.

If Pinburgh was a double elimination heads up tournament, seeded by WPPR ranking, why would anyone in the bottom quarter bother showing up? It’s effectively a one strike tournament for them.


#16

What I was saying was that sports like Tennis, NCAA, NHL, etc go the 1-16, 2-15, 3-14 route. I said almost never happens, because I was sure there would be at least a couple bigger events out there that were random. =]

I’m pretty sure we’re on the same page for the most part.


#17

As for the NCAA and NHL, etc…seeding makes perfect sense because they went through the qualification period of the regular season.

You’re right about tennis though. Don’t follow it closely enough to have an opinion.


#18

I will say with respect to IFPA “corporate” events, using WPPR rank as a seeding metric we are at least eating our own lunch so to speak.

Everyone has the ability to reach as high a ranking as they want, and with that come perks. Whether that’s qualifying for certain events at all, or preferential seeding at others, it’s gives us another motivational point for people to earn as many wpprs as possible before the next Heads Up championship to give themselves that seeding advantage.


#19

Suuuuuure


#20

Sorry, let me clarify. Everyone that’s actually really good at pinball has that ability :slight_smile:

Copy/Paste the example of ZAC’s world ranking if he only showed up to PAPA each year the last three years and went to nothing else (top 15 in the world from just that).


#21

They’re the ones that don’t have this problem though, aren’t they?