We use a similar method for Seattle pinball league finals, except each round 3/4 people move on:
Game 1 with 4 players, winner gains immunity and is safe - advances
Game 2 with the remaining 3 players, winner advances
Game 3 head to head, loser is eliminated.
This could be modified to only two games, but would obviously be more brutal feeling… gotta win first in either a 4 player match or a 3 player match, but honestly that’s not much different than current 4-2-1-0 where 90% of the time the two people that moved on got a 4, OR someone got a 9 or 12 and the other person snuck through with a couple of seconds and a third, with one of the players basically already advanced and thus not really competing hard against them
At my last tournament with tiebreakers, I made people play off on The Flip Side. This was popular and I’m going to do it again, but it probably wouldn’t go over well with a more serious-minded player base…
The main reason 2-game playoff rounds isn’t used more is WPPRs. Two games per round counts as 4 games or 16% towards 100% rating. An event taking the top 16 going 16-8-4 would then have 12 value games for their playoffs instead of 18, meaning the playoffs get you just 48%, not 72% (same with top 24 with top 8 getting a bye). If you’re doing best-game qualifying, you can get no more than 20% from time, so you need at least 32% more from game count, i.e. having to count 8 or more games in the bank. Almost nobody does that. With 3 games per round, no problem.
I have a rule that they can’t lift it off the table or move it outside some tape markings that I put around the legs, which is such a tight arena they basically have to sweat the whole time trying to remember not to nudge it by accident.
In DHPL league play we use the qualifying position to determine the tiebreaker. It puts more onus on qualifying better in the qualifying rounds so that you get the tiebreaker advantage in the playoff rounds. If it so happens that both players were identically tied in qualifying and they tie in playoff round (only ever happened once) we do a one ball playoff. Yes it is brutal but it does move along the event and allows us to finish things on time. It okay for league players but I do understand KD in that if you are playing in a major championship you would hate to lose like that, then again in Pinburgh I have lost out due to having a great 9th round so really you balance keeping the tournament on track and fairness for the players who have come a long way only to be eliminated by a tiebreaker rule.
not sure on TGP but theoretically someone can get through a round in 1 game so that would be shortest path but may be able to do a weighted average if you ask josh nicely
When you get to final four you can just do the same thing to get to a final 3 - then do the same thing but only 2 rounds to go to a final 2, then just play one game for the finals or best of 3
The mathematically calculated TGP for 2 games, 2 advance would be:
Game 1 is standard 4-player game played by everyone, so 8% TGP
Game 2 is a 3-player game which only half of advancing player participate in so
TGP = 1/2 * 1.5 * 4 = 3% TGP
Total TGP for the round: 11%
3 games, 3 advance:
Game 1 is still standard 4-player game played by every, so 8% TGP
Game 2 is a 3-player game which 2/3 of advancing players participate in so
TGP = 2/3 * 1.5 * 4 = 4% TGP
Game is a 2-player game which 1/3 of advancing players participate in so
TGP = 1/3 * 4 = 1.33% TGP
Total TGP for the round: 13.33%
It’s not an efficient format for maximizing TGP because advancing players aren’t playing in a lot of the games. Probably best for a finals where qualifying already provides a lot of value. Sounds fun, though!
Adding fractional points is in fact the way my NWPAS matchplay and pingolf spreadsheets would break ties that are defined as “number of first, number of second” etc., without any complex searching or counting.
Consider 7-5-3-1, Make that 7.1, 5.01, 3.001, and 1.0001, then add up the score, and the fractional part gives you the #1st, #2nd, etc. Also used it in PinGolf, when breaking ties as number of holes in 1, number of holes in 2, etc., but in reverse order. E.g. 5.1, 4,01, 3.001, 2.0001, and 1.00001.
Then you simply sort by the sum, and it all works out. Whether or not you display the fractional part is up to you.
I agree with the above completely: tiebreaker settled by additional H2H game is always preferred… but if time is of the essence, and tiebreaker must be settled by a different factor, this method from @WWJ seems the most fair and robust.
If it is known ahead of time that the tiebreaks will be based on performance from a specific round with no additional game played, then there are some deeper strategies.
In this example, let’s suppose A is coincidentally the bus driver of that group.
Fact: Player A scored (2) 4’s and is automatically through.
Fact: A knows B is ahead of C, also B is tied with D, and C is tied with D with respect to the number of compared wins.
Fact: Player A can elect to “earn a zero” and determine the potential ties for the remaining players.
(Granted this 3rd fact re-opens up the conversation about Player A intentionally losing, but that is very hard to prove, and as was further debated, maybe earning the zero is the Player A’s best competitive effort to make.)
So, using the idea of Player A electing to “earn” the zero and automatically advancing regardless, a strategy might be to choose the 3rd game as a randomizer to have a better chance of eliminating one of the remaining players.
Of course, Player A might choose the 3rd game to practice on and not really care what happens, or can defer picking if the rule was the same game cannot be chosen twice at all in the finals.
In any case, when there is a fixed tiebreak already in place such as using qualifying results or results of games already played that match, additional strategies can potentially be employed to alter the results.
Also used it in PinGolf, when breaking ties as number of holes in 1, number of holes in 2, etc., but in reverse order. E.g. 5.1, 4,01, 3.001, 2.0001, and 1.00001.