If you’re using MatchPlay, you’re going to want to use whatever the fastest entry option ends up being. Note that with ~60 matches, it will probably be easiest either to get Scan Winner working or to have the players use “Pick Winner” and make sure you have a keyboard available.
My experience in 35-player versions of this tournament is an average of 2 or 3 people in line waiting to select the winner at the computer. I wonder if this will translate to a 10 to 15 person line at the computer, if it will somehow be less, or if the queue for the PC will continue to back up and grow throughout the event.
No one will have time for a smoke break or a bathroom break. This format is ridiculous fast paced. Unless you have super long playing games, people are going to be in the for about 3 minutes. You literally do not have time to go to the bathroom with only 3 minutes.
Biggest I’ve run is 50 people. 10 people in queue. Average wait was 3 minutes and 10 seconds. Here is a sampling of average game times used for this tournament.
AC/DC Pro - 6 minutes
Comet - 5:45
Dr Who - 5:53
Iron Man - 5:44
Metallica Pro - 6:57
No Good Gofers - 7:08
Paragon - 5:38
Stars - 5:21
Even with 30 people in queue, games are going to end at a rapid pace, sometimes you will have 10-12 people standing in line waiting fo give results. Good luck, and please don’t underestimate the speed of Flip Frenzy.
I’ve ran about 8 Flip Frenzies now, and every time during the announcements I let people know they will have ZERO time to step out of line/queue for food, bathroom, etc. and most of the time they shrug it off. After we stop for a break, those same people eventually come back to me to tell me they were blown away at the pace of play and had no idea that wouldn’t have time for any of those things.
A FF comp with 150 players, 30 people (20%) in the queue will need 60 machines available to play simultaneously!
10 hours is WAY TOO LONG for this format, you’ll probably max out TGP after 3hrs, 4hrs definitely.
I’ve organized a 101-player Frenzy and a 94-player one.
This is really at the edge of what the system can handle as it is limited to one computer.
Here are some takeaways:
-I had only the winner come up to register scores. That way, if something is wrong, it is their fault and the line isn’t as crowded. They also registered themselves. Then there is no miscommunication with the person registering.
-With this many people at a location with 90 games set up, it took longer for people to find games, and it is definitely better to have people go to the game as soon as the match is assigned.
-We had 10.4 games as average for 2.5 hours. We then took top 16 through to four-player quarter, semi, and final on three machines per round. Average time in the queue was 5:16.
(with the 94-player one, we had 12 games played on average and 3:10 in the queue, but that time we had fewer players new to the location, so they knew where machines were. We also had an easier PC mouse set up. With the larger tournament, we, unfortunately, used a Mac and the mouse was too sensitive, so that drained valuable seconds each time).
I’m also a proponent of hidden standings as some players tend to give up otherwise. I have seen this several times. However, opinions vary on that.
Frenzy is not the best format because of the luck factor of opponents and games and having four-player matches for the finals rounds removes some of that imho.
@wizcat has done some testing with a program that equalizes games played, machines played, time in queue and opponents specifically for FF - not sure it’s ready for being shared yet?
Yeah it was a monte-carlo simulator based on the argument that because your opponents are randomised, there is a chance player A gets a really weak selection of opponents, compared to player B and so on. We wanted to see how it would work if the player selection was more balanced.
So when we’re ready for starting a new match, instead of taking the player from the queue who has been waiting longest, we instead select a player from the queue who hasn’t been paired against that opponent before (or the player who has been paired least).
This was fantastic at balancing the opponents, but now the randomness is added to the queue time instead, so some players would join the queue and be selected again barely 20 seconds or so later, whilst other times you might be in the queue for 10-20 minutes (compared to 5 minutes when always taking longest)
The upside is that it didn’t radically affect how many games people played on average (there was still a spread, but it was no different to the spread seen due to some games taking longer than other games). The downside is the randomness in queue time is more awkward for the players to handle. They don’t know if they can take a five minute refreshment break or whatever.
The code is messy (c# console app) but I can add it to github if anyone wants to play with it. It’s really only useful for running simulations quickly - (it completes a 6 hour FF tournament in 3 minutes or so, then reports the results to log files)
We do this by design. No desire to increase the queue. Our 3 Frenzies average 25 games played and that’s the purpose of this format, to play a lot of pinball in a short amount of time. We split the 3 hour session in half and take a break in the middle. Also, this is just something we do a few times a year as our bread & butter is 4 player match play.
Great format for location to keep playing, gets new players a lot of game time.
One request: can you add the option to change game in a match versus just DQ’ing the game due to malfunction. On location, there are other non-tournament players also playing games, and so if a match is called on a game in use, we need to simply re-draw the game without putting players back in the queue and deactivating the game.
Hey Folks,
How do people manage breaks when doing a Flip Frenzy using Matchplay?
If I pause the timer the game timers still go on… Which skews the stats.
If you close the tournament then reopen it timers still run.
I’d love the pause button to let you submit results for the running games but then not create anymore games and put the arenas in the list on the left where you can create game ready for when you resume.
Wondering if anyone else notices a seemingly large number of repeat match-ups during a Flip Frenzy? It was just an observation made in the last one I played in, but the numbers seem to bear it out. I’m not a statistician so I don’t really know what’s expected, but if someone can contribute on that front, I’d be very interested.
here’s a sample of 5 different flip frenzy tournaments at 3 different venues. the number after the player count is the number of possible match-up combinations
40% being repeats doesn’t surprise me really. The way to test this would be run a random draw of some number of players for some number of rounds and then see what percentage of matchups are duplicated. This would probably be easy to do for some who knew a little bit of programming.
My theory (no data) is that FF is actually more random and therefor appears less random. But there’s also probably something that happens where people get stuck in play loops and I’m guessing this is a function of game times not varying enough.
I agree with that. I think it’s because ball times and game times vary so much. Think someone plays GNR verses some who plays a classic. Sometimes people just get stuck in a loop.
I agree that people often experience true randomness as being less random due to repetitions. When people ask for randomness in pinball tournaments what they most often are really asking for is an even distribution (what the “balanced” options in Match Play provides).
I don’t think that’s what’s happening for Flip Frenzies though. The format is not very random at all. The only random part is the duration of each game and when you look across games that part is pretty constant. If you happen to have spare machines available you can get a little bit of extra randomness because the software have a choice of a few different machines for each game, but in general the format is very predictable.
A group match play tournament is much much much better at providing a varied experience for each player. The software can easily match you with players you haven’t played before on machine you haven’t played.
It’s a trade off you have to make a tournament organizer: Do I value many games played at the cost of more repeat opponents and repeat machines, then pick a Frenzy. Do I absolute certainty that every player plays each other, then pick match play or round robin (head-to-head match play using strict swiss pairing is the extreme in that direction)
I’m not sure I agree (but I’m willing to have my mind changed). It seems like the seeding to start is 100% random so anything that happens after is based on that first random draw. The outcome of the first game and therefore, the second matchup is going to be based on whatever random group finishes first, so more randomness. Over time, trends may evolve but they evolve out of that initial randomness. At some point there will always be a natural progression towards some order, I’m just not sure how much of this feeling people have about frenzy is actually measurable or just the balanced vs. true random as you point out. I also think that much of the hate for FF formats is scrubbyness from good players that want things very predictable because it suits them the most.
What I don’t think you’re considering is that if it’s a group of games that all play the same length (which is common), that’s going to give the same cadence to all players. That’s what leads to players playing the same 3-4 opponents the whole time. I think the longer the frenzy is the better, but 2-3 hour frenzies become very predictable with games that all play the same length.
The distribution of draws is not even. Less skilled players tend to play a lot more because their games are shorter. As a result, when you are up the draw of who you will select as your opponent is biased towards hard games or less skilled players as the most likely next to finish.
Just checked the last tournament I played in and after 2 hours, the bottom half drew ~1.5 more games than the top half (about 11.3 to 12.7 with an average of 12 overall) so definitely something here but how does this skew the machine draws/queue negatively?